JP2002526030A - Identification of molecular interaction sites in RNA for discovery of new drugs - Google Patents

Abstract

  (57) [Summary] Methods for identifying sites of molecular interaction in eukaryotic and prokaryotic nucleic acids, particularly RNA, are described. Secondary structural elements are identified from the highly conserved sequence. A method for preparing a database relating to such molecular interaction sites is also provided herein as the database itself. Therapeutic, agricultural, industrial and other uses result from the interaction of these molecular interaction sites with "small" and other molecules.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001] This application is part of a sequel to US Serial No. 09 / 076,440 filed May 12, 1998, which is a provisional application filed May 12, 1998 US Serial No. 60 / 085,09
2 claim priority, each of which is incorporated herein by reference in its entirety.

FIELD OF THE INVENTION [0002] The present invention shows methods for identifying regions of prokaryotic and eukaryotic nucleic acids, particularly RNA, that can serve as molecular interaction sites. The invention also includes therapeutics and structural databases.

BACKGROUND OF THE INVENTION Recent advances in genomic, molecular and structural biology have shown how RNA molecules participate or regulate many of the phenomena required for protein expression in cells. Is in the limelight. RNA molecules are more DNA than simple vectors
Active in controlling its own transcription, splicing and editing of mRNA and tRNA molecules, synthesizing peptide bonds at ribosomes, catalyzing the transfer of developmental proteins to the cell membrane, and providing sufficient regulation of the rate of information translation To do. RNA molecules can adopt a variety of unique structural motifs and provide the necessary framework for performing their functions.

[0004] "Small" molecule therapy is an organic chemical molecule that specifically binds structural RNA molecules, but is not a polymer. "Small" molecule therapy involves the most potent naturally occurring antibodies. For example, aminoglycoside and macrolide antibodies are "small" molecules that bind to specific regions of the ribosomal RNA (rRNA) structure and are believed to work by inhibiting structural changes in the RNA required for protein synthesis. Changes in the structure of RNA molecules have been shown to control the rate of transcription and translation of mRNA molecules.

[0005] An additional opportunity in target RNA for drug discovery is that cells frequently produce different mRNAs in different tissues that can be translated into the same protein. Processes such as alternative splicing and alternative polyadenylation create unique or abundant transcripts in specific tissues. This provides the opportunity to design drugs that bind to unique RNA regions in desired cells, including tumors, and that do not affect, or to a lesser extent, protein expression in other cells, Provides an additional level of drug specificity that is not generally achieved with treatments targeting.

[0006] Applicants believe that RNA molecules or related RNA molecules have regulatory regions that cells use to regulate protein synthesis. Cells are believed to regulate both the timing and quantity of proteins synthesized by direct and specific interactions with mRNA. This notion is inconsistent with the impression obtained by reading scientific writings about gene regulation that focuses strongly on transcription. RNA mutation, translocation, intracellular placement, and translation processes provide a good opportunity for drug binding
Abundant in recognition sites. Applicants' invention demonstrates the discovery of these sites for RNA molecules in the human genome as well as in other animal and prokaryotic genomes.

[0007] It is therefore a primary object of the present invention to identify molecular interactions in nucleic acids, especially RNA. It is a further object of the present invention to identify secondary structural elements in RNA that strongly trigger important therapies, controls, or other interactions with "small" molecules and the like. Tissue-enriched unique structure rich in tissue in RNA
Identification of structures) is another object of the present invention.

SUMMARY OF THE INVENTION The applicant's invention provides for eukaryotic and prokaryotic organisms referred to as “molecular interaction sites”.
2 shows a method for identifying a secondary structure in an RNA molecule. The molecular interaction site is small, preferably 50 nucleotides or less, or 30 nucleotides or less, and is an independently folded, functional subdomain contained within a larger RNA molecule. Applicant's method preferably involves a family of integrated processes that analyze the sequence of nucleic acids, preferably RNA, and deduce their structure and function. Applicant's method preferably includes a process of executing a subroutine in an array, where the results of one process are used to elicit a specific course of action, or to a numerical or other Provide input for Preferably, there is a decision point in the routed process that is adopted by a clever process that is determined without detailed real-time human intervention. Automation of RNA sequence analysis provides the ability to identify regulatory sites at the rate at which RNA sequences are available from genomic sequence databases and others. For example, by using the present invention,
The site of molecular interactions associated with central nervous system (CNS) diseases, metabolic diseases, pain, degenerative diseases of aging, cancer, inflammatory diseases, cardiovascular diseases and many other conditions can be identified. For example, using the applicant's invention may serve as a site for "small" molecules that are lacking in eukaryotes, especially humans, but that are associated with the concomitant regulation, increase or decrease of prokaryotic RNA. Possible molecular interaction sites can also be determined. Thus, human toxicity can be avoided in the treatment of viral, bacterial or parasitic diseases.

The present invention preferably identifies a molecular interaction site in a target nucleic acid by comparing the nucleotide sequence of the target nucleic acid with the nucleotide sequence of a number of nucleic acids from different taxonomic species, and is effective at the number of nucleic acids and the target nucleic acid. And identifying whether the conserved region has a secondary structure, and if the conserved region has a secondary structure, identifying the secondary structure.

[0010] The present invention also provides databases relating to molecular interaction sites in eukaryotic and prokaryotic RNA. The database is obtained by comparing the nucleotide sequence of the target nucleic acid with the nucleotide sequence of multiple nucleic acids from different taxonomic species, identifying the multiple nucleic acids and at least one sequence region conserved in the target nucleic acid, Determine if has a secondary structure, identify the secondary structure if the conserved region has a secondary structure, and compile a group of such secondary structures.

[0011] The invention also relates to the RNA of the selected organism and the RNA of at least one further organism.
Figure 1 shows an oligonucleotide containing a molecular interaction site present in a selected organism when the molecule binds to the molecular interaction site.
Serves as a binding site for at least one molecule that regulates the expression of A.

[0012] The invention also relates to prokaryotic RNA and at least one additional prokaryotic RNA.
Oligonucleotides containing a molecular interaction site present in a molecule, wherein the molecular interaction site serves as a binding site for at least one molecule that regulates prokaryotic RNA expression when the molecule binds to the molecule interaction site .

[0013] The invention also relates to prokaryotic RNA and at least one additional prokaryotic RNA.
The present invention relates to a pharmaceutical composition comprising an oligonucleotide having a molecular interaction site present in the compound, wherein the molecular interaction site serves as a binding site for at least one "small" molecule. Such a molecule, when bound to a molecular interaction site, regulates the expression of prokaryotic RNA. Preferably, it also contains a pharmaceutical carrier.

The invention also relates to the RNA of the selected organism and the RNA of at least one further organism.
A pharmaceutical composition comprising an oligonucleotide comprising a molecular interaction site present at The molecular interaction site serves as a binding site for at least one molecule that regulates expression of RNA in the selected organism when the molecule binds to the molecular interaction site, and as a pharmaceutical carrier.

[0015] Finally, the method of the present invention identifies physical structures present in the target nucleic acid that are of great importance to the organism in which the nucleic acid is present. Such structures, so-called molecular interaction sites, can alter the properties and effects of nucleic acids by interacting with molecular species. This can be a therapeutic development, as would be appreciated by one of skill in the art. Such structures may also be found in the nucleic acids of organisms of great importance in agriculture, pollution control, industrial biochemistry, and others. Therefore, industrial organisms such as pesticides, herbicides, fungicides, yeasts, recently viruses, and biocatalytic systems may be more beneficial to this result.

While there are a number of ways to characterize the binding between a molecular interaction site and a ligand, such as an organic complex, preferred methodologies are, for example, each filed on May 12, 1998, each of which is incorporated herein by reference. US serial numbers 09 / 076,440, 09 / 076,405, 09 / 076,447, 09 / 076,206, 09 / 076,214,
And 09 / 076,404. All of the foregoing applications are incorporated by reference herein in their entirety.

The present invention provides a method for identifying specific structural elements in eukaryotic and prokaryotic nucleic acids, particularly RNA molecules, that will interact with other molecules to effect RNA regulation. "Modulation" refers to increasing or decreasing RNA activity or expression. Preferred embodiments of the present invention are outlined in the flowchart of FIG. Structural elements in eukaryotes and prokaryotes are referred to as "molecular interaction sites." These elements contain secondary structure, ie, have a three-dimensional form that can undergo interactions with “small” molecules and others, and in therapeutic and other applications “small” molecules, oligomers such as oligonucleotides, And are expected to serve as sites for interacting with other mixtures.

Referring to FIG. 1, a preferred step for identifying molecular interaction sites in a target nucleic acid is shown in a flow diagram. The nucleotide sequence of the target nucleic acid is compared to the nucleotide sequences of multiple nucleic acids from different taxonomic species, ten. The target nucleic acid may be present in a eukaryotic or prokaryotic cell, and the target nucleic acid may belong to a bacterium or a virus, as well as to "higher" organisms such as humans. Any type of nucleic acid can serve as a target nucleic acid. More preferred target nucleic acids are messenger RNA (mRNA), premessenger RNA (pre-mRNA), transfer RNA
(TRNA), ribosomal RNA (rRNA), or small nuclear RNA (snRNA), but is not limited thereto. The initial selection of a particular target nucleic acid can be based on any functional criteria. For example, nucleic acids known to be important in inflammation, cardiovascular disease, pain, cancer, arthritis, trauma, obesity, Huntington's disease, neuropathy, or other diseases or disorders are typical target nucleic acids. For example, nucleic acids known to be involved in pathogenic genomes, such as bacterial, viral and yeast genomes, are typical prokaryotic nucleic acid targets. Pathogenic bacteria, viruses and yeasts are well known to those skilled in the art. Typical nucleic acid targets are shown in Table 1. However, it should be understood that the applicant's invention is not limited to the targets set forth in Table 1, and that the present invention is believed to be fairly universal.

