WO2025250601A2 - Nanopore direct rna sequencing of aminoacylated transfer rnas - Google Patents

Abstract

The present inventive concept relates to methods of nucleic acid sequencing, and applications and uses of the same. Aspects of the inventive concept include nanopore sequencing on nucleic acids, such as ribonucleic acids (RNAs), including, but not limited to, transfer RNAs (tRNAs). Further aspects include kits for performing the methods, and applications of the methods of the inventive concept as substantively described herein.

Description

NANOPORE DIRECT RNA SEQUENCING OF AMINOACYLATED TRANSFER RNAS

RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63/652,234, filed on May 28, 2024, and U.S. Provisional Patent Application No. 63/721 ,762, filed November 18, 2024, the entire content of each of which is hereby incorporated by reference in their entireties.

STATEMENT OF GOVERNMENT SUPPORT

[0002] This invention was made with government support under Grant Numbers GM1 19550, awarded by the National Institutes of Health, and 2330283, awarded by the National Science Foundation. The government has certain rights in the invention.

RESERVATION OF COPYRIGHT

[0003] This disclosure contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.

FIELD OF THE INVENTION

[0004] The present inventive concept relates to methods of sequencing nucleic acids, such as transfer RNA (tRNA), including aminoacylated tRNAs, using nextgeneration sequencing/high throughput approaches. These approaches include nanopore sequencing methods that provide novel tools and applications for tRNA sequencing (tRNA-seq).

BACKGROUND OF THE INVENTION

[0005] Transfer RNAs (tRNAs) are the fundamental adapter molecules of protein synthesis. They share a conserved structure but differ in three key attributes that also dictate how these molecules function: their sequence; the modifications that decorate them; and the amino acids they are charged with. While a number of different approaches for tRNA sequencing have also been developed over the last decade, overall, tRNAs remain challenging molecules to study in a high throughput manner, especially when looking to interrogate multiple attributes of these molecules within the same experiment. Existing approaches for the study of tRNA modifications and aminoacylation include: LCMS and partial digestion approaches, which enable bottoms- up sequencing of tRNA pools and their modifications; acid northern blots to separate aminoacylated from uncharged tRNA; and methods to chemically mark uncharged tRNA via a periodate pretreatment, enabling these molecules to be differentiated by DNA microarrays or later cDNA sequencing after a diagnostic ligation. While a number of different approaches for tRNA sequencing have also been developed over the last decade, each of these approaches also have technical tradeoffs. Moreover, cDNA- based sequencing methods provide only an indirect readout of modification and aminoacylation status. Another approach for studying tRNA pools is nanopore sequencing. Direct RNA-seq has been used to interrogate native tRNA molecules as they are ratcheted through biological nanopores, e.g., alpha hemolysin (oHL) or Mycobacterium smegmatis porin A (MspA), by a helicase. Direct RNA-seq has be proposed, e.g., in U.S. Patent Application Publication No. 2017/0253923. An attractive aspect of this approach is that RNA modifications or other chemical adducts are sensed directly due to disruptions in ionic current, rather than via detection of RT stops or misincorporations during cDNA synthesis. However, examination of tRNA aminoacylation by nanopore sequencing has not been contemplated. Moreover, although methods for di-, tri-, and longer peptide characterizing and sequencing may have been proposed, e.g., in U.S. Patent Application Publication No. 2023/0024319, no sequencing methods appear to apply to tRNAs charged with amino acid monomers.

SUMMARY OF THE INVENTION

[0006] Aspects of the present inventive concept expand the toolkit for high throughput interrogation of nucleic acid biology, such as RNA biology, for example, tRNA biology, by enabling direct detection of RNA/tRNA sequence, modification, and charging within the same nanopore sequencing experiment. A major innovation included in aspects of the present inventive concept is in a chemical ligation step that allows amino acids at the 3'-termini of an RNA/tRNA to be ligated to an imidazolated RNA oligonucleotide adapter, such that the amino acid forms the bridge between a 5' ester and 3' phosphoramidate linkage. This ligation has been optimized in the presence of a catalyst to generate quantitative ligation of all charged RNAs/tRNAs within 30 minutes, as demonstrated by acid northern blotting, and sequenced both synthetic and biological tRNAs for nanopore sequencing. Aspects of the present inventive concept demonstrate that different amino acids produce unique distortions in current and dwell time in nanopore experiments, laying the groundwork for efforts to train machine learning models to differentiate one amino acid from another in nanopore sequencing data. The library has also been extended to enable one-pot capture of charged and uncharged tRNAs from the same sample using two different sets of RNA adapters for ligation to aminoacylated/charged and non-aminoacylated/uncharged RNAs/tRNAs, respectively, facilitating easy read out in changes related to tRNA charging status from biological samples, while simultaneously capturing information about tRNA sequence and modification status. This approach facilitates exploration of new questions in tRNA biology, including interactions between RNA modifications and charging, tRNA misacylation, the charging of biological and/or tRNAs with non-standard amino acids, the study of tRNA modopathies and other diseases, as well as the sequencing of oligopeptidyl-tRNAs, e.g., di, tri, or tetrapeptidyl-tRNAs, such as those produced during translational arrest. Moreover, nanopore sequencing of individual amino acids embedded within a larger RNA context represents an important steppingstone towards nanopore peptide sequencing.

[0007] Aspects of the present inventive concept include methods of sequencing nucleic acids. In some aspects, the methods may include nanopore sequencing. In some aspects, the methods may include sequencing of ribonucleic acid (RNA), such as transfer RNA (tRNA). In some aspects, the sequenced RNA/tRNA is aminoacylated, or "charged," with an amino acid. In some aspects, the charged or uncharged RNA/tRNA that is sequenced is ligated to an RNA oligonucleotide adapter. Further aspects of the present inventive concept include kits for performing the methods of the present inventive concept. [0008] In an aspect of the inventive concept, provided is a method of characterizing a polynucleotide including: (i) providing together: a) a construct including a polynucleotide, wherein the polynucleotide includes at least one oligonucleotide adapter attached to one end of the polynucleotide, and b) a nanopore, wherein the construct and the nanopore are combined under conditions in which the construct associates with the nanopore; (ii) subjecting the construct and the nanopore to a condition that permits the polynucleotide to enter the nanopore and at least partially translocate through the nanopore; (iii) measuring a property associated with translocation of the polynucleotide through the nanopore; and (iv) characterizing the polynucleotide by analyzing the property measured as the polynucleotide translocates through the nanopore, wherein analyzing the property measured provides information regarding a characteristic of the polynucleotide.

[0009] In another aspect of the inventive concept, provided is a method of determining whether a polynucleotide is charged/aminoacylated including: (i) providing together: a) a construct including the polynucleotide, wherein the polynucleotide includes at least one oligonucleotide adapter attached to one end of the polynucleotide, and wherein the oligonucleotide adapter is attached to an amino acid/peptide if the polynucleotide is charged/aminoacylated through a phosphoramidate linkage generated by reacting the amino acid/peptide with an oligonucleotide adapter including an activated phosphate/phosphoramidating agent, and b) a nanopore, wherein the construct and the nanopore are combined under conditions in which the construct associates with the nanopore; (ii) subjecting the construct and nanopore to a condition that permits the polynucleotide to enter the nanopore and at least partially translocate through the nanopore; (iii) measuring a property associated with translocation of the polynucleotide through the nanopore; and (iv) characterizing whether the polynucleotide is charged/aminoacylated by analyzing the property measured as the polynucleotide translocates through the nanopore, wherein analyzing the property measured provides information regarding charging/aminoacylation status of the polynucleotide.

[0010] In another aspect of the inventive concept, provided is a method for characterizing charging/aminoacylation levels of tRNAs including: (i) generating a library of tRNA constructs, the tRNA constructs including at least one oligonucleotide adapter attached to one end of each tRNA, wherein charged/aminoacylated tRNA constructs include an RNA oligonucleotide adapter attached to an amino acid/peptide on the tRNA charged/aminoacylated with an amino acid/peptide through a phosphoramidate linkage generated by reacting the amino acid/peptide with an oligonucleotide adapter including an activated phosphate/phosphoramidating agent; (ii) providing the tRNA constructs together with nanopores, wherein the construct and the nanopore are combined under conditions in which a single tRNA construct associates with a single nanopore; (iii) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore; (iv) measuring a property associated with translocation of the tRNA constructs through the nanopores; and (v) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein analyzing the property measured provides information regarding the charging/aminoacylation levels of the tRNAs.

[0011] In another aspect of the inventive concept, provided is a method of characterizing changes in charging/aminoacylation of tRNAs including: (i) isolating a pool of tRNAs from a subject/organism growing under/subjected to an environmental condition; (ii)generating a library of tRNA constructs, the tRNA constructs each including at least one oligonucleotide adapter attached to one end of each tRNA, wherein the oligonucleotide adapter is attached to an amino acid/peptide if the tRNA is charged/aminoacylated through a phosphoramidate linkage generated by reacting the amino acid/peptide with an imidazolated oligonucleotide adapter; (iii) providing the tRNA constructs together with nanopores, wherein the constructs and the nanopores are combined under conditions in which a single tRNA construct associates with a single nanopore; (iv) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore; (v) measuring a property associated with translocation of the tRNA constructs through the nanopores; and (vi) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein the charging/aminoacylation levels characterized for the tRNA constructs are compared to charging/aminoacylation levels for a pool of tRNAs derived from a subject/organism growing under/subjected to a control environmental condition.

[0012] In another aspect of the inventive concept, provided is a method of characterizing charging/aminoacylation of hypomodified tRNA including: (i) isolating a pool of hypomodified tRNAs from a subject/organism; (ii) generating a library of tRNA constructs, the tRNA constructs including at least one oligonucleotide adapter attached to one end of each tRNA, wherein tRNA constructs charged/aminoacylated with an amino acid/peptide include an RNA oligonucleotide adapter attached to the amino acid/peptide on the tRNA charged/aminoacylated with an amino acid/peptide through a phosphoramidate linkage generated by reacting the amino acid/peptide with an imidazolated oligonucleotide adapter; (iii) providing the tRNA constructs together with nanopores, wherein the constructs and the nanopores are combined under conditions in which a single tRNA construct associates with a single nanopore; (iv) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore; (v) measuring a property associated with translocation of the tRNA constructs through the nanopores; and (vi) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein the charging/aminoacylation levels characterized for the hypomodified tRNA constructs are compared to charging/aminoacylation levels for a pool of normally modified tRNAs. [0013] In another aspect of the inventive concept, provided is a classifier trained to characterize, for example, charging/aminoacylation status of a polynucleotide/nucleic acid and/or charging/aminoacylation levels of a pool of polynucleotides/nucleic acids.

BRIEF DESCRIPTION OF THE DRAWINGS

[0014] FIG. 1 . A chemical ligation strategy for capture of aminoacylated tRNAs.

(Panel A) Schematic of splinted chemical ligation of 5'-phosphorimidazolide activated adapter (blue) to aminoacyl tRNA (black, with amino acid in red) in the presence of the catalyst 1 -(2-Hydroxyethyl)imidazole (HEI). (Panel B) Acidic charging northern of deacylated or untreated wild-type (BY4741 ) S. cerevisiae tRNA after chemical reaction in the presence (+) or absence (-) of activated 3' adapter, compared to tRNA input only (I). Blot has been probed for budding yeast tRNA Gly-GCC; asterisk indicates a presumed ligation intermediate. (Panel C) “Chemical-charging northern blot” resolving charged and uncharged tRNA-Gly-GCC via chemical ligation and analysis on a mini (8x10 cm) TBE 17M urea polyacrylamide gel followed by membrane transfer and hybridization with a Gly-GCC probe. (Panel D) Small RNAs (17-200 nt, 200 ng) from budding yeast were chemically ligated to activated 3' RNA adapter in a splinted ligation as previous panels, followed by an optional enzymatic ligation with T4 RNA ligase 2 (RNL2) to attach a second, splinted RNA adapter to the 5' end of the tRNA. Products were run on a 10% denaturing PAGE gel and stained with SYBR Gold. (Panel E) Schematic illustrating strategy for chemical ligation and sequencing of charged and uncharged biological tRNAs via nanopore direct RNA sequencing. (Panel F) Normalized mean current in picoamps for synthetic tRNA charged with glycine vs an uncharged control. The region visualized includes the 3' terminus of the tRNA (6 nt, positions 68-73), the CCA tail (positions 74-76), the aminoacylated position (dashed line), and the entirety of the 3' adapter sequence. For each colored trace, the solid line is the mean signal, and the shading spans the standard deviation.

[0015] FIG. 2. Model training and classification of aminoacylated tRNA reads from nanopore signal. (Panel A) Schematic illustrating ground truth libraries for model training. Library (i) was prepared by only chemical ligation to biological aminoacylated tRNAs. Library (ii) was prepared by only enzymatic ligation to deacylated biological tRNAs. (Panel B) Strategy for dataset preparation and neural network training of a Remora model to identify charged and uncharged tRNA reads from nanopore current signals. (Panel C) Schematic depicting reference-anchored signal from libraries (i) and (ii), with the 6-nt training window enclosed within the yellow box, and the position of the amino acid via dashed line. Mean current signals in picoamps for charged and uncharged tRNA reads are indicated by solid lines within each colored trace, with corresponding shading spanning the standard deviation. (Panel D) Density plot showing the distribution of aminoacylation prediction scores (modification likelihood or “ML” score, 0-255 scale) assigned by Remora model in validation libraries (95,859 deacylated tRNA reads, 33,494 aminoacylated tRNA reads). (Panel E) Confusion matrix and key metrics illustrating the model's performance at classifying a second set of replicates prepared as in Panel A, using a cutoff of ML > 199 for identification of aminoacylated reads from nanopore signal. (Panel F) Confusion matrix when a sequence alignment approach was used for charged tRNA classification on the same validation dataset.

[0016] FIG. 3. Sequencing and analysis of budding yeast tRNAs via chemicalcharging northern and aa-tRNA-seq. (Panel A) Correlation of tRNA aminoacylation (% charging) in budding yeast, as measured by acid northern in FIG. 1, Panel B (y-axis), and aa-tRNA-seq (mean and standard deviation of 3 replicates; dashed gray Pearson correlation with r and p-value; solid gray y=x line). (Panel B) Chemical-charging northern analysis of Leu-CAA aminoacylation in a budding yeast leucine auxotroph upon 15 minutes of leucine depletion, as measured by the percent of tRNA chemically ligated in each lane (upper band), with the percent-ligated tRNA per sample quantified under each lane. Signals were normalized to 5S rRNA probe (lower inset); relative abundances represent within-replicate normalized levels of total tRNA (with the sample grown in complete media normalized to 1 .0 and compared to the abundance for each leucine-starved sample). (Panel C) Log2 fold change in tRNA abundance and tRNA charging percent charged reads for all isodecoders after 15 minutes growth in synthetic complete (SC) or leucine dropout media. Points represent the mean of the same 3 biological replicates from Panel B, with error bars spanning the standard deviation. (Panel D) Volcano plot of the mean fold change in aminoacylation for the same 3 replicates in Panels B and C, with Z-test p-values on the y-axis and dashed line indicating the a threshold. (Panel E) Chemical-charging northern analysis of chemically ligated tRNA from two biological replicates of the RTD-sensitive budding yeast strain trm8A trm4A and the RTD-resistant strain trm8A trm4A met22A grown at permissive

(28²C) and nonpermissive (37^QC) temperatures for three hours. The percent of chemically ligated Val-AAC tRNA (upper band) is indicated below each lane, with a 5S rRNA probe as a loading control in the lower inset. The relative abundances represent within-replicate normalized levels of total tRNA (with the samples grown at 28^QC normalized to 1 .0 and compared to the abundance for a matched sample shifted to 37^QC). (Panel F) Log2 fold-change in tRNA abundance and tRNA charging percent charged reads across 3 biological replicates upon shift to nonpermissive temperature in trm8A trm4A cells. (Panel G) Volcano plot of the mean fold change in aminoacylation from Panel F, with Z-test p-values on the Y axis and dashed line indicating the a threshold. (Panel H) Log2 fold-change in tRNA abundance and tRNA charging percent charged reads across 3 biological replicates upon shift to nonpermissive temperature in trm8A trm4A met22A cells.

[0017] FIG. 4. Signal analysis and classification of amino acid identity using nanopore sequencing. (Panel A) Mean dwell time for a synthetic tRNA aminoacylated with 20 naturally-occurring amino acids as well as an uncharged control that has been enzymatically ligated to the 3' adapter prior to nanopore direct RNA sequencing. Sequences are ordered top to bottom on the y-axis by amino acid side chain molecular weight. The plotted region includes the 3' terminus of the tRNA (6 nt, positions 68-73), the CCA tail (positions 74-76), the aminoacylated position (dashed line, position 77) and the entirety of the 3' adapter sequence. The schematic underneath the heatmap depicts the direction of translocation and approximate location of the amino acid within the helicase motor protein (in orange) when the 3' adapter sequence is centered in the nanopore (blue) and the largest increases in dwell time are observed. (Panel B) Relative differences in normalized mean current between for each synthetic aminoacylated tRNA compared to the uncharged control, over the same region as in Panel A. The lower schematic illustrates the location of the amino acid within the nanopore reader head when the largest changes in current are observed. (Panel C) Scatter plot with linear regression line showing the relationship between mean dwell time at position 86 and amino acid volume. Points represent individual amino acids, with side-chain volume (in cubic angstroms) on the x-axis and the relative difference in mean current on the y-axis. A dashed line indicates the linear regression fit, with the Pearson’s correlation coefficient (r) and p-value displayed. (Panel D) Scatter plot showing the relationship between mean dwell time at position 86 and amino acid molecular weight (grams I mole). (Panel E) Scatter plot showing the relationship between mean dwell time at position 86 and the hydrophobicity index ⁸⁷ for each amino acid at pH 7, with higher values being more hydrophobic. (Panel F) Scatter plot showing the relationship between mean current differences at the aminoacylated position (position 77, dashed line in Panel C) and amino acid volume (cubic angstroms). (Panel G) Scatter plot showing the relationship between mean current differences at the aminoacylated position (position 77, dashed line in Panel C) and amino acid molecular weight (grams I mole). (Panel H) Scatter plot showing the relationship between mean current differences at the aminoacylated position (position 77, dashed line in Panel C) and the hydrophobicity index for each amino acid. (Panel I) Schematic depicting pairwise Remora model training strategy for differentiation of reads containing individual amino acids, using distribution of current signals for synthetic tRNA charged with proline vs lysine as an illustrative example. The mean current in picoamps for the lysyl-tRNA substrate is represented by the central blue line, the shaded blue area representing the standard deviation around the mean, and tRNA-Pro data plotted analogously in red. A yellow box outlines the window of signal (4 nt, with one representing a placeholder for the amino acid) on which each pairwise model was trained, with the location of the chemically ligated amino acid indicated by the dashed line. (Panel J) F1 scores for all pairwise Remora models trained on the 20 synthetic tRNA substrates using the strategy illustrated in Panel I.

