[go: up one dir, main page]

WO2025072554A1 - Systèmes et procédés de commande d'expérience biologique dynamique et en temps réel - Google Patents

Systèmes et procédés de commande d'expérience biologique dynamique et en temps réel Download PDF

Info

Publication number
WO2025072554A1
WO2025072554A1 PCT/US2024/048706 US2024048706W WO2025072554A1 WO 2025072554 A1 WO2025072554 A1 WO 2025072554A1 US 2024048706 W US2024048706 W US 2024048706W WO 2025072554 A1 WO2025072554 A1 WO 2025072554A1
Authority
WO
WIPO (PCT)
Prior art keywords
automation system
experiment
automation
commands
samples
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2024/048706
Other languages
English (en)
Inventor
Robert Damoiseaux
Ronan BENNETT
Alejandro HUERTA
Michael MELLODY
Rutu SHAH
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of California Berkeley
University of California San Diego UCSD
Original Assignee
University of California Berkeley
University of California San Diego UCSD
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of California Berkeley, University of California San Diego UCSD filed Critical University of California Berkeley
Publication of WO2025072554A1 publication Critical patent/WO2025072554A1/fr
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/22—Social work or social welfare, e.g. community support activities or counselling services
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N35/00—Automatic analysis not limited to methods or materials provided for in any single one of groups G01N1/00 - G01N33/00; Handling materials therefor
    • G01N35/00584—Control arrangements for automatic analysers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/103—Workflow collaboration or project management
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/40—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for data related to laboratory analysis, e.g. patient specimen analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B45/00—ICT specially adapted for bioinformatics-related data visualisation, e.g. displaying of maps or networks
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B50/00—ICT programming tools or database systems specially adapted for bioinformatics

Definitions

  • the present invention relates to control of automated laboratory equipment conducting biological experiments and more specifically to real-time control using an object-oriented programming language.
  • GUIs graphical user interfaces
  • Missing features are fine control on a per plate or per well level that is paired with the ability to change of experimental parameters on the fly to enable interventions by the experimenter while the experiment or experimental campaign is executing.
  • current automation systems do not lend themselves to the substitution of small-scale work by automation as it is hard to make customization on experiments - especially after the actual experimental run on the system has started. The changing of conditions in response to experimental data as they occur - as normally performed during manual experimentation - necessitates a re-do of experiments on automated large-scale systems and also manual labor during disposal and resetup.
  • a laboratory equipment automation system for biological experiments includes one or more pieces of automated laboratory equipment, an interpreter device configured to receive a first user input that describes a biological experiment to perform using automated laboratory equipment and generate from the user input an experiment schedule including automation commands and timing information that describes when the automation commands should be performed, where each automation command is determined based upon biological meaning interpreted from the user input and is addressed to a corresponding automated laboratory equipment of the one or more pieces of automated laboratory equipment to operate on one or more sample holders containing samples, and an execution device configured to conduct the biological experiment by reading the experiment schedule and directing the automation commands to the corresponding automated laboratory equipment based upon the timing information in performance of one or more workflows, and wherein the interpreter device is also configured to receive a second user input that describes at least one biological command that modifies the biological experiment, update the experiment schedule in real-time while the execution device is conducting the biological experiment by adapting the automation commands and timing information to accommodate the described biological command modifications of the second user input, direct the automated laboratory equipment with modified operations on one or more active
  • a method of automating laboratory equipment for biological experiments includes receiving, by an interpreter device, a first user input that describes a biological experiment to perform using one or more pieces of automated laboratory equipment, generating, using the interpreter device, from the user input an experiment schedule including automation commands and timing information that describes when the automation commands should be performed, where each automation command is determined based upon biological meaning interpreted from the user input and is addressed to a corresponding automated laboratory equipment of the one or more pieces of automated laboratory equipment to operate on one or more sample holders containing samples, conducting the biological experiment, using an execution device, by reading the experiment schedule and directing the automation commands to the corresponding automated laboratory equipment based upon the timing information in performance of one or more workflows, receiving, by the interpreter device, a second user input that describes at least one biological command that modifies the biological experiment, updating, using the interpreter device, the experiment schedule in real-time while the execution device is conducting the biological experiment by adapting the automation commands and timing information to accommodate the described biological command modifications of the second user input, and directing
  • the interpreter device is configured to receive the first user input as commands on a command line within an integrated development environment (IDE).
  • IDE integrated development environment
  • the interpreter device is configured to receive the first user input as commands on a command line in Python command line as a direct interpreter.
  • the interpreter device is configured to receive the first user input as spoken audio and the experiment schedule is expressed as an XML (extensible markup language) file, which is imported into a database.