[0019]

[Table 1]

[0020]

[0021]

[0022]

[0023] Additional nucleic acid targets can be determined independently or can be selected from prokaryotic and eukaryotic gene databases available to the public and known to those of skill in the art. Preferred databases include, for example, online Mendelian genetics in humans (
Online Mendelian Inheritance in Man (OMIM), The Cancer Genome Anatomy Project (CGAP), GenBank, EMBL, PIR, SWISS
-PROT etc. are included. OMIM is a database of genetic mutations associated with disease, but is partly a national center for biotechnology information.
Biotechnology Information) (NCBI). OMIM can be accessed via the Internet at, for example, http://www.ncbi.nlm.nih.gov/Omim/. CGAP is an interdisciplinary program that establishes the information and technical tools needed to decipher the molecular analysis of cancer cells. CGAP can be accessed via the Internet at, for example, http://www.ncbi.nlm.nih.gov/ncicgap/. Some of these databases may include complete or partial nucleotide sequences. In addition, nucleic acid targets can be selected from private gene databases. Alternatively, a nucleic acid target can be selected from available publications, or a nucleic acid target can be determined specifically for use in connection with the present invention.

After selecting or providing the nucleic acid target, determining the nucleotide sequence of the nucleic acid target,
And then comparing with the nucleotide sequence of multiple nucleic acids from various taxonomic species. In one embodiment of the invention, the nucleotide sequence of the nucleic acid target is determined by scanning at least one genetic database or identified in available publications. Preferred databases known and available to those skilled in the art include, for example, the Expressed Gene Anatomy Database
e) (EGAD), Unigene-Homo Sapiens Database
database) (Unigene), GenBank, etc. EGAD encompasses a non-redundant set of human transcribed (HT) sequences, and via the Internet, for example,
It can be accessed at http://www.tigr.org/tdb/egad/egad.html. Unigene is a system that automatically divides GenBank sequences into non-redundant sets of gene-adapted clusters. Each unigene cluster contains sequences that display not only a single gene, but also relevant information such as the tissue type in which the gene is expressed, and the location of the genetic map.

In addition, Unigene contains tens of thousands of novel expressed sequence tag (EST) sequences.
Unigene is available via the Internet, for example, http://www.ncbi.nlm.nih.gov/
It can be accessed at UniGene /. These databases can be used, for example, in conjunction with search programs such as Entrez, which are known and available to those skilled in the art. Entrez is available via the Internet, for example, http: //www.ncbi.nlm.nih.
You can access it at gov / Entrez /. Preferably, the most complete nucleic acid sequence representation available from various databases is used. The GenBank database is
Although known and available to those skilled in the art, it can also be used to obtain the most complete nucleotide sequence. GenBank is the NIH gene sequence database, and is an annotated collection of DNA sequences available to all public. GenBank is described, for example, in Nuc. Acids Res., 1 herein incorporated by reference in its entirety.
998, 26, 1-7, and those skilled in the art, for example,
It can be accessed at http://www.ncbi.nlm.nih.gov/Web/Genbank/index.html. Alternatively, if a complete nucleotide sequence is not available, a partial nucleotide sequence of the nucleic acid target can be used.

In another aspect of the invention, the nucleotide sequence of a nucleic acid target is determined by assembling a plurality of overlapping expressed sequence tags (ESTs). EST database (d
bEST) is known and available to those of skill in the art, but includes one million different human mRNA sequences, including approximately about 500-1000 nucleotides, and various numbers of ESTs from many different organisms. dbEST, via the Internet, for example, http: // www
You can access it at .ncbi.nlm.nih.gov / dbEST / index.html. These sequences are derived from a cloning strategy using cDNA expression clones for genomic sequencing. ESTs have applications in discovering new genes, mapping the genome, and identifying coding regions in genomic sequences. ES rapidly becoming available
Another important feature of T sequence information is tissue-specific gene sequence data. This can be extremely useful in targeting the gene (s) selected for therapeutic intervention. Because EST sequences are relatively short, they must be assembled to provide a complete sequence. Each available clone has been sequenced, resulting in a large number of overlapping regions reported in the database.

Combining overlapping ESTs that extend in both the 5 ′ and 3 ′ directions results in a full-length “virtual transcript”. The resulting virtual transcript may represent a nucleic acid that has already been characterized or may be a novel nucleic acid with an unknown biological function. The Institute for Genomic Research (TI
The GR) Human Genome Index (HGI) database is known and available to those of skill in the art, but includes a list of human transcripts. The TIGR can be accessed via the Internet, for example, at http://www.tigr.org/. Virtual transcripts were constructed and transcripts were generated by this method using TIGR-Assembler, known and available to those of skill in the art. TIGR-Assembler is a tool for assembling many sets of overlapping sequence data, such as ESTs, BACs, or small genomes, and can be used to assemble eukaryotic or prokaryotic sequences. TIGR-Assembler is described, for example, in Sutton et al. (Genome Science & Tech., 1995, 1, 9-19), which is hereby incorporated by reference in its entirety, and is described via the Internet. For example, ftp://ftp.tigr.org/pub/s
You can access it at oftware / TIGR assembler. In addition, GLAXO-MRC, which is known and available to those skilled in the art, is another protocol for constructing virtual transcripts. In addition, "Find Neighbo running on UNIX platform
The “rs and Assemble EST Blast” protocol has been developed by Applicants to construct virtual transcripts. The preferred steps in the Find Neighbors and Assemble EST Blast protocol are described in the flow chart set forth in FIG. .PHRAP,
Used for sequence assembly during Find Neighbors and Assemble EST Blast. PHR
AP can be accessed via the Internet, for example, http: //chimera.biotech.washington.ed
You can access it at u / uwgc / tools / phrap.htm. One skilled in the art can construct the source code and perform the preferred steps described in FIG.

[0028] The nucleotide sequence of the nucleic acid target is compared with the nucleotide sequence of multiple nucleic acids from various taxonomic species. Multiple nucleic acids from various taxonomic species, and their nucleotide sequences, can be found in available databases, in genetic databases, or can be specifically determined for use in connection with the present invention. it can. In one aspect of the invention, a nucleic acid target is combined with the nucleotide sequence of multiple nucleic acids from various taxonomic species by performing sequence identity searches, ortholog searches, or both, as known to those of skill in the art. Compare.

[0029] The result of the sequence identity search is a plurality of nucleic acids having at least a portion of the nucleotide sequence of the plurality of nucleic acids, which is based on at least an 8-20 nucleotide region of the target nucleic acid, referred to as a window region. And are homologous. Preferably, the plurality of nucleotide sequences comprises at least a portion having at least 60% homology to any window region of the target nucleic acid. More preferably, the homology is at least 70%. More preferably, the homology is at least 80%. Most preferably, the homology is at least 90%. For example, the window size, which is the portion of the target nucleotide to which multiple sequences are compared, may be from about 8 to about 20 contiguous nucleotides, preferably 10-15, and most preferably about 11-12. The window size can be adjusted in this way. Thereafter, preferably, the plurality of nucleic acids from the various taxonomic species are compared to each appropriate window in the target nucleic acid until all portions of the plurality of sequences are compared to the window of the target nucleic acid. A variety of taxonomics having portions that are at least 60%, preferably at least 70%, more preferably at least 80%, or most preferably at least 90% homologous to any of the window sequences of the target nucleic acid The sequence of multiple nucleic acids from a species is considered a suitable homologous sequence.

[0030] Sequence identity searches can be performed manually, or using any of the available computer programs known to those of skill in the art. Preferably, Blast and Smith-Waterman algorithms and the like available and known to those skilled in the art can be used. Blast is NCBI's sequence identity search tool designed to support analysis of nucleotide and protein sequence databases.
Blast is available via the Internet, for example, http://www.ncbi.nlm.nih.gov/BLAST
You can access it at /. The GCG package provides a local version of Blast that can be used with either a public database or a locally available searchable database. GCG Package v9.0 is a commercially available software package that contains over 100 interrelated software programs that allow sequence analysis by editing, mapping, comparing, and aligning sequences. is there. Other programs included in the GCG package include, for example, programs that facilitate RNA secondary structure prediction, nucleic acid fragment assembly, and evolutionary analysis. In addition, the most prominent gene databases (Ge
nBank, EMBL, PIR, and SWISS-PROT) are distributed with the GCG package,
And it is completely usable for database search programs and database operation programs. GCG can be accessed via the Internet, for example, http://www.gcg.com/
You can access at Fetch is a tool available in GCG that can obtain GenBank records annotated based on accession numbers, and is similar to Entrez. Another sequence identity search using GeneWorld from Pangea
And GeneThesaurus. GeneWorld 2.5 is an automated, flexible, and flexible tool for analyzing polynucleotide and protein sequences.
High throughput application. GeneWorld enables automated analysis and sequence annotation. Like GCG, GeneWorld employs several tools for sequence search, gene discovery, multiple sequence alignment, secondary structure prediction and motif identification. GeneThesaurus 1.Otm is a sequence and annotation data subscription service that provides information from multiple sources and provides a relational data model for public and local data.

Another alternative sequence identity search can be performed with, for example, BlastParse. Blas
tParse is a PERL script that runs on UNIX platforms and automates the strategy described above. BlastParse takes a list of target accession numbers of interest, and takes each through the preferred steps described in the flow chart shown in FIG. BlastParse parses all GenBank ranges in "tab-set" text and then saves the "tab-set" text in a "relational database" format for easier search and analysis, providing flexibility can do. The end result is a series of fully parsed GenBank records that can be easily sorted, filtered, and questionable, and can be an annotation-relational database.

Another toolkit that can perform sequence identity searches and data manipulation is the SE
ALS, also from NCBI. This toolset is written in perl and C
And any computer platform that supports these languages
Can be driven on the system. This is, for example, the following address: http: //www.n
It is available for download from cbi.nlm.nih.gov/Walker/SEALS/. This
The kit provides access to Blast2 or gapped blasts. that is
Also, tax tax to analyze the output of Blast2 in combination with a tool called break  Includes a tool called collector, and the quality for each species present
Returns the identifier of the most homologous sequence to the query sequence. Another useful tool is feat
ure2fasta, which extracts sequence fragments from input sequences based on annotations. This
A typical use of this tool is to generate a sequence file containing the 5 'untranslated region of a cDNA sequence.
Is to create.