[0018] FIG. 5. (Panel A) Chemical ligation of Flexizyme-charged synthetic tRNA in the absence of catalyst for 0, 1 , 2, or 27 hours. The analytical 16 % denaturing gel was stained with SYBR gold to visualize the RNA. (Panel B) Densitometry-based quantifications of the percent of aminoacylated tRNA shifted upon chemical ligation for the budding yeast tRNAs visualized on the acidic northern in FIG. 1 Panel B, after stripping and reprobing for the indicated isodecoders. (Panel C) Schematic of a 16 nt tRNA minisubstrate bearing a fluorophore (FAM, green diamond) undergoing chemical ligation to 5'-phosphorimidazolide activated hairpin adapter in the presence of the catalyst 1 -(2-hydroxyethyl)imidazole (HEI). (Panel D) Normalized levels of ligation product for fluorescently labeled tRNA minisubstrate in (Panel B) aminoacylated with the indicated amino acids using Flexizyme and reacted with the activated hairpin adapter in the presence of HEI for 90 minutes. (Panel E) Chemical-charging northern displaying a time course of budding yeast tRNA deacylated in Tris pH 9 for 30-90 minutes, with an untreated (“acylated”) control loaded next to each timepoint. Each sample was split in half prior to reaction in the presence (+) or absence (-) of the activated 3' adapter to chemically ligate aminoacylated tRNA. The membrane was probed for tRNA-Gly-GCC. (Panel F) The same northern membrane as in (Panel E), stripped and reprobed for tRNA-Val-AAC.

[0019] FIG. 6. Densitometric quantification of acylated and deacylated tRNA species from wild-type (BY4741) S. cerevisiae in the presence (+) or absence (-) of activated 3' adapter, compared to tRNA input only (I). Lanes 1 -3 represent a chemically deacylated control, lanes 4-6 untreated yeast tRNA, and lanes 7-9 a 50/50 mixture of the previous inputs. Asterisks indicate presumed ligation intermediates, while question marks indicate putative background ligation to incompletely deacylated tRNA species. (Panel A) Densitometric quantification of the full membrane shown in FIG. 1 , Panel B, probed for tRNA-Gly-GCC. This membrane was then stripped and re-probed with oligonucleotides complementary to budding yeast (Panel B) His-GUG, (Panel C) Leu- CAA, (Panel D) Pro-UGG, (Panel E) Asn-GUU, (Panel F) Cys-GCA, (Panel G) Phe- GAA, (Panel H) Tyr-GUA, (I) Gln-UUG, (Panel J) Trp-CCA, (Panel K) Thr-UGU, (Panel L) Val-AAC, (Panel M) Lys-UUU, (Panel N) iMet-CAU, (Panel O) Ala-UGC, (Panel P) Arg-UCU, (Panel Q) lle-UAU, (Panel R) Glu-UUC, (Panel S) Arg-GUC, and (Panel T) Ser-GCU isodecoders. The intensities of each band relative are indicated as a relative percentage per lane in the tables below each panel.

[0020] FIG. 7. (Panel A) Percentage of reads with "unblock" end status in nanopore sequencing libraries from synthetic tRNA chemically ligated to an imidazolated 3' RNA adapter and enzymatically ligated to a 5' RNA adapter. (Panel B) Schematic depicting experimental strategies tested for their effects on Asn- and Cys-tRNA pore blocking rates, including incorporation of a DNA/RNA hybrid 5' adapter, alkylation of cys-tRNA, or hydrolysis of Flexizyme-charged asn-tRNA. (Panel C) Percent of Cys-tRNA reads with "unblock" status using each of the relevant strategies above, as well as omission of 5' adapter. (Panel D) Equivalent percentage of "unblocked" Asn-tRNA reads for all three approaches tested.

[0021] FIG. 8. tRNA translocation time by isodecoder in budding yeast tRNA sequencing libraries prepared using only chemical ligation to imidazole-charged adaptors ("charged only", orange), or libraries where tRNA was first chemically deacylated followed by enzymatic ligation with T4 RNL2 ("uncharged only", blue). The y-axis displays the translocation duration for the entire read in seconds. Statistical significance between the distributions in each library was assessed using the Wilcoxon test, with significance levels indicated by asterisks above each comparison (****p < 0.0001 ).

[0022] FIG. 9. Chemical-charging northern analysis of chemically ligated tRNA to visualize relative aminoacylation levels for (Panel A) Ala-UGC, (Panel B) Gly-GCC, (Panel C) Ser-UGA, and (Panel D) Val-AAC isodecoders in a budding yeast leucine auxotroph upon 15 minutes of leucine depletion, as measured by the percent of tRNA chemically ligated in each lane (upper band), with the percent-ligated tRNA per sample quantified under each lane. Signals were normalized to 5S rRNA probe abundance shown in FIG. 3, Panel B; relative abundances represent within-replicate normalized levels of total tRNA (with the sample grown in complete media normalized to 1.0 and compared to the abundance for each leucine-starved sample).

[0023] FIG. 10. Chemical-charging northern analysis of chemically ligated tRNA from budding yeast with single, double, or triple genetic deletions affecting rapid tRNA decay (RTD), compared to a wild-type control (BY4741 ). Biological replicates were grown at permissive (28^QC) and nonpermissive (37^QC) temperatures for three hours. The percent of chemically ligated tRNA (upper band) is indicated below each lane, with a 5S rRNA probe as a loading control. The relative abundances represent within-replicate normalized levels of total tRNA (with the samples grown at 28^QC normalized to 1 .0 and compared to the abundance for a matched sample shifted to 37^QC). (Panel A) Codeletion of TR/WS and TRM4 and triple deletion with MET22 as in Fig. 3, Panel E, but with membrane re-probed for Gly-GCC, (Panel B) Cys-GCA, or (C) Leu-CAA. The inset below Panel A contains the same membrane re-probed for 5S ribosomal RNA as a loading control. (Panel D) Single deletions of TRM4 and MET22, with membrane probed for Val-AAC. The inset below shows the same membrane re-probed for 5S rRNA as a loading control. (Panel E) Single deletions of TRM8 and MET22, with membrane probed for Val-AAC. The inset below shows the same membrane re-probed for 5S rRNA. (Panel F) Single deletions of TRM4 and MET22, where the membrane from (D) has been re-probed for Gly-GCC. (G) Single deletions of TRM8 and MET22, where the membrane from (E) has been re-probed for Gly-GCC. [0024] FIG. 11 . Additional analysis of aa-tRNA-seq data from RTD strains grown at permissive (28^QC) and nonpermissive (37^QC) temperatures. (A) I_og2 fold change in tRNA abundance and tRNA charging percent charged reads in the RTD-sensitive budding yeast strain tanlA trm44A and the RTD-resistant strain tanlA trm44A met22A after 3 hours growth at the nonpermissive or permissive temperatures. Points represent the mean values for each tRNA isodecoder across 3 biological replicates, with error bars spanning the standard deviation. (B) Volcano plot comparing the mean fold change in tRNA abundance for all RTD-sensitive and resistant strains sequenced in this study, with Z-test p-values on the y-axis and the grey box indicating the a threshold. Each panel contains data from triplicate sequencing of two RTD sensitive budding yeast strains (trmSA trm4A and tanlA trm44A)' as well as the corresponding RTD-resistant strains trm8A trm4A met22A and tanlA trm44A met22A), and a control strain with a

MET22 disruption. (C) Volcano plot comparing the mean fold change in tRNA aminoacylation for all RTD-sensitive and resistant strains sequenced in this study. Each panel contains data from triplicate sequencing of two RTD sensitive budding yeast strains (trm8A trm4A and tanlA trm44A)' as well as the corresponding RTD-resistant strains trm8A trm4A met22A and tanlA trm44A met22A), and a control strain with a

MET22 disruption. The y axis contains Z-test p-values, and the grey box indicates the a threshold.

[0025] FIG. 12. Chemical-charging northern analysis of chemically ligated tRNA from two biological replicates of the RTD-sensitive budding yeast strain tanlA trm44A and the RTD-resistant strain tanlA trm44A met22A grown at permissive (28^QC) and nonpermissive (37^QC) temperatures for three hours. tRNA extracted from BY4741 yeast were loaded as an additional control in the last 3 lanes, with the final lane containing tRNA from the 28^QC sample where the imidazolated 3'-adapter was not added to the chemical ligation reaction. (Panel A) The percent of chemically ligated Ser-UGA tRNA (upper band) is indicated below each lane. The relative abundances represent within- replicate normalized levels of total tRNA (with the samples grown at 28^QC normalized to 1 .0 and compared to the abundance for a matched sample shifted to 37^QC). (Panel B) Reprobing of the same membrane using a 5S rRNA probe as a loading control. (C) Additional reprobing of the same membrane with an oligonucleotide complementary to Leu-CAA tRNA.

[0026] FIG. 13. Normalized signal intensity (current in picoamps) for synthetic tRNA charged with 20 naturally-occurring amino acids using Flexizyme. Each panel plots the mean current in picoamps for an uncharged tRNA substrate in the central blue line, with the shaded blue area representing the standard deviation around the mean. Each aminoacylated comparison is plotted analogously in orange. The x-axis spans the same window of interest as in FIG. 2, containing six nucleotides at the 3' terminus of the tRNA, the CCA tail, the aminoacylated position (dashed line, included as an extra nucleotide inserted at position 77 in the reference) and the entirety of the 3' adapter sequence from nt 78 onward.

[0027] FIG. 14. Mean dwell time for synthetic tRNA charged with 20 naturally- occurring amino acids using the Flexizyme. Each panel plots the mean dwell in milliseconds at each nucleotide in an uncharged tRNA substrate in the central blue line, with the shaded blue area representing the standard deviation around the mean. Each aminoacylated comparison is plotted analogously in orange. The x-axis spans the same window of interest as in FIG. 2, containing six nucleotides at the 3' terminus of the tRNA, the CCA tail, the aminoacylated position (dashed line, included as an extra nucleotide inserted at position 77 in the reference) and the entirety of the 3' adapter sequence from nt 78 onward.

DETAILED DESCRIPTION OF THE INVENTION

[0028] The present inventive concept will now be described with reference to the following embodiments. As is apparent by these descriptions, this inventive concept can be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the inventive concept to those skilled in the art. For example, features illustrated with respect to one embodiment can be incorporated into other embodiments, and features illustrated with respect to a particular embodiment can be deleted from that embodiment. In addition, numerous variations and additions to the embodiments suggested herein will be apparent to those skilled in the art in light of the instant disclosure, which do not depart from the instant inventive concept.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this inventive concept belongs. The terminology used in the description of the inventive concept herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the inventive concept.

[0030] All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety.

[0031] The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the description of the invention and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Additionally, as used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items and may be abbreviated

[0032] The abbreviation, "e.g." is derived from the Latin exempli gratia, and is used herein to indicate a non-limiting example. Thus, the abbreviation "e.g." is synonymous with the term "for example." No language in the specification should be construed as indicating any non-claimed element essential to the practice of the application.

[0033] As used herein the term "comprising" or "comprises" is used in reference to compositions, methods, and respective component(s) thereof, that are useful to an embodiment, yet open to the inclusion of unspecified elements, whether useful or not. It will be understood by those within the art that, in general, terms used herein are generally intended as "open" terms (e.g., the term "including" should be interpreted as "including but not limited to," the term "having" should be interpreted as "having at least," the term "includes" should be interpreted as "includes but is not limited to," etc.). [0034] The term "comprise," as used herein, in addition to its regular meaning, may also include, and, in some embodiments, may specifically refer to the expressions "consist essentially of" and/or "consist of." Thus, the expression "comprise" can also refer to, in some embodiments, the specifically listed elements of that which is claimed and does not include further elements, as well as embodiments in which the specifically listed elements of that which is claimed may and/or does encompass further elements, or embodiments in which the specifically listed elements of that which is claimed may encompass further elements that do not materially affect the basic and novel characteristic(s) of that which is claimed. For example, that which is claimed, such as a composition, construct, formulation, method, system, etc. "comprising" listed elements also encompasses, for example, a composition, construct, formulation, method, kit, etc. "consisting of," i.e., wherein that which is claimed does not include further elements, and a composition, construct, formulation, method, kit, etc. "consisting essentially of," i.e., wherein that which is claimed may include further elements that do not materially affect the basic and novel characteristic(s) of that which is claimed.

[0035] The term "about" generally refers to a range of numeric values that one of skill in the art would consider equivalent to the recited numeric value or having the same function or result. For example, "about" may refer to a range that is within ± 1 %, ± 2%, ± 5%, ± 10%, ± 15%, or even ± 20% of the indicated value, depending upon the numeric values that one of skill in the art would consider equivalent to the recited numeric value or having the same function or result. Furthermore, in some embodiments, a numeric value modified by the term "about" may also include a numeric value that is "exactly" the recited numeric value. In addition, any numeric value presented without modification will be appreciated to include numeric values "about" the recited numeric value, as well as include "exactly" the recited numeric value. Similarly, the term "substantially" means largely, but not wholly, the same form, manner or degree and the particular element will have a range of configurations as a person of ordinary skill in the art would consider as having the same function or result. When a particular element is expressed as an approximation by use of the term "substantially," it will be understood that the particular element forms another embodiment. [0036] Embodiments of the inventive concept relate to characterizing a polymer, such as a polynucleotide. In some embodiments, characterizing the polymer may include characterizing a sequence of a polymer units present within a polymer, such as a sequence of nucleotides within a polynucleotide/nucleic acid, may be determined using a system in which the polymer is translocated through a nanopore. The system may make one or more measurements during the translocation that depend in some way on the polymer units in the polymer. For example, a current across the nanopore, dwell time of the polymer in the nanopore, and/or translocation speed may be measured during translocation of the polymer through the nanopore. In some cases, the measurements made by the measurement system depend on the identity of the polymer unit(s) as they translocate through the nanopore, so the signal over time allows the sequence of polymer units to be determined. However, the signal must be decoded to determine the underlying sequence of polymer units that produced the signal.

[0037] In general, the polymer characterized may be of any type, for example a polynucleotide (or nucleic acid), a polypeptide such as a protein, or a polysaccharide. The polymer may be naturally occurring, artificial/synthetically prepared, or engineered. The polynucleotide may include a homopolymer region. In some embodiments, the polymer is a polynucleotide/nucleic acid. In some embodiments, the polymer may be a polynucleotide/nucleic acid including an amino acid/peptide/polypeptide associated with/attached to the polynucleotide/nucleic acid.

[0038] For a polynucleotide or nucleic acid, the polymer units may be nucleotides. The polynucleotide/nucleic acid may be ribonucleic acid (RNA), deoxyribonucleic acid (DNA), a combination of RNA and DNA, i.e., a polynucleotide including both ribonucleotides and deoxyribonucleotides, a cDNA, or any synthetic nucleic acid known in the art, such as peptide nucleic acid (PNA), glycerol nucleic acid (GNA), threose nucleic acid (TNA), locked nucleic acid (LNA) or other synthetic polymers with nucleotide side chains, alone, in combination one another, or in combination with ribonucleotides/deoxyribonucleotides. The PNA backbone is composed of repeating N- (2-aminoethyl)-glycine units linked by peptide bonds. The GNA backbone is composed of repeating glycol units linked by phosphodiester bonds. The TNA backbone is composed of repeating threose sugars linked together by phosphodiester bonds. LNA is formed from ribonucleotides as discussed above having an extra bridge connecting the 2' oxygen and 4' carbon in the ribose moiety. The polynucleotide/nucleic acid may be single-stranded, may be double-stranded or include both single-stranded and doublestranded regions. The polynucleotide/nucleic acid may include one strand of RNA hybridized to one strand of DNA. In some embodiments, the polynucleotide/nucleic acid may be an RNA. In some embodiments, the RNA may be, for example, a transfer RNA (tRNA), or a tRNA fragment. In some embodiments, the polynucleotide may be an RNA including a tRNA-like structure, such as, but not limited to, tRNA-like molecules. In some embodiments, the RNA may be an aminoacylated RNA, such as an aminoacylated tRNA, which may be referred to as a "charged" tRNA, including an amino acid attached to the 3'-OH of the tRNA.

[0039] The polymer units may include any type of nucleotide. The nucleotide may be naturally occurring or artificial. A nucleotide typically contains a nucleobase, a sugar and at least one phosphate group. The nucleobase and sugar form a nucleoside. The nucleobase is typically heterocyclic. Suitable nucleobases include purines and pyrimidines, such as, but not limited to, adenine, guanine, thymine, uracil and cytosine. The sugar typically may be a pentose (five carbon) sugar. Suitable sugars include, but are not limited to, ribose and deoxyribose. The nucleotide typically may be a ribonucleotide or deoxyribonucleotide. The nucleotide typically contains a monophosphate, diphosphate or triphosphate.

[0040] In the case of tRNAs, tRNA fragments, RNAs including tRNA-like structures, and/or tRNA-like molecules, such polynucleotides/nucleic acids may include nucleosides such as inosine, dyhydrouridine, ribothymidine, pseudouridine, in addition to nucleosides adenosine, guanosine, cytidine, and uridine, as well as modified nucleosides, such as, but not limited to A/¹ -methyladenosine, / -methylguanine, and 5- methylcytosine. Further modified nucleosides found in tRNAs are described, e.g., in Vare et al. Biomolecules 2017; 7(1 ):29.