  • the automation commands include a programming technique of iteration loops to process a set of samples with a biologically complex set of workflows.
  • the automation commands include a programming technique of recursion to process a set of samples with a biologically complex set of workflows.
  • the automation commands include an application of a mathematical algorithm of Newton’s iteration to process a set of samples with a biologically complex set of workflows and feed-forward the output to drive biological systems to a desired goal.
  • results from the automated laboratory equipment are captured are written to a database and the results are used to adjust the timing information in processing a set of samples.
  • the execution device is configured to provide access to a database storing the experiment schedule to a user device and to adjust timing information when user input is captured on the use device that changes some aspect of the experiment schedule.
  • the adjusted timing information affects a specific sample holder.
  • the adjusted timing information affects a specific sample within a sample holder.
  • the execution device implements a daemon process to iterate over the experiment schedule while it is stored in a database.
  • the execution device iterates over the experiment schedule while it is stored in a database and readjusts the experiment schedule to avoid conflicts while processing samples using commands that encapsulate biologically complex workflows in a single function call.
  • the execution device iterates over the experiment schedule containing sample processing information and readjusts the experiment schedule to avoid conflicts while processing samples using commands that encapsulate biologically complex workflows in a single function call.
  • the automation commands are implemented as function calls in an object-oriented programming language such that each of the automated laboratory equipment is instantiated as a class having function calls that access instrument actions of the automated laboratory equipment.
  • the execution device is configured to collect data from the automated laboratory equipment.
  • the automation commands are implemented in an object-oriented programming language that is not pre-compiled and has an audit trail generated as it executes for enforcement by security policies.
  • the execution device is configured to transfer image files from a microscope to an image analysis computing device using function calls in an object-oriented programming language where a class, which contains the function calls, is instantiated for the microscope and interactions with the microscope.
  • a monitoring device is configured to perform image analysis on an image stream of a camera overlooking a surrounding area of the automation system to recognize human presence, to adjust speed of a robotic system to a collaborative speed when human presence is found, to adjust speed of the robotic system to full speed when human presence is not found, and to adjust the timing information based on any speed adjustment of the robotic system.
  • an imaging system with a camera configured to recognize gestures of a user for commands to start, stop, and pause execution of the biological experiment.
  • an imaging system with a camera configured to capture images of samples and transfer images to the execution device, and wherein the execution device is configured to evaluate the images to determine proper orientation of the samples for correct processing of the samples.
  • the execution device is configured to resolve timing and programming errors and deadlock situations by readjusting the timing information and updating the database.
  • the execution device is configured to establish real-time communication to a mobile device to obtain user input for resolving an error situation.
  • the execution device is configured to establish real-time communication to a large language model (LLM) for resolving an error situation.
  • LLM large language model
  • the execution device is configured to establish real-time communication to a large language model (LLM) for substitution of a non-identical equipment that has a similar ability to resolve error situations when a piece of laboratory equipment malfunctions.
  • LLM large language model
  • timing information is expressed as start and end times.
  • At least some of the timing information is expressed as frequency of occurrence.
  • the updates to the experiment schedule reassigns at least one sample to a different workflow while processing other samples.
  • the updates to the experiment schedule assigns a different timing to at least one plate within a set of plates to be processed of the biological experiment.
  • the interpreter device is configured to update the experiment schedule based on using a large language model to process the second user input.
  • the interpreter device is configured to receive the second user input as commands on a command line within an integrated development environment (IDE).
  • IDE integrated development environment
  • the second user input comprises changing parameters of the experiment schedule remotely by a user device.
  • the updated experiment schedule changes a workflow on samples that belong to a group of samples that is already being operated on by a laboratory equipment.
  • the updated experiment schedule changes parameters of a sample within a group of samples that is already being operated on by a laboratory equipment.
  • Fig. 1 illustrates a system for automating biological experiments in accordance with several embodiments of the invention.
  • Fig. 2 illustrates a user device in accordance with several embodiments of the invention.
  • Fig. 3 illustrates an interpreter device in accordance with several embodiments of the invention.
  • FIG. 4 illustrates an execution device in accordance with several embodiments of the invention.
  • Fig. 5 illustrates a process for automating biological experiments in accordance with several embodiments of the invention.
  • Fig. 6 illustrates a process for issuing automation commands to laboratory equipment in accordance with several embodiments of the invention.
  • Many embodiments of the invention provide a command line interpreter with wrappers that enable execution of high level experimental/biological requests in a command line format and translate them without further intervention by the experimenter into high level automation commands which are then in turn easily digested down into actionable commands, i.e., timely instructions for the automation equipment.
  • data are collected results for feedback and allows the experimenter to extract, visualize and/or analyze these results and design/optimize further experiments from the command line while the experiment is in progress and make decisions to continue, modify or abandon a run or even parts of a run.