Preferably, a plurality of nucleic acids from various taxonomic species having homology to the target nucleic acid, as described above in the sequence identity search, are further refined to find the ortholog of the target nucleic acid therein. To be described. Orthologs have sequence identity in widely divergent organisms and perform similar functions in the context of organisms.
A term defined in the genetic classification for referring to a gene. In contrast,
Paralogs are genes within a species that arise due to gene duplication, but evolve new functions and are also referred to as isoforms. In some cases, a paralog search can also be performed. By performing an ortholog search, an exhaustive list of homologous sequences from various organisms is obtained. Subsequently, we analyze these sequences,
Select the most representative sequence that fits the classification for orthologs. The ortholog search can be performed by a program available to those skilled in the art, for example, including Compare. Preferably, the ortholog search is performed by accessing the complete and analyzed GenBank annotations for each of the sequences. At present, records obtained from GenBank are "flat-file" and are not ideally suited for automated analysis. Preferably, the ortholog search is a Q-Comp
are done using the program. The preferred steps of the Q-Compare protocol are described in the flow chart described in FIG. The Blast result-relation database shown in Fig. 3 and the annotation-relational database shown in Fig. 3
are used in the are protocol, resulting in a list of ortholog sequences for comparison in the interspecies sequence comparison program described below.

With the above-described identity search, a result based on a cutoff value called an e-score is obtained. The e-score indicates the possibility of random sequence matching within a given nucleotide window. The lower the e-score, the better the fit. Those skilled in the art are familiar with e-scores. The user defines the e-value cutoff based on the stringency or degree of homology desired, as described above. In embodiments of the invention that identify a prokaryotic molecule interaction site, it is preferred that any of the identified homologous nucleotide sequences be non-human.

In another aspect of the invention, the required sequence is obtained by searching an ortholog database. One such database is Hovergen, a curated database of vertebrate orthologs. The set of orthologs is output from this database and used as seed for further sequence identity searches as described above. Further searches may be desirable, for example, to find invertebrate orthologs. Hovergen can be downloaded, for example, from the following address: ftp://pbil.univ-lyonl.fr/pub/hovergen/. The prokaryotic ortholog database, COGS, is available on the Internet, for example, at the following address: http://www.ncbi.nlm.nih.gov/COG/ and can be used interactively. it can.

In another embodiment of the invention, the nucleotide sequences of multiple nucleic acids from various taxonomic species are determined by performing a sequence identity search using dbEST or the like, and by constructing virtual transcripts, Compare with the nucleotide sequence of the target nucleic acid. Using EST information is useful for two distinct reasons. First, the ability of human genes to identify orthologs for human genes in evolutionarily distinct organisms in the GenBank database is limited. As more efforts are being directed at identifying ESTs from these evolutionarily distinct organisms, dbEST appears to be a better source of ortholog information.

Second, the effort to sequence the human genome is less than 10% complete. Thus, it is believed that human dbEST will provide more information to identify key targets as the sequence of the human genome is fully approached. EST sequences are short and need to be assembled and used. Preferably, as described above
A sequence identity search is performed under high stringency conditions on dbEST excluding human sequences using the Smith-Waterman algorithm. Since dbEST encompasses sequencing errors, including insertions and deletions, in order to accurately search for new sequences, the search strategy used should also allow for these gaps. As each available clone has been sequenced, a large number of overlapping regions reported in the database will result. Full-length or partial "virtual transcripts" for non-human RNA are constructed by extending overlapping EST sequences in both 5 'and 3' directions until a "full-length" transcript is obtained. In another aspect of the invention, a chimeric virtual transcript is constructed.

The resulting virtual transcript may represent an already characterized RNA molecule, or it may be a new RNA molecule with unknown biological function. As mentioned above, the TIGR HGI database makes available an engine called TIGR-Assembler for constructing virtual transcripts. GLAXO-MRC and GeneWorld from Pangea also provide for the construction of virtual transcripts. As mentioned above, Find Neighbor
Virtual transcripts can also be constructed using s and Assemble EST Blast.

Referring to FIG. 1, after obtaining the above-described orthologs or virtual transcripts via either a sequence identity search or an ortholog search, multiple nucleic acids from various taxonomic species and the target nucleic acid are compared. 20. identify at least one sequence region that is conserved between Interspecies sequence comparisons can be performed using a number of computer programs, available and known to those of skill in the art. Preferably, interspecies sequence comparisons are performed using Compare, which is available and known to those skilled in the art. Compare
Is a pair-wis comparison of sequences using window / stringency criteria.
This is a GCG tool that enables e comparisons. Compare creates an output file containing heavens where a particular quality match is found. These can be plotted with another GCG tool, DotPlot.

Alternatively, identification of conserved sequence regions is performed by interspecies sequence comparison using the ortholog sequences generated from Q-Compare in combination with CompareOverWins as described above. Preferably, a list of sequences to compare, for example an ortholog sequence generated from Q-Compare as described in FIG. 4, is input into the CompareOverWins algorithm. The preferred steps in CompareOverWins are described in FIGS. 5A, 5B, and 5C. Preferably, interspecies sequence comparisons are performed by paired sequence comparisons, wherein the query sequence slides over a window on the master target sequence. Preferably, the window is from about 9 to about 99 contiguous nucleotides.

The sequence homology between the window sequence of the target nucleic acid and the query sequence of any of the plurality of nucleic acid sequences obtained as described above is preferably at least 60%, more preferably at least 70%, more preferably At least 80%, and most preferably at least 90%. The most preferred way to select the threshold is to use a computer
Let all 0% thresholds be tried automatically and select the thresholds based on the metrics provided by the user. One such criterion is to choose a threshold such that exactly n hits are returned, where n is typically 3
Set to. All the bases of the query nucleic acid, which are members of the above-described multiple nucleic acids,
This procedure is repeated until all bases of the master target sequence have been compared. The resulting score matrix can be plotted as a scatter plot. Based on the fit density at a given location, no dots, separated dots,
Or it could be a set of dots that are so close to be seen as a straight line. The presence of a straight line, although small, indicates primary sequence homology. A representative diffusion plot for such interspecies sequence comparisons is shown in FIG. Sequence conservation within nucleic acid molecules in branched species, especially in the UTR of RNA, appears to be indicative of a conserved regulatory element that also appears to have secondary structure. The results of interspecies sequence comparisons can be analyzed using MS Excel and Visual basic tools in a fully automated fashion as known to those skilled in the art.

Referring to FIG. 1, after the nucleotide sequence of the nucleic acid target and at least one region conserved between nucleic acids from various taxonomic species has been identified, preferably via an ortholog Analyzes the conserved region to determine if it contains secondary structure, 30. Determining whether an identified and conserved region contains secondary structure can be accomplished by a number of procedures known to those of skill in the art. The determination of secondary structure is preferably performed by self-complementarity comparison, juxtaposition and covariance analysis,
It is performed by secondary structure prediction or a combination thereof.

In one embodiment of the present invention, secondary structure analysis is performed by juxtaposition and covariance analysis. Numerous protocols for juxtaposition and covariance analysis are known to those skilled in the art.
Preferably, the juxtaposition is with ClustalW, which is available and known to those skilled in the art. ClustalW is a tool for juxtaposing multiple sequences and is not part of GCG,
It can be added as an extension of the existing GCG toolset and can be used for local arrays. ClustalW can be accessed via the Internet, for example, at the following address: http://dot.imgen.bcm.tmc.edu:9331/multialign/Options/clustal
It can be accessed from w.html. Clusta1W has also been described by Thompson et al. (Nuc. Aci.
ds Res., 1994, 22, 4673-4680), which is hereby incorporated by reference in its entirety. These methods can be scripted to automatically use conserved UTR regions identified in earlier steps. Seqe, a UNIX command line interface available and known to those skilled in the art
d allows selected local regions to be extracted from a larger array. Numerous sequences from many different species can be clustered and juxtaposed for further analysis.

In a preferred embodiment of the invention, all possible pairing CompareOverWindo
Collect the output of the ws comparison and align it with the reference sequence using a program called AlignHits. A diagram of the operation of this program is provided in FIG. 5D. This program can be repeated by those skilled in the art. The preferred purpose of this program is to remap all hits created in pairwise comparisons to locations on the reference sequence. This method, combined with CompareOverWindows and AlignHits, provides more local alignment (20-100 bases or more) than any other algorithm.
This local juxtaposition is necessary for a structure discovery routine described below, such as Covariation or RevComp. This algorithm describes a fasta array of juxtaposed arrays. As indicated, the algorithm does not correct single base insertions or single base deletions. This is usually achieved by placing the output via ClustalW, described elsewhere. It is important to differentiate ClustalW from using ClustalW alone without CompareOverWindows and AlignHits.

Covariation is a method that uses phylogenetic analysis of primary sequence information for consensus secondary structure prediction. Covariation is described in the following documents, each of which is incorporated herein in its entirety (Gutell, et al., "Comparative Sequence An
alysis Of Experiments Performed During Evolution "In Ribosomal RNA Group
I Introns, Green, Ed., Austin: Landes, 1996; Gautheret, et al., Nuc. Aci
ds Res., 1997, 25, 1559-1564; Gautheret, et al., RNA, 1995, 1, 807-814;
Lodmell, et al., Proc. Natl. Acad. Sci. USA, 1995, 92, 10555-10559; Gauthe
ret, et al., J Mol. Biol., 1995, 248, 27-43; Gutell, Nuc. Acids Res., 199.
4,22,3502-3517; Gutell, Nuc. Acids Res., 1993, 21, 30553074; Gutell, Nuc.
Acids Res., 1993, 21, 3051-3054; Woese, Proc. Nad. Acad. Sci. USA, 1989.
, 86, 3119-3122; and Woese, et al., Nuc. Acids Res., 1980, 8, 2275-2293
). Preferably, covariance software is used for covariance analysis. Preferably, a set of programs for covariation, ie, comparative analysis of RNA structure from sequence alignment, is used. Covariation uses phylogenetic analysis of primary sequence information for consensus secondary structure prediction. Covariation can be performed via the Internet, for example, http
It can be obtained at http://www.mbio.ncsu.edu/RNaseP/info/programs/programs.html. A complete description of the program version is published (Brown, J.
W. 1991 Phylogenetic analysis of RNA structure on the Macintosh computer
CABIOS7: 391-393). The current version is v4.1, which can perform various types of covariation analysis derived from RNA sequence alignments, including standard covariation analysis, identification of compensatory base changes, and mutual information analysis. The program is well educated,
And comes with a wide range of example files. It is compiled as a stand-alone program; it does not require Hypercard (although it includes a fairly small 'stack' version). This program runs on MacOS v7.1 or higher
Can be driven in any of the cintosh environments. A faster processor machine (68040 or PowerPC) is suggested for mutual information analysis or analysis of large sequence juxtapositions.