[0041] In the case of modified nucleosides/nucleotides including a modified base, the base modification may confer stability to a polynucleotide/nucleic acid, as well as be a damaged, or an epigenetic base. The nucleotide can be labeled or modified to act as a marker with a distinct signal. This can be used to identify the absence of a base, for example, an abasic unit or spacer in the polynucleotide.

[0042] In some embodiments, the labeled/modified nucleotide may activate a polynucleotide/nucleic acid/oligonucleotide for attachment/ligation to, e.g., a charged/aminoacylated polynucleotide/nucleic acid, such as an aminoacyl-tRNA. The modified nucleotide, such as a modified nucleotide at the 5' end of a polynucleotide/oligonucleotide may include, e.g., an activated phosphate, such as a phosphoramidating agent. An exemplary polynucleotide/oligonucleotide including an activated phosphate/phosphoramidating agent may be a 5'-phosphorimidazolated oligonucleotide, that facilitates attachment of the oligonucleotide to an amino group of an amino acid/peptide attached to the 3'-end of a polynucleotide/nucleic acid, such as the amino acid of a charged/aminoacylated tRNA. In some embodiments, the oligonucleotide including an activated phosphate/phosphoramidating agent does not or cannot ligate to an uncharged polynucleotide/nucleic acid/tRNA.

[0043] In embodiments where the polynucleotide/nucleic acid may include, e.g., an amino acid or a peptide attached to, e.g., the 3' end of the polynucleotide/nucleic acid, such as in a charged/aminoacylated tRNA, attachment/ligation of the activated oligonucleotide facilitates characterization of the polynucleotide/nucleic acid. Characterizing of a polynucleotide/nucleic acid according to embodiments of the inventive concept may include determining whether a polynucleotide/nucleic acid, such as, but not limited to, a tRNA, is charged/aminoacylated with an amino acid or peptide. Additionally, in some embodiments, characterizing the polynucleotide/nucleic acid may include characterizing the amino acid/peptide attached to a charged/aminoacylated polynucleotide/nucleic acid, such as a charged/aminoacylated tRNA, i.e., determining what amino acid/peptide is attached at the 3' end of the tRNA. Further embodiments of characterizing the polynucleotide/nucleic acid, such as a tRNA, may include: characterizing charging/aminoacylation levels (extent to which a pool of nucleic acids/polynucleotides/tRNAs is charged/aminoacylated vs. uncharged) of a nucleic acid/polynucleotide/tRNA; interactions between RNA modifications and charging/aminoacylation; characterizing mischarging/misacylation of a nucleic acid/polynucleotide/tRNA; and/or effects on tRNA charging/mischarging that are the result of or may result from changes in environmental conditions/growth conditions, such as, but not limited to, nutrient depletion/limitation, nutrient starvation and/or other stress conditions, as well as may result in conditions of tRNA hypomodification, and confirming known and identifying unexpected changes in tRNA aminoacylation and abundance that may result from such changes in environmental conditions/growth conditions.

[0044] Although polynucleotides/nucleic acids characterized by method of the inventive concept are more particularly described in relation to tRNAs, the description is not intended to be limited strictly thereto. For example, polynucleotides/nucleic acids that may be charged/aminoacylated and may be characterized by methods according to the present inventive concept may be part of a larger system/macromolecular construct/machine, for example, polynucleotides/nucleic acids associated with a ribosome, such as ribosome-associated peptidyl tRNA constructs, e.g., tRNAs and associated nascent peptide chains located within the ribosome itself.

[0045] In some embodiments, the nanopore for characterizing a polymer, such as a polynucleotide/nucleic acid, is a protein pore, that may have the following properties. The protein pore may be a transmembrane protein pore. Transmembrane protein pores for use in accordance with the inventive concept can be derived from [3-barrel pores or a-helix bundle pores. [3-barrel pores include a barrel or channel that is formed from [3- strands. Suitable [3-barrel pores include, but are not limited to, [3-toxins, such as a- hemolysin, anthrax toxin and leukocidins, and outer membrane proteins/porins of bacteria, such as Mycobacterium smegmatis porin (Msp), for example MspA, MspB, MspC or MspD, lysenin, outer membrane porin F (OmpF), outer membrane porin G (OmpG), outer membrane phospholipase A and Neisseria autotransporter lipoprotein (NalP). a-helix bundle pores include a barrel or channel that is formed from a-helices. Suitable a-helix bundle pores include, but are not limited to, inner membrane proteins and a outer membrane proteins, such as WZA and ClyA toxin. The transmembrane pore may be derived from Msp or from a-hemolysin (a-HL). The transmembrane pore may be derived from lysenin. Suitable pores derived from lysenin are disclosed in WO 2013/153359. Suitable pores derived from MspA are disclosed in WO 2012/107778. The pore may be derived from CsgG, such as disclosed in WO 2016/034591 . The pore may be a DNA origami pore.

[0046] The protein pore may be a naturally occurring pore or may be a mutant pore. Typical pores are described in WO 2010/109197, Stoddart D et aL, Proc Natl Acad Sci, 2009; 106(19):7702-7, Stoddart D et aL, Angew Chem Int Ed Engl. 2010; 49(3):556-9, Stoddart D et aL, Nano Lett. 2010 Sep. 8; 10(9):3633-7, Butler T Z et aL, Proc Natl Acad Sci 2008; 105(52) :20647-52, and WO 2012/107778. The protein pore may be one of the types of protein pore described in WO 2015/140535 and may have the sequences that are disclosed therein.

[0047] The protein pore may be inserted into an amphiphilic layer, such as a biological membrane, for example a lipid bilayer. An amphiphilic layer is a layer formed from amphiphilic molecules, such as phospholipids, which have both hydrophilic and lipophilic properties. The amphiphilic layer may be a monolayer or a bilayer. The amphiphilic layer may be a co-block polymer such as disclosed in Gonzalez-Perez et aL, Langmuir, 2009, 25, 10447-10450 or WO 2014/064444. Alternatively, a protein pore may be inserted into an aperture provided in a solid-state layer, for example as disclosed in WO 2012/005857.

[0048] A suitable apparatus for providing an array of nanopores is disclosed in WO 2014/064443. The nanopores may be provided across respective wells wherein electrodes are provided in each respective well in electrical connection with an ASIC for measuring current flow through each nanopore. A suitable current measuring apparatus may include the current sensing circuit as disclosed in WO 2016/181118.

[0049] The nanopore may include an aperture formed in a solid-state layer, which may be referred to as a solid-state pore. The aperture may be a well, gap, channel, trench or slit provided in the solid-state layer along or into which analyte may pass. Such a solid-state layer is not of biological origin. In other words, a solid-state layer is not derived from or isolated from a biological environment such as an organism or cell, or a synthetically manufactured version of a biologically available structure. Solid-state layers can be formed from both organic and inorganic materials including, but not limited to, microelectronic materials, insulating materials such as SisN4, AhOe, and SiO, organic and inorganic polymers such as polyamide, plastics such as Teflon® or elastomers such as two-component addition-cure silicone rubber, and glasses. The solid-state layer may be formed from graphene. Suitable graphene layers are disclosed in WO 2009/035647, WO 201 1/046706, or WO 2012/138357. Suitable methods to prepare an array of solid-state pores are disclosed in WO 2016/187519.

[0050] Such a solid-state pore is typically an aperture in a solid-state layer. The aperture may be modified, chemically, or otherwise, to enhance its properties as a nanopore. A solid-state pore may be used in combination with additional components which provide an alternative or additional measurement of the polymer such as tunneling electrodes (Ivanov et aL, Nano Lett. 2011 Jan. 12; 11 (1 ):279-85), or a field effect transistor (FET) device (e.g., as described in WO 2005/124888). Solid-state pores may be formed by known processes including for example those described in WO 00/79257. In some embodiments, the nanopore may be a hybrid of a solid-state pore with a protein pore.

[0051] In some embodiments, a series of measurements of a property that depends on the polymer, such as a polynucleotide/nucleic acid, translocating with respect to the pore may be determined. The series of measurements may form a signal. The property that is measured may be associated with an interaction between the polymer and the pore. Such an interaction may occur in a constricted region of the pore. Measurements of a property, in some embodiments, may be used, e.g., in basecalling to provide/identify/characterize the sequence of a polynucleotide/nucleic acid. In some embodiments, measurements of a property may be used in determining whether a polynucleotide/nucleic acid is charged/aminoacylated with an amino acid/peptide. In some embodiments, measurements of a property may be used in identifying/characterizing the amino acid/peptide attached to a charged/aminoacylated polynucleotide/nucleic acid. In some embodiments, measurements of a property may be used characterizing charging/aminoacylation levels (extent to which a pool of nucleic acids/polynucleotides is charged/aminoacylated vs. uncharged) of a nucleic acid/polynucleotide/tRNA.

[0052] In some embodiments, measurements of a property may be used in characterizing charging and/or mischarging of a nucleic acid/polynucleotide/tRNA and/or effects on tRNA charging/mischarging that or may result from, e.g., changes in environmental conditions/growth conditions, such as, but not limited to, nutrient depletion, nutrient starvation, and/or stress, and/or such that may result in translational inefficiencies, translational bottlenecks, and/or stress. Accordingly, measurements of properties, such measurements of tRNA charging in biological samples, may be used, e.g., for detecting and/or diagnosing translational inefficiencies, translational bottlenecks, and/or stress.

[0053] In some embodiments, a property that is measured may be the ion current flowing through a nanopore. These and other electrical properties may be measured using single channel recording equipment as described in Stoddart D et al., Proc Natl Acad Sci, 12; 106(19):7702-7, Lieberman K R et al, J Am Chem Soc. 2010; 132(50):17961 -72, and WO-2000/28312. Alternatively, measurements of electrical properties may be made using a multi-channel system, for example as described in WO 2009/077734, WO 201 1/067559 or WO 2014/064443.

[0054] Ionic solutions may be provided on either side of the membrane or solid-state layer, which ionic solutions may be present in respective compartments. A sample containing the polymer, such as a polynucleotide/nucleic acid of interest, may be added to one side of the membrane and allowed to move with respect to the nanopore, for example under a potential difference or chemical gradient. The signal may be derived during the movement of the polymer with respect to the pore, for example taken during translocation of the polymer through the nanopore. The polymer may partially translocate the nanopore.

[0055] In order to allow measurements to be taken as the polymer, such as a polynucleotide/nucleic acid, translocates through a nanopore, the rate of translocation may be controlled by a polynucleotide/nucleic acid binding moiety. Typically, the moiety can move the polynucleotide/nucleic acid through the nanopore with or against an applied field. The moiety can be a molecular motor using for example, in the case where the moiety is an enzyme, enzymatic activity, or as a molecular brake. There are a number of methods for controlling the rate of translocation including use of polynucleotide binding enzymes. Suitable enzymes for controlling/handling the rate of translocation of polynucleotides/nucleic acids include, but are not limited to, polymerases, helicases, exonucleases, single stranded and double stranded binding proteins, and topoisomerases, such as gyrases. The polynucleotide/nucleic acid handling enzyme may be, for example, one of the types of polynucleotide handling enzymes described in WO 2015/140535 or WO 2010/086603.

[0056] Translocation of the polymer through the nanopore may occur, either cis to trans or trans to cis, either with or against an applied potential. The translocation may occur under an applied potential which may control the translocation.

[0057] Similarly, the properties that are measured may be of types other than ion current. Some examples of alternative types of property include without limitation: electrical properties and optical properties. A suitable optical method involving the measurement of fluorescence is disclosed by J. Am. Chem. Soc. 2009, 131 1652-1653. Possible electrical properties include, e.g., ionic current, impedance, a tunneling property, such tunneling current (Ivanov et aL, Nano Lett. 2011 Jan. 12; 11 (1 ):279-85), and a FET (field effect transistor) voltage (WO 2005/124888). One or more optical properties may be used, optionally combined with electrical properties (Soni et al., Rev Sci Instrum. 2010 January; 81 (1 ) :014301 ). The property may be a transmembrane current, such as ion current flow through a nanopore. The ion current may typically be the DC ion current, although in principle an alternative is to use the AC current flow (i.e., the magnitude of the AC current flowing under application of an AC voltage).

[0058] In some embodiments, properties of the translocation of the polymer through the nanopore, such as translocation speed and/or dwell time, may be affected by characteristics/properties of the polymer/polynucleotide/nucleic acid. For example, aminoacylated tRNAs exhibit variations in dwell time dependent on the particular amino acid the tRNA may be charged with. Moreover, aminoacylated tRNAs exhibit differences in dwell time when compared to their uncharged counterparts, i.e., tRNAs without an amino acid/peptide attached at the 3' end. Accordingly, in some embodiments, the rate at which the polymer/polynucleotide/nucleic acid translocates through a nanopore, including translocation speed and/or dwell time in the nanopore, may be a property measured and used to characterize the polymer/polynucleotide/nucleic acid.

[0059] In some embodiments, characterizing a polynucleotide/nucleic acid, such as a tRNA, according to the inventive concept as set forth herein, may include a construct including, and/or provide a construct including, a polynucleotide/nucleic acid, such as the polynucleotide/nucleic acid to be characterized, and at least one oligonucleotide adapter attached to one end of the polynucleotide/nucleic acid. In some embodiments, the construct may include an oligonucleotide adapter attached to the 3' end of the polynucleotide/nucleic acid. In some embodiments, the construct may include an oligonucleotide adapter attached to the 5' end of the polynucleotide/nucleic acid. In some embodiments, the construct may include an oligonucleotide adapter attached to the 3' end and an oligonucleotide adapter attached to the 5' end of the polynucleotide/nucleic acid.

[0060] In some embodiments, the oligonucleotide for attaching/ligating to the 3' end of the polynucleotide/nucleic acid includes an activated 5' phosphate that reacts with the amino group of an amino acid/peptide attached at the 3' end of the polynucleotide/nucleic acid, e.g., attaching/ligating to an amino acid/peptide of an a charged/aminoacylated tRNA. In some embodiments, the activated 5' phosphate of the oligonucleotide adapter is a phosphoramidating agent, such as a 5' phosphorimidazolated oligonucleotide adapter. In some embodiments, the oligonucleotide adapter, such as the oligonucleotide adapter for attaching/ligating to the 3' end of the polynucleotide/nucleic acid can be attached to the 3' end of the polynucleotide/nucleic acid, e.g., attached to the CCA-3' end of an uncharged tRNA. In some embodiments, the oligonucleotide adapter, such as the oligonucleotide adapter for attaching/ligating to the 5' end of the polynucleotide/nucleic acid can associate/hybridize with, e.g., the CCA-3' end of a tRNA and/or sequences of the oligonucleotide/oligonucleotide adapter for ligating/attaching to the 3' end of the polynucleotide/nucleic acid/tRNA. In some embodiments, the oligonucleotide adapter, such as the oligonucleotide adapter for attaching/ligating to the 5' end of the polynucleotide/nucleic acid oligonucleotide adapter for attaching/ligating to the 5' end of the polynucleotide/nucleic acid may be a DNA/RNA hybrid, which includes both deoxyribonucleotides/nucleosides, as well as ribonucleotides/nucleosides. For example, the oligonucleotide may include/be made up entirely of deoxyribonucleotides, except for, e.g., the last five nucleotides at the 3' end of the oligonucleotide adapter, which may include/be made up of ribonucleotides. [0061] In some embodiments, the oligonucleotide adapters of the inventive concept may include sequences that facilitate association with/binding to, either directly, or indirectly, e.g., through one or more additional polynucleotides/nucleic acids/oligonucleotides that are associated with, a polynucleotide/nucleic acid binding moiety that can translocate the polynucleotide/nucleic acid through a nanopore. Suitable moieties for translocating a polynucleotide/nucleic acid through a nanopore include, but are not limited to, polymerases, helicases, exonucleases, single stranded and double stranded binding proteins, and topoisomerases, such as gyrases, and/or one of the types of polynucleotide handling enzymes described in WO 2015/140535 or WO 2010/086603.

[0062] In some embodiments, the methods for analyzing a signal produced or a property exhibited/measured as a result of translocation of a polymer, such as a polynucleotide/nucleic acid, through a nanopore may include applying a machine learning (ML) operation including one or more neural networks into which data from the signal may be input. In some embodiments, such neural networks may include, for example, convolutional neural networks (CNNs), recurrent neural network (RNNs), and/or combinations thereof. In some embodiments, the neural network combines a CNN with a long short-term memory (LSTM) model, i.e., a convolutional log short-term memory (ConvLSTM) model/network.

[0063] According to some embodiments, the techniques for analyzing a signal produced or a property exhibited/measured as a result of translocation of a polynucleotide/nucleic acid through a nanopore may include selecting windows of a time-ordered signal. A "window" of a signal may, for instance, refer to a contiguous subset of the signal that retains the time-ordering present in the original signal. Each window may be analyzed to determine whether there was a transition in the polymer sequence in the window and which units of the sequence the transition was between. A plurality of windows of the signal may be analyzed in this manner, which may in some cases be overlapping in the time ordered sequence of measurements. For instance, a first window may be analyzed that includes, e.g., the samples 1-20 in the signal, a second window may be analyzed that includes the samples 3-22 in the signal, etc. The number of sequential samples in a window may be referred to as its "length," and the size of the step between successive windows selected for analysis may be referred to as the "stride."

[0064] In some embodiments, techniques for analyzing a signal produced or a property exhibited/measured as a result of translocation of a polynucleotide/nucleic acid through a nanopore may include deriving a feature vector based on a number of samples from the signal. In some cases, a feature vector may be derived from a selected window of the signal. In some embodiments, the feature vector may be generated by a neural network wherein the samples are provided as an input to the neural network and the feature vector is output from the neural network.