  • the platform described here also removes the requirement for the scientist to be in the laboratory to make the required decisions and/or changes.
  • Embodiments of the invention can take the place of a technician and act in the manner of an operating system or programming language that provides layers of abstraction by translating higher level experimental commands (pertaining to biology) to lower level commands (pertaining to automation of equipment) with additional user input or even an artificial intelligence system that completely takes the place of the technician.
  • the system can iterate over active experiments, surveil labware and generates images, quantify and report back data into an associated relational database (e.g., SQL) that can be queried by the scientist remotely.
  • the database is part of the system and can also hold the proposed schedule and experimental priorities and can be directly adjusted by writing directly into the records for each labware in the database down to the well level of labware while the experiments are ongoing.
  • the database can store all aspects of the experiment and automation work.
  • Experimental workflow can also be directed based on rules set by an experimenter, or artificial intelligence (Al) algorithms can take the place of the scientist further freeing up personnel time.
  • the command line interface can be instrumental for usage by Al systems in that is provides an interface that translates biological instruction into high- and low-level automation instruction without having to train the Al algorithm on each piece of equipment.
  • the system can be built on a programming language such as Python or another high-level programming language which is class based and has packages to enable integration of Al algorithms, automation equipment and a SQL database.
  • the utilization of high-level programming languages also enables the simple isolation of desired sub-populations of plates with specific properties or results, which in turn enables rulesetting for these plates via e.g. numpy or pandas Python packages.
  • Experiments to be conducted can also behave like instances of classes or modules rather than physical entities in that they may be executed and are accessible via simple calls on classes or functions encapsulating biologically meaningful actions into instructions for automation.
  • the biologically meaningful functions can call on an automation layer that implements low level commands that then control the instrument.
  • an instantiated class of experiments can pass information back and forth from the instrument to the database during performance of an experiment in a structured and biologically meaningful manner.
  • biologically meaningful can refer to high level descriptions of a specific experiment type, such as propagation of cells, that typically entails a tedious repetitive but biologically frequently required workflow, such as the changing of media on cells or the introduction of foreign DNA into living cells via a transfection agent, which has many steps that are also repetitive in nature but typically require customization across the samples to include specific attempts at optimization of the conditions. Decisions that involve biological meaning can incorporate knowledge and/or experience of relevant workflows for comprehensive consideration of numerous factors that can affect the outcome. For example, when designing an experiment, parameters such as the number of plates to use and the timing of actions can be chosen in a biologically meaningful way that considers potential interactions, failure rates, and repetitions that may be required.
  • a user or an Al model can be in full control of all details pertaining to execution at all times and can change any and all details of an experiment at any time while the system is live.
  • the ability to update parameters on a live system includes also the updating of criteria used to make decisions on experimental flow.
  • Embodiments of the invention can also provide for greater granularity in control of individual samples. For example, in a batch of plates (e.g., 100), the instrument can treat individual wells differently instead of treating all wells in the batch the same or the ability to apply different sets of parameters to different wells in a plate or set of plates.
  • timing and/or how experiments are to behave can be individualized for each plate and/or each well.
  • Additional features of the invention may include recognizing and fixing errors on the fly.
  • a microtiter plate of 1536 wells may be rotationally C2 symmetric, but intended by a user to be placed in one orientation only so the arrangement of samples within the wells is correct and correspondent to the database content. This is of particular issue if e.g. a plate is barcoded asymmetrically to prevent such issues, but the plate was rotated during the actual barcoding procedure. This is of high significance as e.g. commonly used notches or other marks on the microtiter plate may highlight the orientation but humans are error prone.
  • a camera or other sensor may sense the notch or indicator (e.g., by machine vision) and the system can determine whether the plate is in the proper orientation after loading by a scientist and subsequently correct the orientation by issuing an automation command to turn the plate around to reach correct orientation.
  • Contemplated are also uses of the image recognition that flag dangerous conditions, e.g. a plate that was placed incorrectly on e.g. a liquid handler.
  • laboratory personnel can be monitored if in proximity of the system and can use e.g. hand gestures to start and stop the system or decrease of the speed of a robotic system to ensure collaborative behavior of the system for the time that a human is in its vicinity.
  • the adjustive character of the system allows for re-adjusting the schedule of the experiments to make up for the time lost when running slower.
  • libraries include class definitions where each type or model of laboratory equipment has a defined class and functions defined within the class for each action or operation that the equipment may perform.
  • the class definition may also describe how to access and interact with the equipment (e.g., by an address of a serial interface or an IP address).
  • the class associated with the equipment can be instantiated and its functions called to have the equipment operate on samples according to the workflow.