In another embodiment of the present invention, the secondary structure analysis is performed by secondary structure prediction. There are a number of algorithms that predict RNA secondary structure based on thermodynamic parameters and energy calculations. Preferably, the secondary structure prediction is performed using M-fold or RNA Structure 2.
Perform using one of the 52. M-fold, for example, http
It can be accessed at http://www.ibc.wustl.edu/zuker/ma/form2.cgi or downloaded for local use on UNIX platforms. M-fold is also available as part of the GCG package. RNA Structure
2.52 is a Windows adaptation of the M-fold algorithm and can be accessed via the Internet, for example at http://128.151.176.70/RNAstructure.html.

In another embodiment of the present invention, secondary structure analysis is performed by self-complementarity comparison. Preferably, the self-complementarity comparison is performed using Compare as described above. More preferably, Compare can be modified to extend the pairing matrix to account for GU or UG base pairs in addition to traditional Watson-Crick GC / CG or AU / UA pairs. Such a modified Compare program (modified Compare) begins by predicting all possible base pairings within a given sequence. As described above, small but conserved regions, preferably UTRs, are identified based on primary sequence comparisons of a series of orthologs. In modified Compare, each of these sequences is compared to its own reverse complement. FIG. 7 shows a specific self-complementary analysis. Acceptable matches include Watson-Crick AU, GC matches and non-standard GU matches. The overlay of such self-complementary plots of all available orthologs, and the choice for most repetitive patterns in each, results in a minimal number of possible folds. FIG. 8 shows a specific overlay. Such overlays can then be used with additional constraints, including those imposed by the energy considerations described above, to deduce the most likely secondary structure.

In another preferred embodiment of the present invention, the output of AlignHits is read by a program called RevComp. The block diagram of this program is shown in FIG. This program can be repeated by those skilled in the art. The preferred purpose of this program is to use base pairing rules and to use ortholog evolution to predict RNA secondary structure. The RNA secondary structure consists of a single-stranded region and a base-pairing region, the so-called stem. To search for evolutionarily conserved structures, the most likely stem for a given juxtaposition of orthologous sequences is one that can be formed by most sequences. Possible stem formation or base pairing rules are determined, for example, by analyzing the base pair statistics of the stem as determined by other techniques, eg, NMR. The output of RevComp is a sorted list of possible structures, ranked by the percentage of sequences of ortholog set members that can form the possible structures. This approach does not respond to noise sequences because it uses a percentage threshold approach. A noise sequence is a sequence that does not give an example of a structure to be searched, but is not a true ortholog or is created in the output of AlignHits due to high sequence homology. A very similar algorithm is called Visual basic for Applica to run on a PC.
It is performed using Options (VBA) and Microsoft Excel to generate a reverse complementarity matrix overview for a given set of sequences.

Whether performed by juxtaposition and by covariance, self-complementary analysis, secondary structure prediction, eg, M-fold, or the like, the results of the secondary structure analysis described above are: Identify secondary structures in regions conserved between the target nucleic acid and multiple nucleic acids from various taxonomic species, 40. Typical secondary structures that can be identified include bulges, loops, stems, hairpins, knots, triple interacts, cloverleaves, or helices, or combinations thereof. But not limited to these. Alternatively, a novel duplex structure may be identified.

In another aspect of the invention, once the secondary structure of the conserved region has been identified as described above, at least one structural motif for the conserved region having the secondary structure is identified. These structural motifs correspond to the secondary structures identified as described above. For example, analysis of secondary structure by self-complementarity can provide one type of secondary structure, while analysis by M-fold can provide another secondary structure. Thus, all possible secondary structures identified by the secondary structure analysis described above are represented by a family of structural motifs.

[0051] Once the secondary structure (s) of the target nucleic acid and the secondary structure of nucleic acids from various taxonomic species have been identified, rather than by primary nucleotide sequence, as described above, Additional nucleic acids can be identified by structure-based searches. By constructing a family of descriptor elements for the structural motifs described above, additional nucleic acids having a secondary structure similar or identical to the secondary structure found as described above, and corresponding to the descriptor elements By identifying another nucleic acid having a secondary structure, the nucleic acid can be identified. Any or all combinations of nucleic acids having a secondary structure can be collected in a database. The entire method can be repeated for another target nucleic acid to generate multiple different secondary structure groups that can be assembled in a database. Therefore, by performing the method according to the invention described in this specification, a database of molecular interaction sites can be collected.

After determining a hypothetical structural motif from the secondary structure analysis described above, a family of structural descriptor elements is constructed. Preferably, the structural motifs described above are converted into a family of descriptor elements. Typical descriptor elements are shown in FIG. The construction of descriptors is well known to those skilled in the art. Structural descriptors are described, for example, in Laferriere et al. (Comput. Appl. Biosci., 1994, 1), which is hereby incorporated by reference in its entirety.
0, 211-212). A separate structural descriptor element is constructed for each of the structural motifs identified from the secondary structure analysis. Briefly, the secondary structure is converted into a general text string, for example, as shown in FIG. For new motifs, further biochemical analysis, such as chemical mapping or mutagenesis, may be required to confirm the structure prediction. Descriptor elements can be defined to have different stringencies.

For example, referring to FIG. 9, the region described as H1, including the first region of the stem,
NN: Can be described as NNN, which contemplates any of the complementary base pairing including GC, CG, AU, and UA. The H1 region can also be shown to include only base pairing such as CG or AU. In addition, descriptor elements can be defined to allow for agitation. In this manner, the descriptor elements can be defined to have any level of stringency desired by the user. Applicants' invention is thus also directed to a database containing various descriptor elements.

After constructing a family of structural descriptor elements, a nucleic acid having a secondary structure corresponding to the structural descriptor element is identified. Preferably, at least one database is searched, clustered and analyzed to identify orthologs, or a combination thereof, to identify nucleic acids having a secondary structure corresponding to a structural descriptor element. Thus, the identified nucleic acid has a secondary structure that falls within the secondary structure defined by the descriptor element. Thus, the identified nucleic acid has the same or almost the same secondary structure as the target nucleic acid, depending on the stringency of the descriptor element.

In one embodiment of the invention, a nucleic acid having a secondary structure corresponding to a structural descriptor element is identified by searching at least one database. Any gene database can be searched. Preferably, the database is a UTR database that collects untranslated regions in messenger RNA. The UTR database can be accessed via the Internet, for example, ftp://area.ba.cnr.it/pub/embnet/data
You can access it at base / utr /. Preferably, the database is a UNIX-based motif search tool available, for example, from Daniel Gautheret
Search using a computer program such as Rnamot. Thereafter, each "new" sequence having the same motif is questioned against a public database to identify additional sequences. Of these additional ortholog sequences as described above.
Analyze the results for pattern recurrence in the UTR and build a database of RNA secondary structures. Rnamot is well known to those skilled in the art. Simply put, Rn
amot takes a descriptor string as shown in FIG. 9 and searches any of the Fasta format databases for possible matches. Descriptors may be very specific for exact nucleotide matching or may have built-in degeneracy. The length of the stem and loop may be specified. The single-stranded loop region may have a variable length. GU pairs are also allowed and can be specified as wobble parameters. Acceptable mismatches can also be included in the definition of the descriptor. Functional significance is assigned to motifs if the biological role of the motif is known based on previous analysis. Known control regions, such as, for example, an Iron Response Element, have been discovered using this technique (see Example 1 below). In embodiments of the invention that collect a database containing prokaryotic molecular interaction sites, it is preferable to search for human sequences or, if found, to discard human sequences.

In another aspect of the invention, a database, such as a UTR database, is
Nucleic acids identified by searching using namot are clustered and analyzed to determine their location in the genome. The results provided by Rnamot merely identify sequences containing secondary structure, but do not give any indication as to the location of the sequence in the genome. Clustering and analysis are preferably performed using ClustalW as described above.

In another aspect of the invention, after clustering and analysis have been performed as described above, orthologs are identified as described above. However, in contrast to the above-identified orthologs that were simply identified based on their primary nucleotide sequence, the sequences of these novel orthologs are based on structures using nucleic acids identified using Rnamot. Identified. Preferably, orthologs are identified by BlastParse or Q-Compare as described above. In embodiments of the invention that collect a database containing prokaryotic molecule interaction sites, it is preferred that human orthologs be discovered or, if found, to discard human orthologs.

After identifying a nucleic acid having a secondary structure corresponding to a structural descriptor element, any or all of the nucleotide sequence can be assembled in a database by standard editing protocols known to those skilled in the art. One database may contain eukaryotic molecular interaction sites, and another database may contain prokaryotic molecular interaction sites.

The present invention is also directed to oligonucleotides comprising a molecular interaction site present in the RNA of the selected organism and in at least one, but rather in some additional, RNA of the organism. The nucleotide sequence of the oligonucleotide is chosen to provide the secondary structure of the molecular interaction site described above. The nucleotide sequence of the oligonucleotide is preferably the nucleotide sequence of the target nucleic acid described above. Or,
The nucleotide sequence is preferably the nucleotide sequence of a nucleic acid from a plurality of different taxonomic species that also contains a molecular interaction site. The molecular interaction site functions as a binding site for at least one molecule that, when bound to the molecular interaction site, regulates the expression of RNA in the selected organism.

The invention also relates to an oligonucleotide comprising a molecular interaction site present in a prokaryotic RNA and at least one additional prokaryotic RNA, wherein the molecular interaction site binds to the molecular interaction site In this case, it functions as a binding site for at least one molecule that regulates prokaryotic RNA expression. An additional organism is selected from all eukaryotic and prokaryotic organisms and cells, but is not the same organism as the selected organism. Oligonucleotides and modifications thereof are well known in the art. The oligonucleotides of the invention can be used, for example, as a research reagent, for example, to detect naturally occurring molecules that bind to a molecular interaction site. The oligonucleotides of the invention can also be used as decoys to compete with naturally occurring molecular interaction sites in cells for research, diagnostic and therapeutic uses. The molecule that binds to the site of molecular interaction is modified, either by increasing or decreasing expression of the RNA. The oligonucleotide is
It can also be used in agricultural, industrial and other applications.