[0065] In some embodiments, the techniques for analyzing a signal produced or a property exhibited/measured as a result of translocation of the polynucleotide/nucleic acid through a nanopore may comprise generating a plurality of weights for a portion of the signal, wherein each weight is associated with a transition between labeled units of the polynucleotide/nucleic acid. The weights may be indicative of a likelihood that a transition occurred between a first of the labeled units to a second of the labeled units within the portion of the signal. As one example, if the labeled units of the polymer were to correspond to polynucleotides having one of the four bases A, C, G and T, for a given portion of the signal sixteen weights may be generated each corresponding to one of the possible transitions between these four labels, A — ► A, A — > C, etc. In some cases, a set of weights may be derived from a selected window of the signal. In some embodiments, the set of weights may be generated by a neural network wherein samples from the signal (or data derived from the samples, such as a feature vector) are provided as an input to the neural network and the weights are output from the neural network.

[0066] In some embodiments, the neural network may include a trained model/classifier that may operate on data from the signal that has been input into the trained model to provide information regarding a polynucleotide/nucleic acid, such as a tRNA, e.g., nucleotide sequence of the polynucleotide/nucleic acid, charging/aminoacylation status of the polynucleotide/nucleic acid, amino acid identity on a charged/aminoacylated polynucleotide/nucleic acid, and/or extent of charging/aminoacylation of a tRNA pool/population from the signal. The model/classifier may include/may be trained on a dataset or datasets including, for example, properties/characteristics of charged/aminoacylated polynucleotides/nucleic acids, such as aminoacyl-tRNAs, and properties of uncharged/non-aminoacylated polynucleotides/nucleic acids, such as uncharged tRNAs, passing through a nanopore, and used to characterize individual and/or pools/populations of polynucleotides/nucleic acids, such as a pool/population of tRNAs, regarding properties, such as state of charging/aminoacylation, nature of the amino acid/peptide the polynucleotide/nucleic acid/tRNA is charged with, i.e., characterize what amino acid/peptide the polynucleotide/nucleic acid/tRNA is charged with, determine if a polynucleotide/nucleic acid/tRNA is mischarged, i.e., characterize if a tRNA is not charged with/aminoacylated an amino acid appropriate for its codon, which may be a result of, e.g., but not limited to, nutrient depletion, or the like.

[0067] The dataset or datasets used for training may include properties/characteristics that have been determined/measured for, e.g., fully charged tRNA libraries and deacetylated/uncharged tRNA libraries, that may serve as a "ground truth" dataset or datasets. The dataset(s) and properties determined/measured may be derived from biological sources, such as, but not limited to, a pool of all aminoacylated- tRNAs and deacylated/uncharged tRNAs obtained from an organism/subject, unique/uniquely labeled aminoacyl-tRNAs, synthetic tRNAs charged with defined amino acids, and their deacylated/uncharged counterparts. The properties of the training dataset(s) may include, e.g., measurements of ion current flow and/or dwell time for polynucleotides/nucleic acids of known charging status/characteristics passing through a nanopore for use with any suitable training procedure as would be appreciated by one of skill in the art that can result in a trained ML model/classifier suitable for characterizing, e.g., charging/aminoacylation status of a nucleic acid/polynucleotide/tRNA, charging/aminoacylation levels (extent to which a pool of nucleic acids/polynucleotides/tRNAs is charged/aminoacylated vs. uncharged) of a nucleic acid/polynucleotide/tRNA, mischarging of a nucleic acid/polynucleotide/tRNA, and/or effects on tRNA charging/mischarging resulting from changes in environmental and/or growth conditions, such as, but not limited to, nutrient depletion, nutrient starvation, and/or stress. [0068] Training of the model/classifier may be performed on any suitable hardware/software platform that would be appreciated by one of skill in the art without limitation. The ML model/classifier may be trained with any suitable and currently available commercial or open source software packages/API wrappers, such as, but not limited to, Remora or Dorado, operating on a computer apparatus/system, or the model/classifier may be trained on a custom software package/API wrapper developed specifically for characterizing polynucleotides/nucleic acids/tRNAs according to embodiments of the inventive concept, e.g., determine whether the polynucleotide/nucleic acid/tRNA is charged/aminoacylated vs. uncharged. In addition to classification tasks, embodiments and implementations of the classifier/trained model may include regression models, ensemble methods, and transformer-based architectures that, for example: predict amino acid identity from signal; quantify charging fractions across reads or populations; and handle multi-modal data integration (e.g., signal + sequence + metadata). The computer apparatus is likewise not limited, and may any type of system, conventional or custom, as would be appreciated by one of skill in the art.

[0069] The trained classifier may include a threshold or cutoff value for analyzing a property measured, that provides information regarding a characteristic of interest related to a polymer/polynucleotide/nucleic acid/tRNA, such as, but not limited to, charging/aminoacylation status of the polymer/nucleic acid/tRNA. For example, in some embodiments, the threshold or cutoff value sets forth a value above which a measured property is indicative that, e.g., a tRNA is charged, and a value below which a measured property is indicative that, e.g., a tRNA is uncharged.

[0070] Additional embodiments of the inventive concept may include kits for performing the methods of the inventive concept. Exemplary kits may include, any one of, some of, or all of the following, for example: oligonucleotide/oligonucleotide adapters for ligating to/attaching to a polynucleotide/nucleic acid, such as a tRNA at the 5' end, 3' end, or an amino acid attached to the 3' end of the polynucleotide/nucleic acid/tRNA to prepare constructs including at least one oligonucleotide adapter attached to one end of the polynucleotide/nucleic acid as described herein; reagents for ligating/attaching oligonucleotides/oligonucleotide adapters included with the kit to the polynucleotide/nucleic acid/tRNA; a nanopore(s)/protein pore(s) for characterizing the polynucleotide; a moiety for controlling translocation through the nanopore/protein pore; a calibrant and/or internal standard, e.g., a synthetic tRNA with known charging status suitable for, e.g., downstream normalization and/or quality control during sequencing and signal analysis; and/or containers/vessels suitable for carrying out the method of the inventive concept. In some embodiments, the oligonucleotide/oligonucleotide adapter for ligating/attaching to the 3' end of the polynucleotide/nucleic acid/tRNA may be oligonucleotide including an activated 5' phosphate for attaching to an amino acid/peptide of an aminoacyl-tRNA, such as a 5'-phosphorimidazolated oligonucleotide. In some embodiments, the oligonucleotide/oligonucleotide adapter for ligating/attaching to the 5' end of the polynucleotide/nucleic acid/tRNA may be oligonucleotide can associate/hybridize with the CCA-3' end of a tRNA and/or sequences of the oligonucleotide/oligonucleotide adapter for ligating/attaching to the 3' end of the polynucleotide/nucleic acid/tRNA. Kits of the inventive concept may include instructions describing how to perform the methods of the inventive concept as described herein. [0071] The following examples are provided to assist in describing the present inventive concept. The details of these examples and the general description of the examples are for description purposes only and should be seen or taken to limit the scope of the inventive concept in any way.

EXAMPLES

NANOPORE SEQUENCING OF INTACT AMINOACYLATED tRNAS [0072] Described herein is a nanopore sequencing approach (“aa-tRNA-seq”) that directly captures information on tRNA sequence, modifications, and aminoacylation in a single read. The method enables selective capture of tRNAs based on aminoacylation status, selectively embedding the amino acid of aa-tRNAs within the adapter-ligated tRNA molecule. In a separate step, non-aminoacylated tRNA are captured, facilitating comparative analyses of tRNA charging. Nanopore signals produced by 20 proteinogenic amino acids using synthetic tRNA were characterized, and these signals were leveraged to train a recurrent neural network (RNN) to discriminate aminoacylated tRNAs from their uncharged counterparts, and extended this approach for pairwise amino acid classification. The method was applied to study changes in budding yeast tRNA populations during nutrient limitation and in conditions of tRNA hypomodification, confirming known and identifying unexpected changes in tRNA aminoacylation and abundance during these stress conditions.

RESULTS

A chemical ligation approach enables selective capture of intact aminoacylated tRNAs [0073] While investigating prebiotic roles of aminoacyl-RNAs, a splinted ligation reaction was developed that generates amino acid-bridged chimeric RNA molecules using a 5'-phosphorimidazole activated oligoribonucleotide ³³³⁴. It was realized that aminoacyl-tRNAs were analogous to these substrates, so it was tested and found that a synthetic tRNA-Gly-GCC aminoacylated with glycine or lysine using the Flexizyme ³⁵ underwent chemical ligation with moderate kinetics, while the non-aminoacylated tRNA yielded no detectable product (FIG. 5, Panel A). To accelerate the reaction between the a-amino group of the aminoacyl-tRNA and the activated adapter, 1-(2- Hydroxyethyl)imidazole (HEI) was included as an organocatalyst ³⁶ (FIG. 1, Panel A, FIG. 5, Panel B), and subjected S. cerevisiae tRNA to a 30 minute chemical ligation at pH 5.5. Under these conditions, efficient ligation of aminoacylated budding yeast tRNA was achieved as measured by acid northern blot, with <0.1 % background ligation to a chemically deacylated tRNA control for tRNA-Gly-GCC (FIG. 1 , Panel B). To examine the efficiency of this reaction on budding yeast tRNA substrates, the same membrane from FIG. 1 , Panel B was stripped and reprobed using oligonucleotide probes complementary to isodecoders from each of the 20 tRNA isoacceptor families in yeast (FIG. 6). Of the 16 tRNA isodecoders that were separated sufficiently to enable densitometric quantification, the percent of aminoacylated species shifted upon chemical ligation ranged from 62-100%, with quantitative ligation of arginyl, asparaginyl, cysteinyl, glutaminyl, glycyl, and lysyl tRNA species (FIG. 5, Panel B). Using a fluorescently labeled tRNA analog, it was confirmed that this HEI-catalyzed reaction yields nearly quantitative ligation in vitro to substrates that were Flexizyme-charged with 13 different amino acids (FIG. 5, Panels C,D). A chemical-northern blot simplifies analysis of aminoacylated tRNA

[0074] The analysis of tRNA aminoacylation by acidic northern blotting requires careful handling to preserve the labile ester linkage, using a large format, low pH polyacrylamide gel run in cold, acidic buffer conditions to provide adequate resolution of charged and uncharged tRNA, which is achieved by >12 hours of electrophoresis depending on the isodecoder of interest ²⁴, though it was found that some aa-tRNAs (e.g., Glu-UUC, Asp-GUC) are poorly resolved (FIG. 6). Chemical ligation of aminoacyl- tRNA stabilizes the ester linkage ^{3337 38} and significantly increases its size, enabling robust separation from non-acylated tRNA by acidic northern blot (FIG. 1 , Panel B). These features motivated a simpler approach to separate charged and uncharged tRNA using non-acidic denaturing polyacrylamide gel electrophoresis. In this “chemicalcharging northern” (FIG. 1 , Panel C) budding yeast tRNA was chemically ligated as in FIG. 1 , Panel B, followed by a ~30 minute electrophoretic separation on a standard TBE-urea mini-gel, membrane transfer, and probe hybridization for the same glycyl isodecoder. It was found that 35% of Gly-GCC tRNA displays a gel shift after chemical ligation (FIG. 1 , Panel C, lane 4, consistent with the value from FIG. 1, Panel B), with 7% background ligation to the deacylated sample (lane 2) due to incomplete deacylation (FIG. 5, Panels E,F ³⁹). In the absence of the phosphorimidazole-activated 3' oligoribonucleotide (lanes 1 & 3), no gel shift for the Gly-GCC tRNA was apparent (FIG.

1 , Panel C).

Aminoacylated tRNAs can be analyzed by nanopore sequencing

[0075] It was next assessed whether synthetic tRNA ligated to an activated 3' oligoribonucleotide via a bridging amino acid was amenable to nanopore sequencing. It was confirmed that adapters for direct tRNA nanopore sequencing ³² can be attached to budding yeast tRNA via sequential chemical and enzymatic ligations of the 3' and 5' sequences, respectively (FIG. 1, Panel D), suggesting a clear strategy for the preparation of nanopore sequencing libraries containing aminoacylated tRNAs (FIG. 1 , Panel E). Gly-tRNA charged with the Flexizyme ³⁵ was synthesized, chemically ligated this to the same phosphorimidazolated 3' adapter, gel purified the ligation product, and enzymatically ligated the 5' adapter using T4 RNA ligase 2 (RNL2). A nanopore direct RNA sequencing library was then prepared from the individual ligated Gly-tRNA using the RNA004 chemistry from Oxford Nanopore Technologies (ONT). FIG. 1 , Panel F shows the difference in ionic current between a synthetic tRNA and the same sequence charged with glycine, with lower current for Gly-tRNA spanning multiple nucleotides as the aminoacylated molecule is pulled into the nanopore from its 3' end during motor- catalyzed translocation.

[0076] We next synthesized tRNA charged with each of the 20 standard proteinogenic amino acids using the Flexizyme and subjected them to the same library preparation. Quality control assessment revealed no major issues for 18 of 20 of these libraries; however, 41 .0% of Asn-tRNA reads and 27.6% of cysteinyl-tRNA reads terminated aberrantly, consistent with these tRNAs becoming stuck in nanopores and subsequently ejected (“unblocked”) at higher rates than those charged with other, bulkier amino acids or the uncharged tRNA control (FIG. 7, Panel A). As the Asn-tRNA and Cys-tRNA libraries also had lower sequencing yields, we sought to resolve this issue and tested multiple strategies, illustrated in FIG. 7, Panel B. For Cys-tRNA, either (i) alkylation with chloroacetamide after chemical ligation or (ii) omitting the enzymatic ligation to the 5' adapter resolved the pore blocking (FIG. 7, Panel C). This second strategy also reduced pore blocking for tRNA-Asn, suggesting that the thiol of Cys and amide of Asn might interact with the 5' RNA adapter, leading to the molecule becoming stuck in the nanopore during sequencing. Further testing revealed that changing the 5' adapter from an all-RNA oligonucleotide to a DNA oligonucleotide with five ribonucleotides at its 3' terminus (i.e., nearest the amino acid) ameliorated the issue for both Cys and Asn (FIG. 7, Panels C,D); therefore, this DNA/RNA hybrid adapter was used in all subsequent experiments.

Discrimination of charged and uncharged biological tRNAs by nanopore sequencing [0077] Modified basecalling of nanopore sequencing signals is achieved by training neural networks ⁴⁰; in the case of direct RNA sequencing, detecting the presence of a modification via comparison to an unmodified control is often a more trivial computational challenge than precise characterization of a modification’s identity ^41-44. To extend this concept to aminoacylated tRNA sequencing, we sought to classify charged and uncharged tRNA sequencing reads, with the goal of developing a one-pot approach for nanopore tRNA sequencing to capture charged and uncharged molecules within the same sequencing library. Separate “ground truth” libraries were prepared containing either (7) chemically ligated, charged tRNA, or (//) deacylated, enzymatically ligated tRNA from budding yeast (FIG. 2, Panel A). tRNAs were ligated to 3' adapters containing unique sequences for charged and uncharged molecules, and a common 5' adapter incorporating the design changes (DNA/RNA hybrid) from in-FIG. 7, Panel B, and sequenced these libraries on Prometh ION RNA flow cells.

[0078] Informed by our synthetic tRNA sequencing results (FIG. 1 , Panel F), a recurrent neural network was trained on this dataset using the Remora software from ONT (FIG. 2, Panel B), which predicts the presence of modified nucleotides using ionic current signals generated during nanopore sequencing. For training, a six-nucleotide signal window was defined spanning the invariant CCA at the 3' terminus of mature tRNA (“CCA-G-GC”, where “G” represents the first base of the adapter most affected by the amino acid, and “GC” represent the next two bases of the adapter; depicted in FIG. 2, Panel C), as the sequence in this region would be shared across all tRNA isodecoders. A Remora dataset was generated from 80% of these reads, with labels defined by the library of origin (7 and ii, above), and then trained and evaluated a model on the reserved 20% test set. During signal-anchored inference, Remora models output predictions for “modified” positions in the ML (“modification likelihood”) tag of a BAM file; these values range from 0-255, where lower values are more likely to be canonical nucleotides. Inspection of these values in our libraries revealed a bimodal distribution, with 96.6% of reads from the charged library (7) bearing M L scores >200, compared to

1.2% of in the deacylated sample (ii) (FIG. 2, Panel D).

[0079] A second biological replicate was generated of the libraries in FIG. 2, Panel A for model validation, performing reference-anchored inference using the trained model above and using a ML cutoff of >199 to classify aminoacylated tRNAs. FIG. 2, Panel E illustrates the Remora model’s performance on this new dataset, with an F1 score of 0.966. This signal-based approach substantially outperformed an alignment-based one using unique 3' adapter sequences to discriminate charged and uncharged tRNAs, which yielded an F1 score of 0.322 with low sensitivity for identifying charged tRNA reads (FIG. 2, Panel F), and which was attributed to increased base-calling error caused by the embedded amino acid. Based on its clear improvement over the alignment-based approach, we focused on signal-based classification of tRNA aminoacylation state.