The present invention is also directed to a pharmaceutical composition comprising an oligonucleotide as described above together with a pharmaceutical carrier. “Pharmaceutical carrier” refers to a pharmaceutically acceptable solvent, diluent, suspension, or other pharmaceutically inert vehicle for delivering one or more nucleic acids to an animal. Either, and they are well known to those skilled in the art. The carrier can be a liquid or solid and, when combined with the other components of the pharmaceutical composition, provides a desired bulk, consistency, etc., with the intended mode of administration being maintained while maintaining the carrier in a controlled manner. select. Typical pharmaceutical carriers include binders such as pregelatinized corn starch, polyvinylpyrrolidone, or hydroxypropylmethylcellulose; fillers such as lactose and other sugars, microcrystalline cellulose, pectin, Gelatin, calcium sulfate, ethylcellulose, polyacrylic acid, or calcium hydrogen phosphate; a lubricant (eg, magnesium stearate, talc, silica, colloidal silicon dioxide, stearic acid, metal stearate, hydrogenated vegetable oil, corn starch, Disintegrants (eg, starch, sodium starch gluconate, etc.); or wetting agents (eg, sodium lauryl sulfate, etc.). Not.

The following examples show specific examples of preferred embodiments of the present invention, but they are not meant to be limiting.

[0063]

【Example】

Example 1 Iron-Reactive Element 1. Selection of RNA Target To illustrate a strategy for identifying small molecule interaction sites, the iron-reactive element (IRE) in the mRNA encoded by the human ferritin gene was identified. Identify. IRE
Is a typical example of an RNA structural element used to regulate the level of translation of mRNA associated with iron metabolism. The structure of the IRE has recently been determined using NMR spectroscopy. In addition, NMR analysis of the IRE structure is described by Gdaniec et al. (Biochem., 1998, 37, 1505-1512) and Adde
(J. Mol. Biol., 1997, 274, 72-83). IREs are RNA elements of about 30 nucleotides that fold into a hairpin structure and bind to specific proteins. This structure has been very well studied and is known to be evident in many species of mRNA and thus serves as a great example of how Applicants' methodology works.

2. Determination of Nucleotide Sequence of RNA Target The human mRNA sequence for ferritin was converted to the target sequence or the initial mRNA
Used as NA. Ferritin protein sequences are also used in analysis, particularly in the early steps used to find related sequences. For the human ferritin gene, the best input is the full-length annotated mRNA and protein sequence obtained from UNIGENE. However, the same level of detailed information is not available for many genes of interest. In these cases, alternative sources of master sequence information are obtained, for example, from GenBank, TIGR, the GenBank dbEST department, or from sequence information obtained from private laboratories. Applicants' method works using all levels of input sequence information, but requires fewer steps than using high quality annotated input sequences.

3. Identification of Similar Sequences An early step in the process uses the master sequence (nucleotide or protein) to find and rank relevant sequences in databases (orthologs and paralogs). A sequence identity search algorithm is used for this purpose. All sequence identity algorithms calculate a quantitative measure of identity for each result, compared to the master sequence. An example of a quantitative result is Blast (Bl
ast) E-value obtained from the algorithm. Non-redundant GenBank database blast search using ferritin mRNA as a query sequence
The values illustrate the use of quantitative analysis of the sequence identity search. The E value is the probability that a match between the query sequence and the database sequence will occur for a random opportunity. Therefore, E
The lower the value, the more certain the exact association of the two sequences. A plot of the lowest E value score for ferritin is shown in FIG. Sequences meeting the cutoff criteria are selected for a more detailed comparison according to a set of rules described below. Since the purpose of a sequence identity search is to find orthologs and paralogs that are distantly related, the cutoff criteria should be less stringent,
Otherwise, search targets may be excluded.

4. Identification of Conserved Regions Identification of conserved regions was performed using Q-Compare with CompareOverWins.
se) Performed by sequence comparison. Structural conservation between genes with related functions from different species is a key indicator that can be used to find good drug binding sites. Conserved structures can be identified by using distantly related sequences and splicing the remaining conserved sequences together with analysis of the potential structure. Sequence comparisons are performed between sets of mRNAs from different species, using Q-compare, which can identify traces of sequence conservation from very divergent organisms. Q-compare, along with CompareOverWins, slides one array across the other array, creating a window of a specific size (window
)) To compare all regions of each sequence by measuring the number of matches.

Human and mouse mRNA for ferritin each contain an IRE in the 5′-UTR, and when these are analyzed in this manner, plots showing regions of sequence identity are generated as shown in FIG. . Pairwise analysis of human and mouse ferritin mRNA sequences illustrates some important aspects of this type of analysis.
The region of each mRNA that encodes the amino acid sequence has the highest degree of identity, while the untranslated region has lower identity. In FIG. 19, the position of the IRE is shown. In both human and mouse ferritin mRNAs, the IRE is located at the 5 'end of each mRNA. This explains the important point-the conservation of sequence in the region of the IRE structure is not significant against the background of sequence identity between human and mouse ferritin sequences. In contrast, IRE can be readily identified by comparison of human and trout (FIG. 11) or human and chicken (FIG. 12) ferritin mRNA. This is because the sequence of the UTR between human and trout or between human and chicken has been separated due to a greater evolutionary distance than human and mouse, which is relatively small compared to other mammals. It is theoretical considering the evolutionary distance of humans from birds and fish. A comparison of human and bird and fish sequences provides insight because the natural flow of evolution changes many sequences in the UTR. However, the IRE sequence is more constrained because it forms an important structure. Thus, they are better protruding and can be more easily identified.

The same principle applies when comparing trout and chicken ferritin sequences to each other. Both are separated from humans by about 100 million years of evolution, but they are well separated from each other. This illustrates another important strategy for use in the present invention--a comparison of two non-human RNA sequences can be used to discover regulatory RNA structures without the actual human sequence. Non-human comparison studies actually dictate those skilled in the art where to look to find control sites in humans as potential drug targets.

Evolutionary distances can be used to determine which sequences are not to be compared, as well as which sequences are to be compared. For humans and mice, comparisons between trout and salmon are less informative because their species are too closely related and the IRE is less prominent on the UTR background.
Comparison of human and Drosophila ferritin mRNA sequences does not find IRE in either species, despite the presence of IRE. this is,
This is because the sequence of IRE between human and Drosophila diverges despite the conservation of structure. However, a comparison of Drosophila and mosquito ferritin mRNAs identified the IRE and again explains that human sequences are not required to identify regulatory elements associated with drug discovery in humans.

The software used in the present invention uses a look-up table based on the evolutionary distance between species to determine whether to compare sequence pairs or not. An example of a small look-up table using the embodiment described above is shown in FIG. The look-up table in the present invention includes all species having sequences deposited in GenBank. CompareOverWins
Together with Q-Compare, it determines which sequences are paired.

5. Identification of Secondary Structures Sets of sequences that show evidence of conservation in orthologs and paralogs or other related genes are analyzed for their ability to form internal structures. This involves plotting the sequence on the X-axis from 5 'to 3' and its reverse complement (rev) as in, for example, self-complementary analysis.
This is achieved by analyzing each sequence in a matrix that plots the complement on the Y-axis from 5 'to 3'. Matches that match potential intramolecular base pairs are:
Scored according to the values in the table. When the human ferritin IRE sequence is analyzed in this manner, the diagonal indicates a potential self-complementary region. Each of the embodiments described in this embodiment
Thirteen IRE sequences were analyzed in a similar manner. While each sequence can form a variety of different structures, the most likely structures are common to all sequences. Superposition of plots of all 13 independent sequences (see FIG. 8) leads to a potential structure common to all sequences.

Example 2: Iron-reactive element (method B) 2. Determination of the nucleotide sequence of the RNA target
Used as NA. Ferritin protein sequences are also used in analysis, particularly in the early steps used to find related sequences. For the human ferritin gene, the best input is the full length annotated mRNA from gigene (gi507251)
And the protein sequence. However, the same level of detailed information is not available for many genes of interest. In these cases, alternative sources of master sequence information are obtained, such as from Hovergen and GenBank. The method works using all levels of input sequence information, but requires fewer steps than using high quality annotated input sequences.

3. Identification of Similar Sequences An alternative and preferred approach to discovering orthologs is described in Duret et al. (Nuc.
Acids Res., 1994, 22, 2360-2365), using the Hovergen database and query tool, which is hereby incorporated by reference in its entirety.

The use of Hovergen to identify related sequences is shown in FIG. 16 (species-level phylogenetic tree classification) and FIG. 17 (eye-level classification). Sequences corresponding to each of these orthologs were saved in GenBank format and grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted using SEALS and COWX, as shown in FIG.

4. Identification of Conserved Regions The IRE sequence forms a key structure and is therefore more repressed. Thus, they are more prominent and can be more easily identified even in closely related sequences. However, to make this work for any gene, rewrite the comparison algorithm (see FIGS. 5A-C). This new tool, CompareOverWins, allows dynamic selection of both window size ranges and hit thresholds. This algorithm requires 5 'and 3' UTR sequences parsed and separated as input. We accomplish this using tools available in the Seals Genome Analysis Package described earlier. FIG. 18 describes the steps involved.

To identify iron-reactive elements using the methods described herein, the Compare Over
The Winows algorithm was used and the results were visualized using AlignHits (for the algorithm, FIG. 5D). Representative results are shown in FIG. In addition to the threshold optimization, CompareOverWins also extracts sequences corresponding to hits. ClustalW (version 1.74) was used on the extracted sequences to create positional gapped juxtapositions (Figure 24).
reference). A representative flow scheme for this approach is shown in FIG.

5. Identification of Secondary Structures Sets of sequences that show evidence of conservation in orthologs and paralogs or other related genes are analyzed for their ability to form internal structures. This means that for each matrix in which the sequence is plotted 5 'to 3' on the X-axis and its complement is plotted 5 'to 3' on the Y-axis, as in, for example, a self-complementary analysis, etc. Achieved by analyzing the sequence. Matches corresponding to potential intramolecular base pairs are scored according to the values in the table. When the human ferritin IRE sequence is analyzed in this manner, the diagonal indicates a potential self-complementary region. Each of the 13 IRE sequences described in this example were analyzed in a similar manner. While each array can form a variety of different structures,
The most likely structures are common to all sequences. Superposition of plots of all 13 independent sequences (see FIG. 26) leads to a potential structure common to all sequences.

The above scheme is algorithmically executed in a program called RevComp (see FIG. 14). RevComp creates a sorted list of all structures. Typical results are “dome” outputs (see Figure 27) or many RNAs.
RNA structure viewing programs (RNAStructure, RNAV
iz) can be found either as a "connect" or "ct" file that can be used in one of them. A typical example of such a structure diagram is shown in FIG.

Example 3: Histone The histone 3'UTR represents another typical stem-loop structure that has been extensively studied (EMBO, 1997, 16, 769). Post-transcriptional l
evel) show that the stem-loop structure in the 3 'untranslated region of histone mRNA is very important (Son, Saenghwahak Nyusu, 1993, 13, 64-70).
. The analysis described below describes the use of this known structure and confirms the strategies and methods described herein.

FIGS. 29 and 30 show the output of the phylogenetic tree for all histone orthologs in the Hovergen database. Each of these orthologs was saved in GenBank format and grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted and compared using SEALS and COWX as described previously (see Figures 18 and 25).

Following extraction and comparison with SEALS and COWX, potential regions of interest were determined using AlignHits (see FIG. 31). One such area is shown enclosed. Sequences corresponding to the region of interest were extracted from all species for alignment using CLUSTAL W (1.74). Following extraction of sequence information from AlignHits, CLUSTAL W (1.74) was used to provide the indicated multiple sequence alignments (see Figure 32). Each putative hit sequence was analyzed for its ability to form internal structures. This was accomplished by analyzing each sequence in a matrix that plots the sequence from 5 'to 3' on the X axis and its complement from 5 'to 3' on the Y axis. Base pairs along the diagonal indicate potential self-complementary regions that can form secondary structures. FIG. 33 shows a representative inverse complement matrix. FIG. 34 shows a representative sequence in a dome format showing potential stem structures between base pairs. Following the conversion of the dome format file to a ct file, the structure is visualized using RNA Structure 3.21 (see Figure 35).

Example 4 Vimentin Vimentin is an intermediate filament protein whose 3 ′ UTR is highly conserved among species. Previous work by Zehner et al. (Nuc. Acids Res., 1997, 25, 3362-3370) indicated that the proposed complex stem-loop structure contained within this region was mRNA
It has been shown that placement can be important for vimentin mRNA function.
This region was identified using this analysis, confirming the approach. In addition, based on the analysis described herein, a second stem-loop structure has been identified that occurs downstream of a previously proposed structure that may also have a role in regulating vimentin function (see FIG. 36).

A representative phylogenetic tree output for all vimentin orthologs in the Hovergen database is shown in FIG. Each of these orthologs was saved in GenBank format and grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted and compared using SEALS and COWX as described previously (see Figures 18 and 25).

Following extraction and comparison with SEALS and COWX, potential regions of interest were determined using Align Hits. Two such regions appeared and were used for subsequent analysis (see FIG. 38). Following extraction of sequence information from Align Hits for Region 1, CLUSTAL W was used to provide the indicated multiple sequence alignment (Figure 3).
9). The potential stem structure between base pairs given on the sequence alignment in the dome format is shown in FIG. Following the conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 (see Figure 41). This structure is very similar to that proposed by Zehner et al. (See Figure 42).
. Zehner et al. Presented a detailed chemical analysis of the proposed structure for the smallest binding domain in the 3 'UTR of vimentin. This analysis is single-strand specific (ChS or T
This included cleavage with 1) or duplex-specific (V1) nucleases, as well as acetate induction after exposure.

Following extraction of sequence information from Align Hits for region 2, CLUSTAL W was used to provide the multiple sequence alignment shown in FIG. The potential stem structure between base pairs in region 2 is given on the sequence alignment in dome format (see FIG. 44). Following the conversion of the dome format file to a ct file, the structure for region 2 was visualized using RNA Structure 3.21 (see Figure 36).

Example 5 Transferrin Receptor As with the control of ferritin (Examples 1 and 2), another known function of IRE is in the control of transferrin receptor. Five IREs have been identified in the 3 'UTR of a known transfer receptor mRNA (Kuhn et al., EMBO J., 1987, 6,
1287-93 and Casey et al., Science, 1988, 240,924-928), each of which is hereby incorporated by reference in its entirety. All five IREs are iron regulatory proteins (IRP
) Have been shown to interact independently. This technique was applied to the identification of these conserved elements in the transferrin receptor.

The output of a representative phylogenetic tree for all transferrin receptor orthologs in the Hovergen database is shown in FIG. Each of these orthologs was saved in GenBank format and grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted and compared using SEALS and COWX as described previously (see Figures 18 and 25).

Following extraction and comparison with SEALS and COWX, FIG.
Potential target areas were determined as shown in This allows one to see where the vertical lines cross a series of horizontal lines representing sequence information from a set of species. This region between 920 and 990 base pairs in the 3-prime UTR of the transferrin receptor was extracted from all species for alignment using CLUSTAL W (1.74).

Following extraction of sequence information from Align Hits for region 1, CLUSTAL W (1.74
) Was used to provide the multiple sequence alignment shown in FIG. Representative potential stem structures between base pairs are given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 (see Figure 49). This allows one to see where the vertical lines cross a series of horizontal lines representing sequence information from a set of species. 99 in the 3-prime UTR of the transferrin receptor
This region between 0 and 1050 base pairs was extracted from all species for alignment using CLUSTAL W (1.74) (see FIG. 50).

Following extraction of sequence information from Align Hits for Region 2, CLUSTAL W (1.74
) Was used to provide the multiple sequence alignment shown in FIG. Potential stem structures between base pairs are given on the sequence alignment in dome format as shown in FIG. Following conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 as shown in FIG. Following extraction and comparison by SEALS and COWX, Align Hits were used to determine potential regions of interest. This allows one to see where the vertical lines cross a series of horizontal lines representing sequence information from a set of species. This region between 1372 and 1423 base pairs in the 3-prime UTR of the transferrin receptor is designated CLUSTAL W (1.
Extracted from all species for alignment using (74) (see FIG. 54).

Following extraction of sequence information from Align Hits for region 3, CLUSTAL W (1.Ex.
34) was used to provide the multiple sequence alignment shown in FIG. Potential stem structures between base pairs are given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 as shown in FIG. SEALS
Align Hits were used to determine potential regions of interest following extraction and comparison with COWX. This allows one to see where the vertical lines cross a series of horizontal lines representing sequence information from a set of species. Transferrin receptor 3
This region between 1439 and 1479 base pairs in the prime UTR is designated CLUSTAL W (
1.74) was extracted from all species for alignment using (see FIG. 58).

Following extraction of sequence information from Align Hits for region 4, CLUSTAL W (1.Ex.
34) was used to provide the multiple sequence alignment shown in FIG. The potential stem structure between base pairs is given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 as shown in FIG. SEALS
Align Hits were used to determine potential regions of interest following extraction and comparison with COWX. This allows one to see where the vertical lines cross a series of horizontal lines representing sequence information from a set of species. Transferrin receptor 3
This region between 1479 and 1542 base pairs in the prime UTR is designated CLUSTAL W (
1.74) were extracted from all species for alignment using (see FIG. 62).

Following extraction of sequence information from Align Hits for region 5, CLUSTAL W (1.Ex.
34) was used to provide the multiple sequence alignment shown in FIG. Potential stem structures between base pairs are given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, the structure was visualized using RNA Structure 3.21 as shown in FIG.

Example 6 Ornithine Decarboxylase Ornithine decarboxylase (ODC) is the first enzyme in the polyamine biosynthetic pathway. Studies have shown the presence of translational regulatory elements in both the 5 'and 3' untranslated regions (Grens et al.,
J. Biol. Chem., 1990, 265, 11810). It has been proposed that secondary structure exists in both of these regions, but there is no definitive evidence for it. The method described here identified two structures in the 3 'UTR as shown below. The presence of these structures (see FIG. 66) was verified using mass spectrometry probing (Griffey et al., Proc. SPIE-Int. Soc. Opt. Eng., 2985 (Ultrasensitive production). Chemical diagnosis II
(Ultrasensitive Biochemical Diagnostics II)): 82-86, which is hereby incorporated by reference in its entirety). Two representative sequences showing slight changes in length (see FIG. 67) were made in the RNA and subjected to MS structural investigation. Figure 6
The results shown in 6 confirm the presence of the stem-loop structure. Therefore, the identification of new secondary structures can be identified from the methods described herein, and such presence has been independently verified by structural investigations.

The output of the phylogenetic tree for all ornithine decarboxylase orthologs in the Hovergen database is shown in FIGS. 68 and 69. GenBan each of these orthologs
Saved in k format and grouped together in a single data file. As described previously for untranslated regions on both the 5 'and 3' sides of the coding region, SEAL
Extracted and compared using S and COWX (see FIGS. 18 and 25).

Following extraction and comparison with SEALS and COWX, using Align Hits, FIG.
Potential target areas were determined as shown in Two such regions appeared and they were used for subsequent analysis. Following extraction of sequence information from region 1, CLUSTAL
W (1.74) was used to provide the indicated multiple sequence alignment. Each putative mid-sequence was analyzed for its ability to form internal structures as shown in the reverse complementation matrix depicted in FIG. This was accomplished by analyzing each sequence in a matrix that plots the sequence from 5 'to 3' on the X axis and its complement from 5 'to 3' on the Y axis. Base pairs along the diagonal indicate potential self-complementary regions that can form secondary structures. A dome-like overview of potential stem structures between base pairs in region 1 is given on the sequence alignment determined using RevComp (see FIG. 72). The structure was visualized using RNA Structure 3.21 (see Figure 73).