Measurement of tRNA aminoacylation with aa-tRNA-seq

[0080] To determine whether this approach could report on tRNA charging levels, replicate aa-tRNA-seq libraries were generated from wild-type budding yeast under conditions identical to FIG. 1, Panel B (wild-type yeast in rich media), enabling an orthogonal comparison of tRNA aminoacylation measured by acidic northern (FIG. 1 , Panel B) and model-based classification of nanopore sequencing reads. Across the 16 tRNA isodecoders quantifiable by densitometry (FIG. 5, Panel B), a Pearson’s correlation of 0.68 was observed; however, estimates of isodecoder charging by sequencing were uniformly lower than charging levels measured by northern (FIG. 3, Panel A). While in principle this difference might be explained by differences in the rates of aberrant read termination between charged and uncharged molecules, the magnitude of this result was inconsistent with our observations in synthetic tRNA sequencing data after optimization of the 5' adapter sequence (FIG. 7). Therefore, the initial dataset used for Remora model training was re-analyzed, comparing the translocation time for tRNA reads in the first library prepared via chemical ligation to aminoacylated tRNA to the second library where tRNAs were deacylated and enzymatically ligated. This analysis revealed that the total translocation time for aminoacylated tRNA — across all isodecoders — is on average ~1 .2-fold longer than their uncharged counterparts (FIG. 8), which is likely the primary driver of the underestimation of charging observed in aa-tRNA-seq libraries (FIG. 3, Panel A): in a complex mixture of biological tRNAs, uncharged tRNAs are sequenced faster than charged tRNAs, leading to a sampling bias toward uncharged tRNAs. As such, while absolute quantitation of aa-tRNAs requires further optimization, we focused on relative quantitation of tRNA abundance and charging. aa-tRNA-seq detects changes in tRNA aminoacylation during nutrient stress

[0081] It was next sought to assess whether aa-tRNA-seq could detect dynamic changes in tRNA charging and focused on nutrient limitation, which causes rapid changes to the aminoacyl-tRNA pool. A S. cerevisiae leucine auxotroph was subjected to leucine starvation for 15 minutes, and found that, as previously reported ¹¹ , this caused rapid depletion of aminoacylated leucyl-tRNA isodecoders (FIG. 3, PANELS B- D). Using a chemical-charging northern (FIG. 3, Panel B), a mean 28.4% decrease was observed in the abundance of aminoacylated leucyl-tRNA. We performed aa-tRNA-seq on the same material, and examined global changes in tRNA abundance and aminoacylation upon leucine starvation. This analysis revealed a significant decrease in the charging of all four leucyl tRNA isodecoders present in budding yeast, while most other tRNAs remained unaffected. Surprisingly, a concomitant increase in the charging of Ala-tRNA isodecoders was also found (3.97-fold for Ala-UGC, p < 0.001 ; 2.12-fold for Ala-AGC, p = 0.08; BH-adjusted Z test; FIG. 3, Panels C,D), an effect that was not detected in earlier microarray-based experiments examining tRNA charging in response to leucine starvation ¹¹. This isodecoder-specific effect was validated by chemicalcharging northern, where a mean 2.62-fold increase was observed for Ala-TGC compared to a mean 1 .10-fold increase across three other unaffected isodecoders (FIG.

9)- aa-tRNA-seq detects interdependence between tRNA modifications and aminoacylation during rapid tRNA decay

[0082] Hypomodified tRNAs are susceptible to “rapid tRNA decay” (RTD) in budding yeast, fission yeast, and bacteria ^45-53. For example, Val-AAC tRNA lacking 5-methyl cytidine and 7-methyl guanine in trm8A trm4A budding yeast is rapidly destabilized and degraded at high temperature by 5'-3' exonucleolytic decay ⁵⁴. A trm4A trm8A strain was cultured and a control bearing an additional disruption of MET22 at the permissive temperature of 28^QC followed by a shift to the non-permissive temperature of 37^QC for three hours. Deletion of MET22 causes accumulation of pAp (adenosine 3', 5' bisphosphate, a competitive inhibitor of 5'-3' exonucleases ⁵⁵), suppressing RTD ⁵⁴. Small RNA was isolated from each strain and temperature in biological triplicate, and prepared this material for analysis by chemical-charging northern and aa-tRNA-seq. Both approaches confirmed defects in stability and aminoacylation for Val-AAC ⁵⁴. By chemical-charging northern, aminoacylation levels for Val-AAC drop ~2-fold upon shift to the nonpermissive temperature in trm8A trm4A cells, and this effect was suppressed by met22A (FIG. 3, Panel E). This effect is readily detected by aa-tRNA-seq, where

Val-AAC undergoes significant changes in both tRNA abundance (~3.6-fold reduction, p = 2.93 x 10⁸, BH-adjusted Z test) and aminoacylation (~3.3-fold decrease, p = 1 .99 x

10⁶, BH-adjusted Z test) upon temperature shift (FIG. 3, Panel F). Consistent with previous reports ⁴⁵, in yeast strains bearing single deletions of TRM4 or TRM8, Val-AAC aminoacylation remains near constant upon shift to the non-permissive temperature

(FIG. 10, Panels D,E).

[0083] Statistically significant increases in aminoacylation were also observed (2.54- fold for Gly-GCC, p-value = 0.009; 2.73-fold for Gly-CCC, p-value = 0.004, BH-adjusted Z test) for two glycyl-tRNA isodecoders upon a shift to 37^QC (FIG. 3, Panels F,G), which were eliminated by MET22 co-deletion in our sequencing data (FIG. 3, Panel H). However, validation of this result by chemical-charging northern was less conclusive. While a 1 .22-fold increase in Gly-GCC aminoacylation was detected in the double mutant, aminoacylated Gly-GCC tRNA in the trm8A trm4A met22A triple mutant also increased by a similar magnitude (FIG. 10, Panel A). Two additional isodecoders, Cys- GCA and Leu-CAA, also displayed more modest (~1 .1 -fold increases) changes in aminoacylation upon temperature shift (FIG. 10, Panels B,C). To investigate this result further, we again turned to single deletions of TRM4, TRM8, and MET22. For Gly-GCC, the mean change in aminoacylation upon temperature shift was 1 .54-fold for trm4A,

1 .20-fold for trm8A, and 1 .27-fold for met22A, respectively (p-value of 0.09 by ANOVA,

FIG. 10F,G). While not statistically significant across biological replicates, the results hint at a potential distinct role for TRM4 in modulating Gly-GCC aminoacylation during temperature stress, in contrast to TRM8, which is not known to act directly on Gly-GCC 56 [0084] RTD has also been described in tan1 trm44A cells, which are temperature sensitive due to lack of acetylation at C12 by TAN1 and loss of methylation at U44 by TRM44; these enzymes share activity on multiple serine and leucine isodecoders in budding yeast ⁵⁴⁵⁶. As in trm8A trm4A, temperature sensitivity is accompanied by exonuclease-mediated decay of specific tRNA isodecoders (Ser-UGA, Ser-CGA) and both the molecular and growth phenotypes are alleviated by co-deletion of MET22⁵⁴⁵⁷. We grew tanl trm44A and tanl trm44A met22A cells under the same RTD- inducing conditions described above, isolated small RNA, and again performed chemical ligation and analysis by chemical-charging northern and aa-tRNA-seq. Our results indicate that the targets of RTD in tan1 trm44A cells are restricted to the tRNA species originally identified; however, aminoacylation of these hypomodified tRNAs is unaffected upon temperature stress. After three hours of temperature stress, we detected modest decreases in abundance of both serine tRNA isodecoders previously reported to be affected in tan1A trm44A cells via aa-tRNA-seq, with levels of Ser-CGA (the more strongly affected isodecoder in previous studies) decreasing by a statistically significant 22.2% and Ser-UGA showing a 13.8% reduction in abundance upon temperature stress (FIG. 11 , Panels A-C). When validated by chemical-charging northern, tanlA trm44A cells displayed 6-1 1% drops in Ser-UGA abundance upon shift to the nonpermissive temperature (FIG. 12, Panels A-C), while levels of the unaffected isodecoder Leu-CAA remained constant or increased.

A machine learning approach to discriminate embedded amino acids in aa-tRNA-seq [0085] Because the identity of the amino acid on an aminoacylated tRNA is invisible to other tRNA sequencing approaches, a major advance would be the detection of misaminoacylation events directly from high throughput tRNA sequencing libraries. Towards this goal, detailed analysis was performed of signals produced during nanopore sequencing for aminoacyl-tRNA charged with each of the 20 proteinogenic amino acids using the Flexizyme system. While nanopore basecallers and other machine learning approaches for detection of modified bases are typically trained on ionic current signatures produced at specific residues ⁴⁴’⁵⁸’⁵⁹, the raw signal generated during nanopore sequencing is a composite of changes in current and translocation speed (measured in “dwell time”, the time between inferred translocation states). Because they are sequence-identical (with the exception of Cys- and Asn-tRNA libraries, which were prepared using the DNA/RNA 5' adapter sequence tested in FIG. 8 and used in all biological sequencing experiments), our synthetic aa-tRNA-seq libraries enable the isolation of amino acid specific signals by comparing each of the 20 aa- tRNAs to an uncharged control.

[0086] Aminoacylated tRNAs yield specific distortions in nanopore signals, generating unique signatures that vary between amino acids. FIG. 4, Panel A displays the mean dwell time in milliseconds for each of our 20 synthetic aa-tRNA libraries and an uncharged control tRNA. Large increases in dwell time for charged tRNAs were observed at a position located 9 nucleotides downstream from the amino acid (position 86), with dwell times exceeding 1 second for 9 of 20 amino acids. Because nanopore direct RNA sequencing proceeds in a 3' to 5' direction, it is speculated that this signal represents specific interactions between amino acid and the motor protein as the 3’ adapter is transiting through the nanopore reader head. FIG. 4, Panel B shows the relative change in normalized current for each of the aminoacylated tRNA reads over this same window, compared to the uncharged tRNA control. Charged tRNA libraries display lower mean current values than the non-aminoacylated substrate across most of this region, suggesting that amino acids occlude ionic flow as they transit through the helicase/pore assembly. Consistent with this explanation, the largest reductions in mean current were observed at the precise site of aminoacylation for the bulkiest amino acids, indicating that the largest distortions in ionic current distortions occur within the narrowest aperture of the nanopore itself.

[0087] The correlation between the largest current and dwell time effects and various amino acid properties were examined. While the putative helicase interactions at position 86 are poorly correlated with amino acid volume and molecular weight (FIG. 4, Panels C,D), strong correlations were identified between dwell time and hydrophobicity (FIG. 4, Panel E), as well as correlations between amino acid mass and volume, and the observed current differences at position 77 (the position of the amino acid; FIG. 4, Panels F,G). Together, these observations likely reflect straightforward physical occlusion of ion flow within the nanopore reader head and more complex interactions between the tRNA-embedded amino acid and the motor protein.

[0088] Closer examination of the data revealed substantial variation in the magnitude of current (FIG. 13) and dwell time (FIG. 14) signals produced by different synthetic aminoacyl-tRNAs. We trained 380 pairwise models to discriminate one AA from another (FIG. 4, Panel I), and evaluated their performance on our synthetic data (FIG. 4, Panel J). The performance of these models was generally high with a median F1 score of 0.86, indicating that differences in nanopore current between several pairs of amino acids represent a strong signal for classification. Notable underperforming outliers included isoleucyl-tRNAs, which are poorly distinguished from other hydrophobic side chains, and other specific comparisons (e.g., Phe-Tyr, Pro-Ala, Cys-Arg). Improvement of these models by explicit inclusion of dwell time information during model training are being explored, and these models are applied to understand specific cases of potential tRNA misaminoacylation (FIG. 3, Panels D,G).

DISCUSSION

[0089] Transfer RNAs are long-lived and undergo repeated cycles of charging and deacylation throughout their lifetime. While tRNA biogenesis is recognized as an intricately coordinated process — spanning transcription, processing, modification, and aminoacylation — less is known about the molecular transformations that tRNAs undergo throughout their functional lifetime and the specific events that trigger their turnover. A novel chemical ligation approach was developed to selectively capture aminoacylated tRNAs and used it to study tRNA aminoacylation in a variety of contexts. Increased accuracy of aa-tRNA identification may further enhance our understanding of how tRNA modifications act as determinants or anti-determinants of synthetase recognition and specificity. Moreover, by enabling simultaneous screening of tRNA modifications, charging, and stability, this approach provides a versatile framework for engineering synthetic or therapeutic tRNAs with precisely tuned properties ^60-64. Direct identification of amino acids and their corresponding tRNA sequences could also advance studies on aminoacyl-tRNA synthetase evolution and engineering, bypassing the indirect readouts and negative selections commonly employed by current synthetase engineering approaches ⁶⁵⁶⁶.

[0090] Chemical ligation was leveraged to simplify the analysis of aminoacylated tRNAs via northern blotting (FIG. 1 , Panel C). This approach offers key advantages, including simplified electrophoretic separation of chemically ligated species, and is therefore a compelling alternative to more laborious techniques. However, it is important to note that under the existing reaction conditions, there is some variability in ligation efficiency among biological tRNA isoacceptors and only some were ligated quantitatively (FIG. 5, Panel B), limiting the performance of this method compared to a traditional acidic northern blot. It is also noteworthy that chemical ligation is likely useful beyond the capture of aminoacylated tRNA: a variety of nucleophiles react with phosphorimidazole-activated oligonucleotides, including RNA terminal hydroxyl groups at elevated pH ⁶⁷. Juxtaposition of the phosphorimidazolide near the a-amino group of the aa-tRNA was facilitated by splinted base-pairing to the 3'-CCA overhang (FIG. 1 , Panel A), but one could imagine other sequence-specific targeting scenarios, including the 3’-ends of mRNA poly-A tails.

[0091] Next, chemical ligation was adapted to enable direct nanopore sequencing of aminoacylated tRNA ligated to a downstream oligonucleotide (“aa-tRNA-seq”). A Remora classification model was trained to distinguish charged and uncharged tRNAs from nanopore sequencing data, and showed that this signal-based classification approach outperformed an alignment-based one (FIG. 2). Our application of aa-tRNA- seq confirmed known effects of hypomodification and nutrient deprivation on tRNA stability and aminoacylation (Fig. 3), revealing that despite the broad specificity of TRM8 and TRM4, their co-deletion uniquely impacts the stability of Val-AAC (FIG. 3, Panel F). Similarly, the combined impact of TAN1 and TRM44 deletion is largely restricted to Ser-OGA and Ser-UGA stability (FIG. 11). Significantly increased aminoacylation for two isoacceptor families was also found in cases of hypomodification (Gly-GCC and Gly-CCC, FIG. 3, Panel G) and nutrient deprivation (Ala-TGC, FIG. 3, Panel D). These may simply be increases in cognate aminoacylation of these isodecoders that were not previously observed. Alternatively, they may represent new cases wherein a lack of modification or charging enables misaminoacylation of Gly and Ala isodecoders. Methionine misaminoacylation is a common response to stress ⁶⁸ wherein the methionyl tRNA synthetase non-specifically charges several tRNA isodecoders with methionine. However, the pattern of misaminoacylation we observe is restricted to a few tRNAs, suggesting an alternative mechanism.

[0092] Like other tRNA sequencing methods, aa-tRNA-seq has tradeoffs and technical biases. First, intrinsic differences were identified in the total translocation time between charged and uncharged tRNA during nanopore sequencing (FIG. 8), which is a barrier to absolute quantitation of tRNA charging. Second, under these reaction conditions not all biological tRNA isoacceptors are quantitatively ligated (FIG. 5, Panel B), reflecting an additional source of bias that may be further optimized. Ligation inefficiency reflects a broader challenge: enzymatic adapter ligation bias has complicated other tRNA sequencing methods, often making comparisons of tRNA abundances within the same sample unreliable ⁶⁹⁷°. While methods relying on periodate oxidation to distinguish charged from uncharged tRNAs have proved valuable to the field, they also face technical challenges, including tRNA damage during periodate treatment ²¹ , variability in the sensitivity of tRNA 3' termini to oxidation, and inconsistency in the number of nucleotides removed during treatment ⁷¹. Additionally, as periodate-based methods require cDNA generation, they are subject to biases from tRNA modifications that inhibit reverse transcriptase processivity. In protocols where these modifications are instead removed, such methods can introduce additional RNA damage from harsh buffer conditions during enzymatic pretreatment ^{72 74}.

[0093] Chemically ligated aa-tRNAs generate discrete signals during nanopore sequencing due to interactions between the embedded amino acid and the motor protein and nanopore architecture (FIG. 4). This unique feature enabled accurate identification of bona fide aa-tRNAs while minimizing false positives: we classified 3.6% of reads from the "charged-only" validation library as uncharged. However, some of these may be true non-acylated tRNA that were enzymatically ligated to hydrolyzed phosphorimidazole-adapter, which has a 5'-phosphate competent for T4 RNL2 ligation. While our current approach demonstrates that different amino acids produce distinct signal properties that enable their discrimination in pairwise comparisons (FIG. 4, Panel J), further advances in our machine learning approach may enable the direct, de novo identification of amino acids in biological samples, unlocking new questions about tRNA charging and misaminoacylation.

[0094] It is anticipated that incorporation of translocation time information into models for classifying aminoacylated tRNAs will generate additional improvements to aa-tRNA-seq. Dwell time provided useful information in the detection of RNA modifications using the previous direct RNA (RNA002) ONT chemistry, including pseudouridine ⁷⁵, 2'-O-methyl ⁷⁶, and 2'-phosphate ³¹ modifications, but existing approaches for training models on nanopore signal do not leverage this information directly, due to the fact that Remora and other software re-anchor ionic current information onto an aligned sequence. While dwell time information is retained in this process, it is transformed and thereby de-emphasized in model training. Notably, the increased dwell times observed for aminoacylated tRNAs — likely due to unique helicase interactions— are a robust signal in our data. However, these effects are less strongly correlated with amino acid properties than the current differences at the aminoacylated position (FIG. 4, Panels C-H), where we have focused our model training.

[0095] The translocation rate for aa-tRNAs also impacts the pore blocking effects described for Cys- and Asn-tRNA (FIG. 7), as read ejection (“unblocking”) is initiated during nanopore sequencing when a constant signal (indicating a stalled molecule) exceeds a set time threshold. While pore blocking issues were resolved via optimization of the 5' adapter sequence, it is not fully understood why chemical ligation of these substrates produced these artifacts, or why they were resolved by the substitution of deoxyribonucleotides in our 5' RNA/DNA splint adapter. While Asn-tRNA yielded significant pore blocking, Gln-tRNA did not (FIG. 7, Panel A), suggesting an issue beyond simply the presence of an amide side chain. It is noted that the Asn side is uniquely capable of cyclization rearrangements during intein catalysis, which may contribute to its unique pore blocking phenotype ⁷⁷.