The structure was investigated using mass spectrometry analysis techniques. FIG. 67 showed the presence of gaps / insertions in the multiplex alignment. Two representative RNAs from the alignment shown in FIG. 67 (gi404561 and gi35135) were used in this experiment. Pattern analysis of the induced fragmentation revealed that the stem half of the stem-loop structure (indicated in the figure) (t
It showed very strong prospects for base pairs along op half). This is 404561
Correspond to bases 11-14 and 20-23 or 35135 at bases 8-11 and 18-21. The base at the bend (G9 at 404561 or U22 at 35135) also showed a characteristic fragmentation pattern. The bottom-half of the structure appeared to be less stable, with some fragmentation at the base pairs predicted by our analysis.
This was especially true for sequence 35135. However, this region has several adjacent AU or GU base pairs that tend to be less stable and therefore have a higher fragmentation potential.

Following extraction of sequence information from Align Hits for region 2, CLUSTAL W was used to provide a multiple sequence alignment as shown in FIG. The potential stem structure between base pairs in region 2 is given on the sequence alignment in the dome format as shown in FIG. Following the conversion of the dome format file to a ct file, RNA Structure 3.21 was used to visualize the structure for region 2 as shown in FIG.

Example 7: Interleukin-2 (IL-2) A representative phylogenetic tree output for all IL-2 orthologs in the Hovergen database is shown in FIG. Save each of these orthologs in GenBank format,
Grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted and compared using SEALS and COWX as described previously (see Figures 18 and 25).

Following extraction and comparison with SEALS and COWX, 3 ′ UT using Align Hits
The potential target area in R area was determined. Two such regions appeared and they were used for subsequent analysis (see Figure 78). Following extraction of sequence information from Align Hits for region 1, CLUSTAL W (1.74) was used to provide the multiple sequence alignment shown in FIG. A dome outline of potential stem structures between base pairs in region 1 is given on the sequence alignment determined using RevComp (see FIG. 80). The structure was visualized using RNA Structure 3.2 as depicted in FIG. 81 (see FIG. 73). Following extraction of sequence information from Align Hits for region 2, CLUSTAL W (1.74) was used to provide the multiple sequence alignment shown in FIG. The potential stem structure between base pairs in region 2 is given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, RNA Structure 3.21 was used to visualize the structure for region 2 as shown in FIG.

In addition to the two regions described above, a third region that partially overlaps downstream of region 2 was identified using an alternative reference sequence (3087784.fa) and is shown in FIG. .
Following extraction of sequence information from Align Hits for this region, CLUSTAL W (1.74
) Was used to provide the multiple sequence alignment shown in FIG. The potential stem structure between base pairs in region 3 is shown in FIG. 87 on a sequence alignment in the dome format. Following the conversion of the dome format file to a ct file,
The structure for region 3 was clarified using RNA Structure 3.21 (see FIG. 88).

Example 8: Interleukin-4 (IL-4) A representative phylogenetic tree output for all IL-4 orthologs in the Hovergen database is shown in FIG. Save each of these orthologs in GenBank format,
Grouped together in a single data file. Untranslated regions on both the 5 'and 3' sides of the coding region were extracted and compared using SEALS and COWX as described previously (see Figures 18 and 25).

Following extraction and comparison with SEALS and COWX, using Align Hits, FIG.
The potential region of interest in the 5 'UTR region was determined as shown in Following extraction of sequence information from Align Hits for the above regions, CLUSTAL W (1.74) was used to provide the multiple sequence alignment shown in FIG. A dome-like overview of potential stem structures between base pairs in the region is given on the sequence alignment determined using RevComp (see FIG. 92). The structure was visualized using RNA Structure 3.2 as shown in FIG.

FIG. 94 depicts a representative Align Hits view of hits in the 3 ′ UTR region of IL-4. 3 '
Following extraction of sequence information from Align Hits for the UTR region, CLUSTAL W (1.74
) Was used to provide a multiple sequence alignment as shown in FIG. Region 2
The potential stem structure between base pairs in is given on the sequence alignment in dome format as shown in FIG. Following the conversion of the dome format file to a ct file, the structure for region 2 was visualized using RNA Structure 3.21 (see Figure 97).

[Brief description of the drawings]

FIG. 1 illustrates a flowchart comprising one preferred set of steps of a method for identifying molecular interaction sites in eukaryotic and prokaryotic RNA.

FIG. 2 is a flowchart describing a preferred set of procedures in the Find Neighbors And Assemble ESTBlast protocol.

FIG. 3 is a flowchart describing the preferred steps in the BlastParse protocol.

FIG. 4 is a flowchart describing the preferred steps in the QC protocol.

FIGS. 5A, 5B, 5C and 5D illustrate flow charts describing preferred steps in the CompareOverWins protocol.

FIG. 6 is a representative scatter plot of a mouse and human interspecies sequence comparison for ferritin RNA.

FIG. 7 shows self complementation analysis of a single sequence.
) Is shown.

FIG. 8 shows a self-complementar plot of an ortholog.
diagonal strings of blocks (diago), showing the overlay for each of the city plots, and the choice for the most repeated pattern in each.
nal strings), resulting in the smallest possible number of possible folds.

FIG. 9 shows an exemplary descriptor.

FIG. 10 shows a set of e-volume scores for ferritin.

FIG. 11 is a representative diffusion plot of a human and trout interspecies sequence comparison for ferritin RNA.

FIG. 12 is a representative diffusion plot of a human and chicken interspecies sequence comparison for ferritin RNA.

FIG. 13 is an exemplary look-up table used in Q-Compare or CompareOverWins.

FIG. 14 is a typical block diagram of a program called RevComp.

FIG. 15 shows an exemplary flowchart illustrating the preferred steps of a preferred database search strategy for ortholog discovery.

FIG. 16 shows a representative Hovergen phylogenetic tree for ferritin species classification.

FIG. 17 shows a representative Hovergen phylogenetic tree for ferritin mammalian hierarchical classification.

FIG. 18 shows an exemplary flow scheme showing preferred steps for a preferred SEALS strategy.

FIG. 19 shows a representative plot showing regions of homologous sequence between the human and mouse ferritin 5 ′ UTRs.

FIG. 20 depicts a genetic map showing conserved iron-responsive elements in the 5 ′ UTR of mouse and human ferritin.

FIG. 21 shows a representative plot showing regions of homologous sequence between the human and trout ferritin 5 ′ UTRs.

FIG. 22 shows a representative plot showing regions of homologous sequence between human and chicken ferritin 5 ′ UTRs.

FIG. 23 shows a representative Align Hits view of ferritin 5 ′ UTR.

FIG. 24 shows a representative Clustal Alignment of ferritin 5 ′ UTR.

FIG. 25 shows an exemplary flow chart illustrating preferred steps for a preferred Structure Predictor strategy.

FIG. 26 shows a representative reverse complement matrix for the ferritin 5 ′ UTR (
reverse complement matrix).

FIG. 27 shows an outline of a typical dome-shaped structure of a ferritin 5 ′ UTR structure.

FIG. 28 shows a representative structural diagram of ferritin 5 ′ UTR.

FIG. 29 shows a representative Hovergen phylogenetic tree for histones.

FIG. 30. Representative Hovergen showing vertebrate classification for histones.
The phylogenetic tree is shown.

FIG. 31 shows a representative Align Hits view of the histone 3 ′ UTR.

FIG. 32 shows a representative Clustal Alignment for the histone 3 ′ UTR.

FIG. 33 shows a representative inverse complementation matrix for the histone 3 ′ UTR.

FIG. 34 shows a typical dome-shaped schematic structure of a histone 3 ′ UTR.

FIG. 35 shows a representative structural diagram for the histone 3 ′ UTR.

FIG. 36 shows a representative structural diagram of region 2 of the vimentin 3 ′ UTR.

FIG. 37 shows a representative Hovergen phylogenetic tree for vimentin.

FIG. 38 shows a representative Align Hits summary of the vimentin 3 ′ UTR.

FIG. 39 shows a representative Clustal Alignment of region 1 of the vimentin 3 ′ UTR.
Is shown.

FIG. 40 shows a schematic representation of a representative dome-like structure of region 1 of the vimentin 3 ′ UTR.

FIG. 41 shows a representative structural diagram of region 1 of the vimentin 3 ′ UTR.

FIG. 42 shows the structure proposed by Zehner et al. For the vimentin 3 ′ UTR.

FIG. 43 shows a representative Clustal Alignment of region 2 of the vimentin 3 ′ UTR.
Is shown.

FIG. 44 shows a schematic representation of a representative dome-like structure of region 2 of the vimentin 3 ′ UTR.

FIG. 45 shows a representative Hovergen phylogenetic tree of transferrin receptors.

FIG. 46 shows representative A of region 1 of the transferrin receptor 3 ′ UTR.
Here is an overview of lign Hits.

FIG. 47 shows representative C of region 1 of the transferrin receptor 3 ′ UTR.
Indicates lustal alignment.

FIG. 48 shows a schematic representation of a representative dome-like structure of region 1 of the transferrin receptor 3 ′ UTR.

FIG. 49 shows a representative structural diagram of region 1 of the transferrin receptor 3 ′ UTR.

FIG. 50 shows a representative A of region 2 of the transferrin receptor 3 ′ UTR.
Here is an overview of lign Hits.

FIG. 51 shows a representative C of region 2 of the transferrin receptor 3 ′ UTR.
Indicates lustal alignment.

FIG. 52 shows a schematic representation of a representative dome-like structure of region 2 of the transferrin receptor 3 ′ UTR.

FIG. 53 shows a representative structural diagram of region 2 of the transferrin receptor 3 ′ UTR.

FIG. 54 shows a representative A of region 3 of the transferrin receptor 3 ′ UTR.
Here is an overview of lign Hits.

FIG. 55 shows a representative C of region 3 of the transferrin receptor 3 ′ UTR.
Indicates lustal alignment.

FIG. 56 shows a schematic representation of a representative dome-like structure of region 3 of the transferrin receptor 3 ′ UTR.

FIG. 57 shows a representative structural diagram of region 3 of the transferrin receptor 3 ′ UTR.

FIG. 58. Representative A of region 4 of the transferrin receptor 3 ′ UTR.
Here is an overview of lign Hits.

FIG. 59 shows a representative C of region 4 of the transferrin receptor 3 ′ UTR.
Indicates lustal alignment.

FIG. 60 shows a schematic representation of a representative dome-like structure of region 4 of the transferrin receptor 3 ′ UTR.