[0096] Transfer RNA modifications are installed in evolutionarily conserved but incompletely understood circuits ^{78 79}. Described links between tRNA modifications and aminoacylation ⁴⁵ remain sparse and poorly characterized, due in part to the lack of incisive and accessible tools to study these relationships. Comprehensive identification and characterization of circuits linking modification and aminoacylation along with tRNA abundance will require further optimization of aa-tRNA-seq and ongoing work to map the 67+ unique RNA modifications present in the tRNA epitranscriptome ⁵⁶, which will likely require a combination of nanopore sequencing, mass spectrometry ⁸⁰⁸¹ , and other technologies, with the aim of characterizing the complete collection of tRNA modifications (using lower throughput approaches) and understanding the signals they produce during high throughput, direct RNA sequencing experiments ⁸².

MATERIALS AND METHODS

Preparation and chemical ligation of synthetic aminoacylated tRNAs

[0097] Oligonucleotides listed in Table 2 were either purchased from IDT or synthesized on the K&A H-6 RNA/DNA synthesizer. Phosphoramidite coupling times and the remaining synthesis method parameters were as instructed by the manufacturer (ChemGenes and Glen Research). After solid-phase synthesis, oligonucleotides were cleaved and the nucleobases deprotected as recommended by ChemGenes and Glen Research. The cleaved and deprotected solutions were evaporated using a speed-vac for 2 hours followed by overnight lyophilization. The dry material was dissolved in 100 pL DMSO to which 125 pL of TEA. 3HF was added followed by incubation at 65^QC for 2.5 hours. The fully deprotected oligonucleotides were precipitated with 0.1 volumes of 5 M ammonium acetate and 5 volumes of cold isopropanol. The precipitated material was dissolved in 5 mM EDTA, 99 % v/v formamide and purified by denaturing PAGE. The desired gel bands were visualized by UV shadowing, cut out, crushed, and soaked in 2 mM EDTA, 5 mM sodium acetate on a rotator overnight. The rotated solutions were filtered through a 5 pm syringe filter after which the filtered solutions were concentrated using Amicon MWCO filters. The concentrated solutions were finally precipitated using 0.1 volumes of 3 M sodium acetate and 5 volumes of ethanol, washed twice with 80 % v/v ethanol, and air dried.

[0098] The 3,5-dinitrobenzyl esters of amino acids (DBE-aas) were synthesized as described in ref. 35 with the following modifications:

1. Boc protecting groups were removed by dissolving the dry crude DBE-Boc-aa material in 2 mL neat TFA and incubating it at room temperature for 10 minutes. The TFA was removed under a stream of nitrogen and the deprotected product was washed twice with diethyl ether. The diethyl ether was removed under vacuum and the final DBE-aa product was dissolved in DMSO and used in the aminoacylation assays.

2. Boc-Ser, Boc-Thr, and Boc-Tyr (Chemlmpex) were purchased with O-te/ -butyl protection on the side chain. The deprotection was performed in 90:10 TFA:triethylsilane for 2 hours.

3. Boc-Met (Chemlmpex) was used without additional side chain protection, but during the TFA deprotection two side products were observed: oxidation to produce a DBE-Met dimer and tert-butylation of the sulfur. This necessitated reversed-phase purification. RediSep Gold® C18 Reversed Phase Column was used with the 5-90 % gradient of solvent B (solvent A = 2 mM TEAB pH 8; solvent B = acetonitrile).

4. Boc-GIn and Boc-Asn were purchased with Xan protection on the side chain amide (Chemlmpex). DBE-Asn required reversed-phase purification. RediSep Gold® C18 Reversed Phase Column was used with the 5-90 % gradient of solvent B (solvent A = 2 mM TEAB pH 8; solvent B = acetonitrile). DBE-Asn additionally requires immediate use after purification due to the presumed rapid intramolecular attack of the side chain amide onto the activated ester.

5. Boc-Cys was purchased with Trt protection on the side chain thiol (Chemlmpex). The deprotection was performed in 90:10 TFA:triethylsilane for 2 hours.

Activation

[0099] The 5'-phosphorimidazolide adapter was generated by incubating a solution containing 200 pM of the 5'-phosphorylated adapter, 100 mM imidazole pH 7, and 100 mM EDC.HCI for 2 hours at room temperature. The activated adapter was then precipitated by adding 0.1 volumes of saturated sodium perchlorate in acetone and 3 volumes of cold acetone. The pellet was washed twice with a 1 :1 v/v solution of acetone:diethyl ether followed by drying under vacuum. The activated adapter was dissolved in 1 mM imidazole pH 8 and stored at -80 ^QC until use. The same stock of the activated adapter was used throughout the experiment, but care was taken to thaw the stock immediately prior to the experiment, to store it on ice while using it, and to return it to -80 ^SC as quickly as possible.

Aminoacylation

[00100] The aminoacylation reactions containing 100 mM HEPES pH 8, 10 m /l MgCh, 40 pM synthetic tRNA, 36.7 pM dFx Flexizyme, 5 mM DBE-aa (20 % v/v DMSO), were incubated on ice for 16 hours.

Chemical ligation

[00101] The ligation reactions were set up by diluting the aminoacylation reactions 8- fold so that the final solution contained 5 pM of the aminoacylated tRNA, 50 pM 5'- phosphorimidazolide adapter, 50 pM splint (also called the 5'-adapter below), 5 mM EDTA, and 37.5 mM HEPES pH 8. The reactions were allowed to proceed for 24 hours on ice, before being diluted with an equal volume of a solution of 5 mM EDTA, 99 % v/v formamide and purified by 16 % denaturing PAGE. The ligated products were cut out from the gel, crushed, and soaked in a solution of 2 mM EDTA, 5 mM sodium acetate acidified to pH 5 on a rotator for 3 hours at 4 -C. The extracted aa-bridged tRNA products were then filtered using 0.22 pm spin filters, concentrated using Amicon 10k MWCO filters, and desalted using the Oligo Clean and Concentrator kit (Zymo Research).

Cys-bridged tRNA alkylation

[00102] After chemical ligation and gel purification, the Cys-bridged tRNA was reduced with DTT for 1 hour at room temperature. The reaction contained 1 .2 pM Cys- bridged tRNA, 50 mM HEPES pH 8, and 10 mM DTT. After the 1 hour incubation, the reduction reaction was diluted 1 .33-fold so that the final alkylation solution contained 0.9 pM Cys-bridged tRNA, 37.5 mM HEPES pH 8, 7.5 mM DTT, and 50 mM chloroacetamide. The alkylation reaction was allowed to proceed for 30 mins in the dark, after which it was cleaned up using the RNACIean XP beads (Beckman Coulter) according to the manufacturer protocol with the following change: immediately after the addition of the bead suspension to the ligation reaction, isopropanol equal to volume of the reaction+beads was added.

Nanopore library preparation and sequencing of synthetic aminoacylated tRNAs [00103] The chemically ligated tRNA products from above were enzymatically ligated to the 5'-adapter/splint for 30 minutes at room temperature. The ligation reactions contained 16 pmol of the chemically ligated tRNA, 80 pmol of the 5'-adapter, 1 x NEB T4 RNA Ligase 2 buffer supplemented with 5% PEG 8000, 2 mM ATP, 6.25 mM DTT, 6.25 mM MgCI2, and 0.5 units/pL T4 RNA ligase 2 (10,000 units/mL). The ligated material was purified using the RNACIean XP beads (Beckman Coulter) as above. This material was then prepared for nanopore direct RNA sequencing via RTA ligation, which was performed using tRNA purification specific magnetic beads (BioDynami Cat.# 40054S). The remaining library prep and nanopore sequencing was performed as described below on P2solo sequencing instruments, using MinKNOW version 23.11.7.

Synthetic tRNA mini-substrate experiments

[00104] The acceptor stem mimic oligonucleotide (Table 2) was aminoacylated with all 20 amino acids as described above. At the end of the 16 hour incubation:

1. 1 pL aliquots were diluted in 9 pL of acidic quenching buffer (10 mM EDTA pH 8.0, 1 x bromophenol blue, 100 mM sodium acetate pH 5.0, 150 mM HCI, 75 % v/v formamide) and analyzed by 20 % acidic denaturing PAGE (acidic gels contained 100 mM sodium acetate pH 5.0 instead of the usual 1 x Tris-Borate-EDTA). The acidic gels were run in 100 mM sodium acetate pH 5 at 25 W for 3 hours at 4 ^QC. Aminoacylation percentage was obtained by quantifying the per-lane normalized band intensity in the ImageQuant TL software. The amino acids that displayed sufficient aminoacylated versus nonaminoacylated gel band resolution were subjected to the next step.

2. The remaining aminoacylation reaction was immediately diluted 10-fold in the chemical ligation buffer in three separate replicates. The ligation reaction contained 1 pM of the aminoacylated RNA, 4 pM of the 5'-phosphorimidazolide activated hairpin adapter (Table 2), 10 pM of the acceptor stem mimic complement (Table 2), 200 mM HEPES pH 6.5, 5 mM MgCh, and 100 mM of 1 -(2- Hydroxyethyl)imidazole pH 6.5. After 90 mins at room temperature, 1 pL aliquots were diluted in 9 pL of acidic quenching buffer, and analyzed by standard 20 % denaturing urea-PAGE. The efficiency of the ligation reaction was obtained by quantifying the per-lane normalized band intensity in the ImageQuant TL software. The normalized ligation efficiency was obtained by dividing the fraction ligated by the fraction aminoacylated and multiplying by 100 %.

Yeast strains and growth conditions

[00105] Yeast strains used in this study are listed in Table 1 . For chemical ligation validation by acid northern (FIG. 1 , Panel B), a single colony of S288C was inoculated into YEP glucose (yeast extract, peptone, 2% glucose) and incubated at 30^QC overnight with rotation before dilution to an OD₆6o of 0.2 in YEPD media. The culture was grown to log phase shaking at 30^QC before a pellet was collected, flash frozen in liquid nitrogen, and stored at -80^QC.

[00106] For validation of chemical charging northerns (FIG. 1, Panel C), a single colony of WY798 was inoculated into synthetic complete media and incubated at 30^sC overnight with rotation. This culture was diluted to 50 ml_ of synthetic complete media the next day and allowed to shake overnight at 30⁹C before dilution to an OD₆6o of 0.2 in 100 mL synthetic complete media. The culture was grown to log phase at 30^QC, collected by centrifugation, and resuspended in 100 mL room temperature synthetic complete media. The culture was allowed to grow in the new media for 15 minutes, shaking at 30^QC before it was pelleted, washed with water, flash frozen in liquid nitrogen, and stored at -80^QC.

[00107] For the nutrient stress experiment (FIG. 3, Panels B-D), 3 colonies of WY795 were inoculated into synthetic complete media and incubated at 30^QC overnight with rotation before dilution to an OD₆6o of 0.2 in 50 mL the same media. The cultures were grown to log phase shaking at 30³C before pellets were collected by centrifugation.

Each pellet was washed by resuspension in a small volume of synthetic media lacking uracil, tryptophan, histidine, and leucine and split into two tubes. Cells were pelleted again and the wash media was removed. One pellet from each original culture was resuspended in 25 mL 30²C synthetic complete media and the other in 25 ml_ 30²C synthetic media lacking leucine. They were allowed to grow in the new media for 15 minutes, shaking at 30²C before they were pelleted, washed with water, flash frozen in liquid nitrogen, and stored at -80²C.

[00108] For the temperature stress experiment (FIG. 3, Panels E-H), 3 colonies of JMW 009 and 3 colonies of JMW 510 were inoculated into YEPD media and incubated at 30²C overnight with rotation before dilution to an OD₆6o of 0.2 in 60 mL YEPD media. The cultures were grown to log phase shaking at 30²C. At this point, 50 mL of culture was pelleted by centrifugation, washed with water, flash frozen in liquid nitrogen, and stored at -80²C. 40 mL of 40²C YP glucose media was added to the remaining 10 mL of culture and these cells were grown for 3 hours at 37²C. They were then pelleted, washed with water, flash frozen in liquid nitrogen, and stored at -80²C.

Isolation and chemical ligation of aminoacylated tRNAs from budding yeast

[00109] Yeast pellets were thawed on ice and resuspended in 400 pL of cold AES (10 rnM NaOAc pH 4.5, 1 mM ETDA pH 8, 0.5% SDS). 400 pL of cold 25:24:1 acid phenol:chloroform:isoamyl alcohol was added. Samples were vortexed for 15 seconds and allowed to rest on ice for 20 minutes, vortexing every 5 minutes. They were then spun at 18000 g for 10 minutes at 40²C and the aqueous phase was moved to a new tube.

[00110] A 0.4X volume of Ampure XP beads (Fisher Scientific A63881 ) were added to 100 pL of aqueous phase. They were rotated for 2 minutes at RT and placed on a magnet until the beads had settled. The supernatant was moved to a new tube and quantified via nanodrop. Small RNAs were isolated from 100 pg of this supernatant using a Zymo Research RNA Clean and Concentrator kit (R1018) according to the manufacturer’s instructions. Dilution of the bead supernatant for the first step of the kit was done with 10 mM NaOAc pH 4.5, not with water. Small RNA was eluted in 30 pL of 10 mM NaOAc pH 4.5 and quantified via nanodrop. It was stored at -80²C.

[00111] Two 3' DNA-RNA hybrid splint adapters were designed with different internal sequences, one to ligate to deacylated tRNAs (“uncharged 3' adapter”) and the other to acylated tRNAs (“charged 3' adapter”, see Table 2). A universal 5' adapter was designed to pair with either of the 3' adapters. Syntheses of these adapters were ordered from IDT and resuspended in water to a concentration of 2 mM. Adapters were run on a 1 .5 mM 6% TBU (Tris, boric acid EDTA) V16 polyacrylamide gel with 10 nmol loaded per lane (10-well comb). Staining was not performed and UV shadowing was used to excise the adapters. Gel slices were cut into small pieces and rotated end over end in crush + soak buffer (300 mM NaOAc pH 5.5, 1 mM EDTA pH 8.0, 0.1 % SDS) overnight at 4^QC. Nucleic acids were precipitated with 100% ethanol and resuspended in water to a final concentration of 100 pM.

[00112] Gel-purified charged 3' adapter was incubated with a 500-fold molar excess of both imidazole and EDC (1 -Ethyl-3-(3-dimethylaminopropyl)carbodiimide) for 2 hours at 25^QC to imidazolate the 5' end of the adapter. 30 pL of cold 99.5%+ acetone saturated with perchlorate and 1 mL of cold 99.5%+ acetone were added to precipitate the imidazolated adapter. The sample was incubated for 20 minutes on dry ice and then centrifuged at maximum speed for 10 minutes at 4^QC. The supernatant was removed and the pellet was washed twice with 1 mL 1 :1 acetone: diethyl ether. The pellet was dried in a speed vacuum and resuspended in 10 mM imidazole pH 7.0 to a final concentration of 200 pM.

[00113] Small RNA (15-50 pmol) was incubated in 100 mM MES pH 5.5, 2.5 mM MgCh, a 5-fold molar excess of both imidazolated 3' adapter and gel-purified 5' adapter, and 50 mM HEI pH 6.5 for 30 minutes at 25^QC establishing a phosphoramidate covalent linkage between the 3' splint adapter and aminoacylated tRNAs. Ligated products were purified by crush and soak (0.3 M NaOAc pH 5.5, 1 mM EDTA pH 8.0, 0.1 % SDS) at 4 ^QC, overnight) from a 10% TBU polyacrylamide gel, isolating the regions between 70 and 150 nts. The eluate was precipitated by addition of ethanol and GlycoBlue coprecipitant (Invitrogen) resuspended in a small volume of 10 mM NaOAc pH 4.5 and quantified via absorbance at 260 nm (Nanodrop).

Northern blotting

[00114] Transfer RNA charging was measured by acidic northern blot, resolving 75 ng of small RNA on a gel (6% 19:1 acrylamide, 0.1 M sodium acetate, pH 4.5, 8 M urea) 42 cm in length which was run at 450 V for 22 hours in a cold room. For chemical-charging northern blots, chemically ligated tRNA (220 ng) was loaded onto 10% TBU polyacrylamide gels (7.5 cm, 6% 19:1 acrylamide, 1X TBE, 8 M urea) and electrophoresed in 1X TBE at room temperature at 250 V for 40 minutes. Acid-urea and TBU gels were transferred to charged nylon membranes (Hybond N+, GE) via electroblot transfer at 1 Amp for 1 hour for acid gels and 3 mA/cm² based on the membrane area for 35 minutes for TBU gels. After transfer, membranes were UV- crosslinked at 254 nm using a 120 mJ dose and blocked in ULTRAhyb-Oligo (Thermo) before an incubation with ³²P-labeled oligonucleotide probes in ULTRAhyb-Oligo overnight at 42°C (Table 2). Membranes were washed four times at 42°C (2X SSC, 0.1 % SDS), wrapped in plastic, and exposed to a phosphor-imager screen before imaging on a Typhoon 9400 (GE Healthcare). Membranes were stripped with two 30 minute washes in 2% SDS at 80^QC, prior to reblocking and incubation with labeled probe.

Nanopore library generation for budding yeast tRNAs

[00115] Gel-purified tRNAs from chemical ligation were enzymatically ligated to capture deacylated tRNAs with 3' splint adapters and attach 5' adapters to all tRNA. tRNA from the first ligation (20 pmol) was incubated in a 20 pL reaction consisting of 10% PEG 8000, 1 pL of RNase inhibitor (Watchmaker Genomics), 9 pmol gel-purified uncharged 3' splint adapter, 9 pmol gel-purified 5' adapter, 1 X T4 RNA ligase 2 buffer, and 2 pL of T4 RNA ligase 2 (homemade preparation, 0.74 mg/mL). This ligation was incubated at 25^QC for 30 minutes.

[00116] Ligation products were purified by addition of a 1 .8X volume of tRNA beads (BioDynami), mixing by pipetting, and incubation on ice for 4 minutes, followed by magnetic separation. The supernatant was discarded. Beads were washed with 180 pL 80% EtOH and air dried. Beads were resuspended in 13 pL of water, and the elution was moved to a new tube and quantified.

[00117] Splint-adapter-ligated tRNAs are next ligated to RT adapters (RTA) (provided in the RNA004 ONT kit): 12.5 pL of sample was incubated with 1 .5 pL RTA, 0.5 pL RNase inhibitor (Watchmaker Genomics), 4 pL T4 DNA ligase buffer, and 1 .5 pL T4 DNA ligase (Watchmaker Genomics) for 30 minutes at 25^QC, and cleaned up at RT using the tRNA beads as above, using a 1 .35X volume of beads, and elution in 26 pL water. Each sample was quantified with the Quant-iT Qubit dsDNA HS kit and the library size distribution was confirmed by Agilent TapeStation (HS DNA 1000).

[00118] Finally, ligation products were ligated to ONT’s RNA ligation adapter (RLA) on the same day that sequencing was conducted. 50-400 fmol of sample in 23 pL was incubated with 6 pL RLA (ONT RNA004), 8 pL T4 DNA ligase buffer, and 3 pL T4 DNA ligase (Watchmaker Genomics) for 30 minutes at 25^QC. These final ligations were cleaned up using a 1 .8X volume of Ampure XP SPRI beads (Beckman Coulter), and washed with WSB wash buffer (ONT) following the protocol for ONT SQK-RNA004.

Sequencing run conditions for budding yeast tRNAs

[00119] Libraries were loaded onto “RNA” flow cells on a Prometh ION P2 Solo instrument connected to a A5000 GPU workstation or a Prometh ION P2integrated instrument, using MinKNOW software version 24.06.10. The throughput of aa-tRNA-seq was found to be comparable or superior to previous nanopore tRNA sequencing approaches ^32,83, collecting a median >8 million reads for biological tRNA sequencing libraries, and median -250 thousand reads for synthetic tRNA sequencing libraries.

Base-calling and alignment

[00120] Libraries were basecalled with Dorado v0.7.2 (Oxford Nanopore Technologies, httDs://github.com/nanoDoretech/dorado) using the “super high accuracy” (rna004_130bps_sup) v5.0.0 model and -emit-moves parameter. Basecalled bams were converted to fastq format using samtools v. 1 .21 ⁸⁴ with -T flag to retain move tables, and then aligned to using BWA-MEM version 0.7.16-r1 181 ⁸⁵ with the parameters bwa mem -c -w 13 -k 6 -x ont2d, enabling transfer of move tables to aligned bams.

[00121] To evaluate the performance of an alignment-based classification approach for identifying charged vs. uncharged tRNA reads, aligned bams were further filtered to contain reads mapped to full length tRNAs. tRNA reference files were constructed by appending CCA sequences to each mature tRNA sequence, along with the unique 3' and universal 5' adapter sequences, and primary alignments for each read assessed. A Snakemake ⁸⁶ analysis pipeline is available at https://qithub.com/rnabioco/aa-tRNA-seq- pipeline.

Remora model training and validation

[00122] To train a machine learning model distinguishing charged and uncharged tRNAs, the Remora software (v3.2, https://qithub.com/nanoporetech/remora) and training procedure for modified nucleosides was used. First, fully charged yeast tRNA libraries (treated as modified base in the training) and deacetylated libraries (treated as modified base control) were prepared. To make the model universal for all tRNAs, a 6-nt modification kmer was defined spanning the universal CCA 3' end of tRNA and the first three nucleotides of the 3' adapter (CCAGGC), where the underlined G was defined as the modification site. Chunks from both libraries are extracted using remora dat aset prepare with the default remora 9mer table for rescaling and following parameters: — re f ine-rough-rescale -reverse- s ignal -mot i f CCAGGC 3. Training configuration files were prepared with remora dataset make_conf ig us ing - dat aset -weight s 1 1. Finally, the model was trained using the parameters: — model ConvLS TM_w_re f . py -chunk-context 2 0 0 20 0 --num-test -chunks 2 0 0 00. After internal remora validation on 20000 chunks, this model was validated on independent libraries using reference-anchored remora inference (remora infer f rom_pod5_and_bam -reference-anchored). ML tags from inference Output bams were used to calculate model performance, with the following assumption: ML < 200 = uncharged tRNA, ML > 200 = charged tRNA.

[00123] Pairwise machine learning classifiers to distinguish individual amino acids (binary recognition between pairs of amino acids totalling 380 models) were trained using Remora v. 3.2 using a procedure similar to that described above, using Flexizyme-charged synthetic tRNA reads for the training. Synthetic tRNA was aligned to a reference sequence with one nucleotide (“T”) inserted between the CCA sequence at the 3' terminus and the start of the 3' adapter sequence. Pairwise models were trained with one amino acid treated as modified base and the second one as modified base control on the CCAT motif, with the inserted T identified as the modification position for Remora training. 10000 chunks were used for internal Remora validation, except for pairwise comparisons with alanine, where 9000 chunks were used due to lower library depth for the alanine-charged library. F1 scores were calculated for all models using the equation: 2 TP / ( 2 TP+FP+FN) .

Nanopore signal analysis

[00124] Signal metrics (dwell time and trimmed means of ionic current) at reference- anchored positions were extracted from POD5 files using the Remora API and stored in TSV files for analysis and plotting in R. Example scripts for signal extraction and plotting will be available at https://github.com/rnabioco/aa-tRNA-seq.

TABLES

Table 1. Yeast strains

Table 2. Oligonucleotide sequences

REFERENCES

1 . Crick, F. H. The origin of the genetic code. J Mol Biol 38, 367-379 (1968).

2. Phizicky, E. M. & Hopper, A. K. The life and times of a tRNA. RNA 29, 898-957 (2023).

3. Yared, M.-J., Marcelot, A. & Barraud, P. Beyond the Anticodon: tRNA Core Modifications and Their Impact on Structure, Translation and Stress Adaptation. Genes (Basel) 15, (2024).

4. Perret, V. et al. Relaxation of a transfer RNA specificity by removal of modified nucleotides. Nature 344, 787-789 (1990).

5. Giege, R. & Eriani, G. The tRNA identity landscape for aminoacylation and beyond. Nucleic Acids Res 5 , 1528-1570 (2023).

6. Agris, P. F., Vendeix, F. A. P. & Graham, W. D. tRNA’s wobble decoding of the genome: 40 years of modification. J Mol Biol 366, 1-13 (2007).

7. Avcilar-Kucukgoze, I. et al. Discharging tRNAs: a tug of war between translation and detoxification in Escherichia coli. Nucleic Acids Res 44, 8324-8334 (2016).

8. Subramaniam, A. R., Pan, T. & Cluzel, P. Environmental perturbations lift the degeneracy of the genetic code to regulate protein levels in bacteria. Proc Natl Acad Sci US A G, 2419-2424 (2013). Dever, T. E. et a/.ePhosphorylation of initiation factor 2 alpha by protein kinase GCN2 mediates gene-specific translational control of GCN4 in yeast. Cell 68, 585-596 (1992). Sorensen, M. A. Charging levels of four tRNA species in Escherichia coli Rel(-i-) and Rel(-) strains during amino acid starvation: a simple model for the effect of ppGpp on translational accuracy. J Mol Biol 307, 785-798 (2001 ). Zaborske, J. M. et al. Genome-wide analysis of tRNA charging and activation of the elF2 kinase Gcn2p. J Biol Chem 284, 25254-25267 (2009). Passarelli, M. C. et al. Leucyl-tRNA synthetase is a tumour suppressor in breast cancer and regulates codon-dependent translation dynamics. Nat Cell Biol 24, 307-315 (2022). Hyeon, D. Y. et al. Evolution of the multi-tRNA synthetase complex and its role in cancer. J Biol Chem 294, 5340-5351 (2019). Netzer, N. etal. Innate immune and chemically triggered oxidative stress modifies translational fidelity. Nature 462, 522-526 (2009). Sun, L. et al. Evolutionary Gain of Alanine Mischarging to Noncognate tRNAs with a G4:U69 Base Pair. J Am Chem Soc 38, 12948-12955 (2016). Ling, J., Yadavalli, S. S. & Ibba, M. Phenylalanyl-tRNA synthetase editing defects result in efficient mistranslation of phenylalanine codons as tyrosine. RNA 13, 1881-1886 (2007). Shepherd, J. & Ibba, M. Relaxed substrate specificity leads to extensive tRNA mischarging by Streptococcus pneumoniae class I and class II aminoacyl-tRNA synthetases. mBio 5, e01656-14 (2014). Kelly, P. et al. Alanyl-tRNA Synthetase Quality Control Prevents Global Dysregulation of the Escherichia coli Proteome. mBio 10, (2019). Evans, M. E., Clark, W. C., Zheng, G. & Pan, T. Determination of tRNA aminoacylation levels by high-throughput sequencing. Nucleic Acids Res 45, e133 (2017). Watkins, C. P., Zhang, W., Wylder, A. C., Katanski, C. D. & Pan, T. A multiplex platform for small RNA sequencing elucidates multifaceted tRNA stress response and translational regulation. Nat Common 13, 2491 (2022). Davidsen, K. & Sullivan, L. B. A robust method for measuring aminoacylation through tRNA-Seq. Elite 12, (2024). Hernandez-Alias, X. et al. Single-read tRNA-seq analysis reveals coordination of tRNA modification and aminoacylation and fragmentation. Nucleic Acids Res 5 , e17 (2023). Carlile, T. M. et al. Pseudouridine profiling reveals regulated mRNA pseudouridylation in yeast and human cells. Nature 515, 143-146 (2014). Varshney, U., Lee, C. P. & RajBhandary, U. L. Direct analysis of aminoacylation levels of tRNAs in vivo. Application to studying recognition of Escherichia coli initiator tRNA mutants by glutaminyl-tRNA synthetase. J Biol Chem 266, 24712- 24718 (1991 ). Grosjean, H., Droogmans, L., Roovers, M. & Keith, G. Detection of enzymatic activity of transfer RNA modification enzymes using radiolabeled tRNA substrates. Methods Enzymol 425, 55-101 (2007). Henley, R. Y. et al. Electrophoretic Deformation of Individual Transfer RNA Molecules Reveals Their Identity. Nano Lett 16, 138-144 (2016). Smith, A. M., Abu-Shumays, R., Akeson, M. & Bernick, D. L. Capture, Unfolding, and Detection of Individual tRNA Molecules Using a Nanopore Device. Front Bioeng Biotechnol3, 91 (2015). Wang, Y. et al. Structural-profiling of low molecular weight RNAs by nanopore trapping/translocation using Mycobacterium smegmatis porin A. Nat Common 12, 3368 (2021 ). Thomas, N. K. et al. Direct Nanopore Sequencing of Individual Full Length tRNA Strands. ACS Nano 15, 16642-16653 (2021 ). Garalde, D. R. et al. Highly parallel direct RNA sequencing on an array of nanopores. Nat Methods 15, 201-206 (2018). White, L. K., Strugar, S. M., MacFadden, A. & Hesselberth, J. R. Nanopore sequencing of internal 2’-PO modifications installed by RNA repair. RNA 29, 847-861 (2023). Lucas, M. C. et al. Quantitative analysis of tRNA abundance and modifications by nanopore RNA sequencing. Nat Biotechnol 42, 72-86 (2024). Radakovic, A., Wright, T. H., Lelyveld, V. S. & Szostak, J. W. A Potential Role for Aminoacylation in Primordial RNA Copying Chemistry. Biochemistry 60, 477-488 (2021 ). Radakovic, A., DasGupta, S., Wright, T. H., Aitken, H. R. M. & Szostak, J. W. Nonenzymatic assembly of active chimeric ribozymes from aminoacylated RNA oligonucleotides. Proc Natl Acad Sci U S A 119, (2022). Murakami, H., Ohta, A., Ashigai, H. & Suga, H. A highly flexible tRNA acylation method for non-natural polypeptide synthesis. Nat Methods 3, 357-359 (2006). Mansy, S. S. et al. Template-directed synthesis of a genetic polymer in a model protocell. Nature 454, 122-125 (2008). Radakovic, A. et al. A potential role for RNA aminoacylation prior to its role in peptide synthesis. Proc Natl Acad Sci U S A 121 , e2410206121 (2024). Liu, Z., Ajram, G., Rossi, J.-C. & Pascal, R. The Chemical Likelihood of Ribonucleotide-a-Amino acid Copolymers as Players for Early Stages of Evolution. J Mol Evol 87, 83-92 (2019). Peacock, J. R. et al. Amino acid-dependent stability of the acyl linkage in aminoacyl-tRNA. RNA 20, 758-764 (2014). Rand, A. C. et al. Mapping DNA methylation with high-throughput nanopore sequencing. Nat Methods 14, 41 1-413 (2017). Furlan, M. et al. Computational methods for RNA modification detection from nanopore direct RNA sequencing data. RNA Biol 18, 31-40 (2021 ). White, L. K. & Hesselberth, J. R. Modification mapping by nanopore sequencing. Front Genet 13, 1037134 (2022). Begik, O. et al. Quantitative profiling of pseudouridylation dynamics in native RNAs with nanopore sequencing. Nat Biotechnol 39, 1278-1291 (2021 ). Leger, A. et al. RNA modifications detection by comparative Nanopore direct RNA sequencing. Nat Common 12, 7198 (2021 ). Alexandrov, A. et al. Rapid tRNA decay can result from lack of nonessential modifications. Mol Cell 21 , 87-96 (2006). Tasak, M. & Phizicky, E. M. Initiator tRNA lacking 1 -methyladenosine is targeted by the rapid tRNA decay pathway in evolutionarily distant yeast species. PLoS Genet 18, e1010215 (2022). De Zoysa, T. & Phizicky, E. M. Hypomodified tRNA in evolutionarily distant yeasts can trigger rapid tRNA decay to activate the general amino acid control response, but with different consequences. PLoS Genet 16, e1008893 (2020). Dewe, J. M., Whipple, J. M., Chernyakov, L, Jaramillo, L. N. & Phizicky, E. M. The yeast rapid tRNA decay pathway competes with elongation factor 1 A for substrate tRNAs and acts on tRNAs lacking one or more of several modifications. R/V 18, 1886-1896 (2012). Guy, M. P. etal. Identification of the determinants of tRNA function and susceptibility to rapid tRNA decay by high-throughput in vivo analysis. Genes Dev 28, 1721-1732 (2014). Payea, M. J. et al. Widespread temperature sensitivity and tRNA decay due to mutations in a yeast tRNA. RNA 24, 410-422 (2018). De Zoysa, T. et al. A connection between the ribosome and two tRNA modification mutants subject to rapid tRNA decay. bioRxiv (2023) doi :10.1 101/2023.09.18.558340. Payea, M. J., Hauke, A. C., De Zoysa, T. & Phizicky, E. M. Mutations in the anticodon stem of tRNA cause accumulation and Met22-dependent decay of pre- tRNA in yeast. RNA 26, 29-43 (2020). Han, L, Guy, M. P., Kon, Y. & Phizicky, E. M. Lack of 2’-O-methylation in the tRNA anticodon loop of two phylogenetically distant yeast species activates the general amino acid control pathway. PLoS Genet 14, e1007288 (2018). Chernyakov, L, Whipple, J. M., Kotelawala, L., Grayhack, E. J. & Phizicky, E. M. Degradation of several hypomodified mature tRNA species in Saccharomyces cerevisiae is mediated by Met22 and the 5’-3' exonucleases Rati and Xrn1 . Genes Dev 22, 1369-1380 (2008). Dichtl, B., Stevens, A. & Tollervey, D. Lithium toxicity in yeast is due to the inhibition of RNA processing enzymes. EMBO J 16, 7184-7195 (1997). Cappannini, A. et al. MODOMICS: a database of RNA modifications and related information. 2023 update. Nucleic Acids Res 52, D239-D244 (2024). Kotelawala, L., Grayhack, E. J. & Phizicky, E. M. Identification of yeast tRNA Um(44) 2’-O-methyltransferase (Trm44) and demonstration of a Trm44 role in sustaining levels of specific tRNA(Ser) species. RNA 14, 158-169 (2008). Stoiber, M. et al. De novo Identification of DNA Modifications Enabled by Genome-Guided Nanopore Signal Processing. bioRxiv 094672 (2017) doi:10.1 101/094672. Diensthuber, G. et al. Enhanced detection of RNA modifications and read mapping with high-accuracy nanopore RNA basecalling models. Genome Res (2024) doi: 10.1101 /gr.278849.123. Lueck, J. D. etal. Engineered transfer RNAs for suppression of premature termination codons. Nat Common 10, 822 (2019). Albers, S. et al. Engineered tRNAs suppress nonsense mutations in cells and in vivo. Nature 6 8, 842-848 (2023). Anastassiadis, T. & Kohrer, C. Ushering in the era of tRNA medicines. J Biol Chem 299, 105246 (2023). Uhlenbeck, O. C. & Schrader, J. M. Evolutionary tuning impacts the design of bacterial tRNAs for the incorporation of unnatural amino acids by ribosomes. Curr Opin Chem Biol 46, 138-145 (2018). Katoh, T. & Suga, H. In Vitro Genetic Code Reprogramming for the Expansion of Usable Noncanonical Amino Acids. Annu Rev Biochem 91 , 221-243 (2022). Dunkelmann, D. L. etal. Adding a,a-disubstituted and [3-linked monomers to the genetic code of an organism. Nature 625, 603-610 (2024). Soni, C. et al. A Translation-Independent Directed Evolution Strategy to Engineer Aminoacyl-tRNA Synthetases. ACS Central Science (2024) doi :10.1021/acscentsci.3c01557. Zhou, L., O’Flaherty, D. K. & Szostak, J. W. Template-Directed Copying of RNA by Non-enzymatic Ligation. Angewandte Chemie International Edition 59, 15682-15687 (2020). Wiltrout, E., Goodenbour, J. M., Frechin, M. & Pan, T. Misacylation of tRNA with methionine in Saccharomyces cerevisiae. Nucleic Acids Res 40, 10494-10506 (2012). Shigematsu, M. et al. YAMAT-seq: an efficient method for high-throughput sequencing of mature transfer RNAs. Nucleic Acids Res 45, e70 (2017). Warren, J. M. etal. Combining tRNA sequencing methods to characterize plant tRNA expression and post-transcriptional modification. RNA Biol 18, 64-78 (2021 ). Ceriotti, L. F., Warren, J. M., Virginia Sanchez-Puerta, M. & Sloan, D. B. The landscape of Arabidopsis tRNA aminoacylation. bioRxiv 2024.09.09.612099 (2024) doi: 10.1101/2024.09.09.612099. Cozen, A. E. et al. ARM-seq: AlkB-facilitated RNA methylation sequencing reveals a complex landscape of modified tRNA fragments. Nat Methods 12, 879- 884 (2015). Zheng, G. et al. Efficient and quantitative high-throughput tRNA sequencing. Nat Methods 12, 835-837 (2015). Scheepbouwer, C. et al. ALL-tRNAseq enables robust tRNA profiling in tissue samples. Genes Dev 37, 243-257 (2023). Fleming, A. M., Mathewson, N. J., Howpay Manage, S. A. & Burrows, C. J. Nanopore Dwell Time Analysis Permits Sequencing and Conformational Assignment of Pseudouridine in SARS-CoV-2. ACS Cent Sci7, 1707-1717 (2021 ). Stephenson, W. et al. Direct detection of RNA modifications and structure using single-molecule nanopore sequencing. Cell Genom 2, (2022). Shah, N. H. & Muir, T. W. Inteins: Nature’s Gift to Protein Chemists. Chem Sci5, 446-461 (2014). Schultz, S. K. et al. Modifications in the T arm of tRNA globally determine tRNA maturation, function, and cellular fitness. Proc Natl Acad Sci U S A 121 , e2401 154121 (2024). Han, L. & Phizicky, E. M. A rationale for tRNA modification circuits in the anticodon loop. RNA 24, 1277-1284 (2018). Jones, J. D., Simcox, K. M., Kennedy, R. T. & Koutmou, K. S. Direct sequencing of total tRNAs by LC-MS/MS. RNA 29, 1201-1214 (2023). 81 . Shaw, E. A. et al. Combining Nanopore direct RNA sequencing with genetics and mass spectrometry for analysis of T-loop base modifications across 42 yeast tRNA isoacceptors. Nucleic Acids Res 52, 12074-12092 (2024).