FIG. 61 shows a representative structural diagram of region 4 of the transferrin receptor 3 ′ UTR.

FIG. 62 shows a representative A of region 5 of the transferrin receptor 3 ′ UTR.
Here is an overview of lign Hits.

FIG. 63 shows a representative C of region 5 of the transferrin receptor 3 ′ UTR.
Indicates lustal alignment.

FIG. 64 shows a schematic representation of a representative dome-like structure of region 5 of the transferrin receptor 3 ′ UTR.

FIG. 65 shows a representative structural diagram of region 5 of the transferrin receptor 3 ′ UTR.

FIG. 66 shows a representative mass-spec structural probe analysis of region 1 of the ornithine decarboxylase 3 ′ UTR.

FIG. 67 shows a representative Clustal Alignment of region 1 of ornithine decarboxylase 3 ′ UTR.

FIG. 68 shows a representative Hoverg of ornithine decarboxylase 3 ′ UTR.
The en tree is shown.

FIG. 69 shows a representative Hovergen phylogenetic tree of vertebrate ornithine decarboxylase 3 ′ UTRs.

FIG. 70 shows a representative Align of ornithine decarboxylase 3 ′ UTR.
Here is an overview of Hits.

FIG. 71 shows a representative reverse complementation matrix of region 1 of the ornithine decarboxylase 3 ′ UTR.

FIG. 72 shows a schematic representation of a representative dome-like structure of region 1 of the ornithine decarboxylase 3 ′ UTR.

FIG. 73 shows a representative structural diagram of region 1 of the ornithine decarboxylase 3 ′ UTR.

FIG. 74 shows a representative Clustal Alignment of region 2 of ornithine decarboxylase 3 ′ UTR.

FIG. 75 shows a schematic representation of a representative dome-like structure of region 2 of ornithine decarboxylase 3 ′ UTR.

FIG. 76 shows a representative structural diagram of region 2 of the ornithine decarboxylase 3 ′ UTR.

FIG. 77 shows a representative Hovergen phylogenetic tree for interleukin-2 (IL-2).

FIG. 78 shows an outline of typical Align Hits of IL-2 3 ′ UTR.

FIG. 79 shows a representative Clustal Alignment of region 1 of the IL-2 3 ′ UTR.

FIG. 80 shows a schematic representation of a representative dome-like structure in Region 1 of the IL-2 3 ′ UTR.

FIG. 81 shows a representative structural diagram of region 1 of the IL-2 3 ′ UTR.

FIG. 82 shows a representative Clustal Alignment of region 2 of the IL-2 3 ′ UTR.

FIG. 83 shows a schematic representation of a representative dome-like structure of region 2 of the IL-2 3 ′ UTR.

FIG. 84 shows a representative structural diagram of region 2 of the IL-2 3 ′ UTR.

FIG. 85 shows an overview of typical Align Hits of the IL-2 3 ′ UTR.

FIG. 86 shows a representative Clustal Alignment of region 3 of the IL-2 3 ′ UTR.

FIG. 87 shows a schematic representation of a representative dome-like structure of region 3 of the IL-2 3 ′ UTR.

FIG. 88 shows a representative structural diagram of region 3 of the IL-2 3 ′ UTR.

FIG. 89 shows a representative Hovergen phylogenetic tree of interleukin-4 (IL-4).

FIG. 90 shows an overview of typical Align Hits of the IL-4 5 ′ UTR.

FIG. 91 shows a representative Clustal Alignment of the IL-4 5 ′ UTR.

FIG. 92 shows an outline of a typical dome-shaped structure of the IL-4 5 ′ UTR.

FIG. 93 shows a representative structural diagram of the IL-4 5 ′ UTR.

FIG. 94 shows an overview of representative Align Hits of the IL-4 3 ′ UTR.

FIG. 95 shows a representative Clustal Alignment of the IL-4 3 ′ UTR.

FIG. 96 shows an outline of a typical dome-shaped structure of the IL-4 3 ′ UTR.

FIG. 97 shows a representative structural diagram of the IL-4 3 ′ UTR.

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SL, SZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR , BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS , JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, US, UZ, VN, YU, ZA, ZW (72) Inventor Sampath, Ranga 92129 California, USA Griffin, Richard Inventor Griffin, Richard 92084, California, United States 92084; Vista, Barsby Street 360 (72) Inventor McNail, John 92037, California, United States, La Jolla, Retaheim Way 427 F-term ( Reference) 4B024 AA01 AA03 AA07 AA11 AA12 CA01 CA11 CA12 DA05 DA11 DA12 HA11 4B063 QA13 QA18 QA19 QQ05 QQ06 QQ07 QQ10 QQ41 QQ42 QQ52 QQ53 QQ54 QR31 4C084 AA13 NA14 ZA011 ZA081 ZA361 ZB111 ZB212 ZB261 ZB332 ZB352 ZB372 ZC011 ZC412 ZC521 4C086 AA01 AA03 EA16 MA01 MA02 MA03 MA04 MA05 NA14 ZA01 ZA08 ZA36 ZB11 ZB21 ZB26 ZB33 ZB35 ZB33 ZB35 ZB35 ZB35 ZB35 ZB35 ZB35 ZB35 ZB33 ZB35 ZB33 ZB35 ZB33 ZB35 ZB33 ZB35 ZB35 ZB33

Claims (34)

[Claims] 1. A method for identifying a molecular interaction site in a target nucleic acid, comprising the steps of: comparing the nucleotide sequence of the target nucleic acid with the nucleotide sequences of a plurality of nucleic acids from various taxonomic species. Identifying at least one sequence region conserved between the plurality of nucleic acids and the target nucleic acid; determining whether the conserved region has a secondary structure; Identifying the secondary structure for the conserved region having: 2. The method of claim 1, further comprising identifying at least one structural motif for the conserved region having a secondary structure. 3. The method of claim 2, further comprising constructing a set of descriptor elements for the structural motif. 4. The method of claim 3, further comprising further identifying a nucleic acid having a secondary structure corresponding to said descriptor element. 5. The method of claim 1, wherein said target nucleic acid is present in a eukaryotic cell. 6. The target nucleic acid may be mRNA, pre-mRNA, tRNA, rRNA and snRNA.
The method of claim 5, wherein the method is selected from the group consisting of: 7. The method of claim 1, wherein said target nucleic acid is present in a prokaryotic cell. 8. The method according to claim 7, wherein said target nucleic acid is RNA. 9. The method of claim 7, wherein said target nucleic acid is bacterial. 10. The method of claim 7, wherein said target nucleic acid is of a virus. 11. The method of claim 7, wherein said target nucleic acid is derived from a parasite. 12. The method of claim 1, wherein at least some of the nucleic acid sequence information is from a gene database. 13. The method of claim 1, wherein said nucleotide sequence of said target nucleic acid is determined by assembling a plurality of expressed sequence tags. 14. The method of claim 1, further comprising comparing the target nucleic acid with a paralog nucleic acid. 15. The method of claim 1, wherein the plurality of nucleic acids from different taxonomic species are obtained by performing a sequence identity search, an ortholog search, or a combination thereof. 16. The method of claim 1, wherein said plurality of nucleic acids from different taxonomic species is obtained by performing a sequence identity search and constructing a virtual transcript. 17. The method of claim 1, wherein determining whether the conserved region has a secondary structure is performed by self-complementary comparison, juxtaposition and covariance analysis, secondary structure prediction, or a combination thereof. The described method. 18. The method of claim 18, wherein the secondary structure comprises at least one bulge, loop, stem, hairpin, knot, triple interact,
18. The method of claim 17, comprising cloverleaf or helix. 19. The method of claim 2, wherein said structural motif is identified by performing a self-complementary comparison, juxtaposition and covariance analysis, secondary structure prediction, or a combination thereof. 20. The method of claim 3, wherein the set of descriptor elements is constructed utilizing a descriptor database. 21. The other nucleic acid having a secondary structure corresponding to the descriptor element is identified by searching at least one database, performing clustering and analysis, searching for orthologs, or a combination thereof. Claim 4
The method described in. 22. A database comprising a molecular interaction site identified by the method of claim 1. 23. The database of claim 22, which contains a eukaryotic cell molecule interaction site. 24. The database according to claim 23, comprising a human molecule interaction site. 25. The database of claim 22, which contains prokaryotic molecule interaction sites. 26. An oligonucleotide comprising a molecular interaction site present in the RNA of the selected organism and in the RNA of at least one additional organism, wherein the molecular interaction site is The oligonucleotide, wherein the oligonucleotide functions as a binding site for at least one molecule that, when bound, regulates expression of the RNA in the selected organism. 27. The prokaryotic RNA and at least one additional prokaryotic RNA
An oligonucleotide comprising a molecular interaction site present therein, wherein said molecular interaction site binds to at least one molecule such that when bound to said interaction site, said molecule regulates expression of said prokaryotic RNA. The oligonucleotide, which functions as a. 28. The oligonucleotide of claim 27, wherein said molecular interaction site is absent from eukaryotic cell RNA. 29. The oligonucleotide of claim 27, wherein said molecular interaction site is not present in human RNA. 30. An oligonucleotide comprising a molecular interaction site present in prokaryotic RNA and in at least one additional prokaryotic RNA, wherein said molecular interaction site is said molecular interaction site A pharmaceutical composition comprising: said oligonucleotide, which functions as a binding site for at least one molecule that, when bound to, regulates expression of said prokaryotic RNA, and a pharmaceutical carrier or diluent. 31. The molecular interaction site is not present in eukaryotic RNA.
A pharmaceutical composition according to claim 30. 32. The pharmaceutical composition according to claim 30, wherein said molecular interaction site is not present in human RNA. 33. An oligonucleotide comprising a molecular interaction site present in the RNA of the selected organism and in at least one additional organism's RNA, wherein the molecular interaction site is the molecule. Said oligonucleotide functioning as a binding site for at least one molecule that, when bound to an interaction site, regulates the expression of said RNA in said selected organism; and a pharmaceutical carrier or diluent. , Pharmaceutical compositions. 34. A pharmaceutical composition comprising: an oligonucleotide comprising a molecular interaction site present in prokaryotic RNA but not present in mammalian RNA; and a pharmaceutical carrier or diluent.