82. National Academies of Sciences, Engineering & Medicine. Toward Sequencing and Mapping of RNA Modifications: Proceedings of a Workshop-in Brief. (2023) doi:10.17226/27149.

83. White, L. K. et al. Comparative analysis of 43 distinct RNA modifications by nanopore tRNA sequencing. bioRxiv (2024) doi:10.1 101/2024.07.23.604651.

84. Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078-2079 (2009).

85. Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. (2013).

86. Molder, F. etal. Sustainable data analysis with Snakemake. FWOORes 10, 33 (2021 ).

87. Monera, O. D., Sereda, T. J., Zhou, N. E., Kay, C. M. & Hodges, R. S. Relationship of sidechain hydrophobicity and alpha-helical propensity on the stability of the single-stranded amphipathic alpha-helix.

[00125] The foregoing is illustrative of embodiments of the inventive concept and is not to be construed as limiting thereof. Although some embodiments have been described, those skilled in the art will readily appreciate that many modifications are possible in embodiments without materially departing from the novel teachings and advantages of the inventive concept. Accordingly, all such modifications are intended to be included within the scope of the inventive concept as defined in the claims that follow.

Claims

THAT WHICH IS CLAIMED:

1 . A method of characterizing a polynucleotide comprising:

(i) providing together: a) a construct comprising a polynucleotide, wherein the polynucleotide comprises at least one oligonucleotide adapter attached to one end of the polynucleotide, and b) a nanopore, wherein the construct and the nanopore are combined under conditions in which the construct associates with the nanopore;

(ii) subjecting the construct and the nanopore to a condition that permits the polynucleotide to enter the nanopore and at least partially translocate through the nanopore;

(iii) measuring a property associated with translocation of the polynucleotide through the nanopore; and

(iv) characterizing the polynucleotide by analyzing the property measured as the polynucleotide translocates through the nanopore, wherein analyzing the property measured provides information regarding a characteristic of the polynucleotide.

2. The method of claim 1 , wherein the construct comprises a first oligonucleotide adapter attached to the 3' end of the polynucleotide, and a second oligonucleotide adapter attached to the 5' end of the polynucleotide.

3. The method of claim 2, wherein the first oligonucleotide adapter can associate and/or hybridize with the second oligonucleotide adapter

4. The method of any one of claims 1-3, wherein characterizing the polynucleotide comprises determining a nucleotide sequence of the polynucleotide.

5. The method of any one of claims 1-4, wherein the polynucleotide comprises ribonucleic acid (RNA).

6. The method of claim 5, wherein the RNA is a transfer RNA (tRNA), a tRNA fragment, or an RNA comprising a tRNA-like structure.

7. The method of any one of claims 1-6, wherein characterizing the polynucleotide comprises determining whether the polynucleotide comprises an aminoacylation and/or is charged with an amino acid or a peptide.

8. The method of claim 7, wherein characterizing the polynucleotide comprises characterizing and/or identifying the amino acid or peptide attached to the polynucleotide if the polynucleotide is charged and/or aminoacylated.

9. The method of claim 7 or 8, wherein generating the construct comprises reacting an oligonucleotide adapter comprising an activated phosphate/phosphoramidating agent with the amino acid/peptide of a charged and/or aminoacylated polynucleotide to generate a phosphoramidate linkage between the oligonucleotide adapter and the amino acid/peptide of the charged and/or aminoacylated polynucleotide.

10. The method of claim 9, wherein the oligonucleotide adapter is a 5'- phosphorimidazolated oligonucleotide that reacts with the amino group of the amino acid/peptide on the charged/aminoacylated polynucleotide at the 3' end of the polynucleotide.

1 1 . The method of claim 9 or 10, wherein the oligonucleotide adapter does not and/or cannot attach the polynucleotide if the polynucleotide is not charged/aminoacylated.

12. The method of any one of claims 1-11 , wherein the condition is a potential difference across the nanopore.

13. The method of any one of claims 1-12, wherein the property measured is current flowing through the nanopore and/or dwell time/translocation speed of the polynucleotide/nucleic acid passing through the nanopore.

14. The method of any one of claims 1-13, wherein characterizing the polynucleotide comprises applying machine learning to characterize the polynucleotide.

15. The method of claim 14, wherein the machine learning is applied to analyze the property measured to characterize the polynucleotide.

16. The method of claim 14 or 15, wherein the machine learning is applied to analyze the property measured to characterize if the polynucleotide is charged/aminoacylated, and/or to characterize/identify the amino acid/peptide attached to the polynucleotide if the polynucleotide is charged and/or aminoacylated.

17. A method of determining whether a polynucleotide is charged/aminoacylated comprising:

(i) providing together: a) a construct comprising the polynucleotide, wherein the polynucleotide comprises at least one oligonucleotide adapter attached to one end of the polynucleotide, and wherein the oligonucleotide adapter is attached to an amino acid/peptide if the polynucleotide is charged/aminoacylated through a phosphoramidate linkage generated by reacting the amino acid/peptide with an oligonucleotide adapter comprising an activated phosphate/phosphoramidating agent, and b) a nanopore, wherein the construct and the nanopore are combined under conditions in which the construct associates with the nanopore;

(ii) subjecting the construct and nanopore to a condition that permits the polynucleotide to enter the nanopore and at least partially translocate through the nanopore; (iii) measuring a property associated with translocation of the polynucleotide through the nanopore; and

(iv) characterizing whether the polynucleotide is ch arg ed/ami noacylated by analyzing the property measured as the polynucleotide translocates through the nanopore, wherein analyzing the property measured provides information regarding charging/aminoacylation status of the polynucleotide.

18. The method of claim 17, wherein the oligonucleotide adapter comprises a 5'-phosphorimidazole that attaches to the amino acid/peptide on the charged/aminoacylated polynucleotide at the 3' end of the polynucleotide.

19. The method of claim 17 or 18, wherein the oligonucleotide adapter does not and/or cannot attach the polynucleotide if the polynucleotide is not charged/aminoacylated.

20. The method of any one of claims 17-19, wherein the construct further comprises a second oligonucleotide adapter attached at the 5' end of the polynucleotide.

21 . The method of claim 20, wherein the first oligonucleotide adapter can associate and/or hybridize with the second oligonucleotide adapter.

22. The method of any one of claims 17-21 , wherein characterizing the polynucleotide comprises determining a nucleotide sequence of the polynucleotide.

23. The method of any one of claims 17-22, wherein the polynucleotide is an RNA.

24. The method of claim 23, wherein the RNA is a tRNA.

25. The method of any one of claims 17-24, wherein characterizing the polynucleotide comprises characterizing and/or identifying the amino acid or peptide attached to the polynucleotide if the polynucleotide is charged and/or aminoacylated.

26. The method of any one of claims 17-25, wherein the condition is a potential difference across the nanopore.

27. The method of any one of claims 17-25, wherein the property measured is current flowing through and/or dwell time/translocation speed of the polynucleotide/nucleic acid passing through the nanopore.

28. The method of any one of claims 17-27, wherein characterizing the polynucleotide comprises applying machine learning to characterize the polynucleotide.

29. The method of claim 28, wherein the machine learning is applied to analyze the property measured to characterize the polynucleotide.

30. The method of claim 28 or 29, wherein the machine learning is applied to analyze the property measured to characterize the amino acid/peptide attached to the polynucleotide if the polynucleotide is charged/aminoacylated.

31 . A method for characterizing charging/aminoacylation levels of tRNAs comprising:

(i) generating a library of tRNA constructs, the tRNA constructs comprising at least one oligonucleotide adapter attached to one end of each tRNA, wherein charged/aminoacylated tRNA constructs comprise an RNA oligonucleotide adapter attached to an amino acid/peptide on the tRNA charged/aminoacylated with an amino acid/peptide through a phosphoramidate linkage generated by reacting the amino acid/peptide with an oligonucleotide adapter comprising an activated phosphate/phosphoramidating agent; (ii) providing the tRNA constructs together with nanopores, wherein the construct and the nanopore are combined under conditions in which a single tRNA construct associates with a single nanopore;

(iii) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore;

(iv) measuring a property associated with translocation of the tRNA constructs through the nanopores; and

(v) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein analyzing the property measured provides information regarding the charging/aminoacylation levels of the tRNAs.

32. The method of claim 31 , wherein the oligonucleotide adapter comprises a 5'-phosphorimidazole that reacts with the amino acid/peptide on the charged/aminoacylated tRNA at the 3' end of the tRNA to generate the phosphoramidate linkage.

33. The method of claim 31 or 32, wherein the oligonucleotide adapter does not and/or cannot attach the tRNA if the tRNA is not charged/aminoacylated.

34. The method of any one of claims 31-33, wherein the construct further comprises a second oligonucleotide adapter attached at the 5' end of the tRNA.

35. The method of claim 34, wherein the first oligonucleotide adapter can associate and/or hybridize with the second oligonucleotide adapter.

36. The method of any one of claims 31-35, wherein characterizing the polynucleotide comprises determining a nucleotide sequence of the tRNA.

37. The method of any one of claims 31-36, wherein characterizing the polynucleotide comprises characterizing and/or identifying the amino acid or peptide attached to the tRNA if the tRNA is charged and/or aminoacylated.

38. The method of any one of claims 31-37, wherein the condition is a potential difference across the nanopore.

39. The method of any one of claims 31-38, wherein the property measured is current flowing through and/or dwell time/translocation speed of the polynucleotide/nucleic acid passing through the nanopore.

40. The method of any one of claims 31-39, wherein characterizing charging/aminoacylation levels of the tRNA comprises applying machine learning to characterize charging/aminoacylation levels of the tRNA.

41 . The method of claim 40, wherein the machine learning is applied to analyze the property change to characterize charging/aminoacylation levels of the tRNA.

42. The method of claim 40 or 41 , wherein the machine learning is applied to analyze the property change to characterize the amino acid/peptide on the tRNA if the tRNA is charged and/or aminoacylated.

43. A method of characterizing changes in charging/aminoacylation of tRNAs comprising:

(i) isolating a pool of tRNAs from a subject/organism growing under/subjected to an environmental condition;

(ii) generating a library of tRNA constructs, the tRNA constructs each comprising at least one oligonucleotide adapter attached to one end of each tRNA, wherein the oligonucleotide adapter is attached to an amino acid/peptide if the tRNA is ch arg ed/ami noacylated through a phosphoramidate linkage generated by reacting the amino acid/peptide with an imidazolated oligonucleotide adapter;

(iii) providing the tRNA constructs together with nanopores, wherein the constructs and the nanopores are combined under conditions in which a single tRNA construct associates with a single nanopore;

(iv) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore;

(v) measuring a property associated with translocation of the tRNA constructs through the nanopores; and

(vi) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein the charging/aminoacylation levels characterized for the tRNA constructs are compared to charging/aminoacylation levels for a pool of tRNAs derived from a subject/organism growing under/subjected to a control environmental condition.

44. The method of claim 43, wherein generating the tRNA constructs comprises reacting a 5'-phophoimidazolated oligonucleotide adapter with the amino acid/peptide on charged/aminoacylated tRNAs at the 3' end of the tRNAs.

45. The method of claim 43 or 44, wherein the imidazolated oligonucleotide adapter does not or cannot attach the tRNA if the tRNA is not charged/aminoacylated.

46. The method of any one of claims 43-45, wherein the construct further comprises a second oligonucleotide adapter attached at the 5' end of the tRNA.

47. The method of claim 46, wherein the first oligonucleotide adapter can associate and/or hybridize with the second oligonucleotide adapter.

48. The method of any one of claims 43-47, wherein the condition is a potential difference across the nanopore.

49. The method of any one of claims 43-48, wherein the property change measured is current flowing through the nanopore.

50. The method of any one of claims 43-49, wherein characterizing charging/aminoacylation levels of the tRNA comprises applying machine learning to characterize the tRNA.

51 . The method of claim 50, wherein the machine learning is applied to analyze the property change to characterize charging/aminoacylation levels of the tRNA.

52. The method of any one of claims 43-51 , wherein the environmental condition is nutrient stress/limitation/depletion.

53. The method of claim 52, wherein the nutrient stress/limitation/depletion comprises essential amino acid starvation.

54. The method of any one of claims 43-53, wherein characterizing charging/aminoacylation levels comprises characterizing mischarging and/or mis- aminoacylation of tRNA in the subject/organism grown under/subjected to the environmental condition.

55. A method of characterizing charging/aminoacylation of hypomodified tRNA comprising:

(i) isolating a pool of hypomodified tRNAs from a subject/organism;

(ii) generating a library of tRNA constructs, the tRNA constructs comprising at least one oligonucleotide adapter attached to one end of each tRNA, wherein tRNA constructs charged/aminoacylated with an amino acid/peptide comprise an RNA oligonucleotide adapter attached to the amino acid/peptide on the tRNA charged/aminoacylated with an amino acid/peptide through a phosphoramidate linkage generated by reacting the amino acid/peptide with an imidazolated oligonucleotide adapter;

(iii) providing the tRNA constructs together with nanopores, wherein the constructs and the nanopores are combined under conditions in which a single tRNA construct associates with a single nanopore;

(iv) subjecting the tRNA constructs and nanopores to a condition that permits a single tRNA construct to enter a single nanopore and at least partially translocate through the single nanopore;

(v) measuring a property associated with translocation of the tRNA constructs through the nanopores; and

(vi) characterizing charging/aminoacylation levels by analyzing the property measured as the tRNA constructs translocate through the nanopores, wherein the charging/aminoacylation levels characterized for the hypomodified tRNA constructs are compared to charging/aminoacylation levels for a pool of normally modified tRNAs.

56. The method of claim 55, wherein the imidazolated oligonucleotide adapter comprises a 5'-phosphorimidazole that attaches to the amino acid/peptide on the charged/aminoacylated tRNA at the 3' end of the tRNA.

57. The method of claim 55 or 56, wherein the imidazolated oligonucleotide adapter does not or cannot attach the tRNA if the tRNA is not charged/aminoacylated.

58. The method of any one of claims 55-57, wherein the construct further comprises a second oligonucleotide adapter attached at the 5' end of the tRNA.

59. The method of claim 58, wherein the first oligonucleotide adapter can associate and/or hybridize with the second oligonucleotide adapter.

60. The method of any one of claims 55-59, wherein the condition is a potential difference across the nanopore.

61 . The method of any one of claims 55-60, wherein the property change measured is current flowing through the nanopore.

62. The method of any one of claims 55-61 , wherein characterizing charging/aminoacylation levels of the tRNA comprises applying machine learning to characterize the tRNA.

63. The method of claim 62, wherein the machine learning is applied to analyze the property change to characterize charging/aminoacylation levels of the tRNA.

64. The method of claim 62 or 63, wherein the machine learning is applied to analyze the property change to characterize the amino acid/peptide attached to the tRNA if the tRNA is charged and/or aminoacylated.

65. A classifier trained to characterize charging/aminoacylation status of a polynucleotide/nucleic acid and/or charging/aminoacylation levels of a pool of polynucleotides/nucleic acids.

66. The classifier of claim 65, wherein the polynucleotide/nucleic acid is a tRNA, a tRNA fragment, or an RNA comprising a tRNA-like structure.

67. The classifier of claim 65 or 66, wherein characterizing charging/aminoacylation of the polynucleotide/nucleic acid comprises: determining if the polynucleotide/nucleic acid is charged/aminoacylated; characterizing/identifying the amino acid/peptide attached if the polynucleotide/nucleic acid is charged/aminoacylated; and/or characterizing an extent to which a pool of polynucleotides/nucleic acids is charged/aminoacylated.

68. The classifier of any one of claims 65-67, wherein training the classifier comprises inputting a ground truth dataset comprising measurements of a property suitable for characterizing the charging/aminoacylation of the polynucleotide/nucleic acid.

69. The classifier of claim 68, wherein the ground truth dataset comprises measurements of properties of fully charged/aminoacylated tRNAs and measurements of properties of uncharged tRNAs.