  • higher level functions can be written (or provided by a library) and used as automation commands that incorporate equipment functions defined within the equipment class and also facilitate the transfer of data to and from a piece of equipment.
  • Import_plate (plate, plate format, cell type, well, density, scanning frequency)
  • Read_plate (source plate, source plate format, timing, wells to be read)
  • Function Flags a plate/wells to be read on a reader and updates the inventory with new results.
  • Function Flags a plate/wells with cells for harvesting of cells and instructs the system to re-plate the cells into another plate at an indicated ratio and make the indicated number of cells with cells for suspension cells.
  • Split_adhesion source plate, wells, target plate, wells source plate format, target plate format, cell type split ratio, number of wells
  • [0077] Function Flags a plate/wells with cells for harvesting of cells by washing with buffer, subsequent release of adherent cells from the plastic of the plate through trypsinization and instructs the system to re-plate the cells into another plate at an indicated ratio and make the indicated number of cells with cells.
  • Function Flags a plate/wells of cells to be transfected by mixing DNA from a source plate with optimem, add transfection reagent, incubate and add to cell target plate
  • [0081] Function Flags a plate/wells with virus producing cells for harvest by moving virus containing supernatant from a source plate and moving it to target plate.
  • Transduce (virus_source plate, wells, cell target plate, wells, source plate format, target plate format, volume of virus to be used, number of wells transduced, transduction enhancer (opt))
  • Function Flags a plate/wells with cells for addition of virus to cells.
  • the import plate function may be called as: Import_plate(plate-1, 384, HeLa, A1-P24, 10, 12)
  • a change plate function may be called as: Change_media(target plate, target plate format, media source plate, source well format, sourcewells, volume)
  • the Change media function may be called as: Change_media(plate-1, 384, reservoir- 1, 1, A1-P24, 50)
  • the system 100 includes an interpreter device 101, an execution device 102, a database 104, a machine learning model computing device 106, one or more pieces of laboratory equipment 108, and may include client devices 110 communicating over a network 112 as illustrated in Fig. 1.
  • the interpreter and execution devices may be on the same platform or computing system. While in other embodiments, the interpreter device and execution device are separate.
  • the database 104 is illustrated as a single entity here, it is understood that data sources and data stores be implemented in many forms, such as distributed systems or cloud services. Databases can include SQL databases such as Microsoft SQL server or Oracle SQL databases, MySQL, or noSQL databases.
  • Data can be moved using any of a variety of available mechanisms, such as using such as Application Programming Interfaces (API).
  • API Application Programming Interfaces
  • the database 104 can communicate with interpreter device 101 and/or execution device 102 directly, while in other embodiments they can communicate over a network such as network 112. Further embodiments of the invention may not utilize a database. Experiment schedules and returned data may be stored in other types of data structures or computing systems.
  • laboratory equipment 108 may communicate with the execution device over a network (e.g., TCP/IP) while embodiments also contemplate laboratory equipment 111 that may communicate directly (e.g., serial connection).
  • Additional embodiments of the invention also include user devices 110, which can be any of a variety of computing devices, such as personal computers, mobile devices or phones, or tablets, that may capture user instructions for conducting an experiment or for making changes to an experiment in ways that will be described further below.
  • a user interface on such devices can also be used for tasks such as to view information, generate reports, and/or send files.
  • an interpreter device 101 may receive user input, e.g., directly through a console or from a user device 110, that describes an experiment to conduct using the automated laboratory equipment.
  • the user input can be converted into an experiment schedule that includes automation commands to be issued to the automated laboratory equipment for performing actions and timing information that describes when the automation commands are to be performed (e.g., by frequency or by start/end time).
  • the experiment schedule may be stored in a database 104.
  • An execution device 102 may read the experiment schedule from the database 104 and issue the automation commands to the automated laboratory equipment based on the timing information.
  • the user device 200 includes a processor 202 and memory 204 that includes an operating system 205, web interface 206 and user interface application 23
  • the user interface application 207 can configure or direct the processor to perform or execute processes such as those described further below.
  • the interpreter device may be a computing system 300 that includes a processor 310 and memory 311 that includes an interpreter application 312.
  • the interpreter application can configure or direct the processor to perform or execute processes such as those described further below to convert a user-provided description of an experiment to automation commands.
  • the integration platform can also access an experiment command database 318 to store automation commands that make up an experiment.
  • an interpreter computing system may be implemented using other computing architectures, for example, as a virtual machine, as a cluster of computers, or using a cloud computing service.
  • the execution device may be a computing system 400 that includes a processor 410 and memory 411 that includes an execution service application 412.
  • the execution service can configure or direct the processor to perform or execute processes such as those described further below to execute automation commands and send commands to laboratory equipment.
  • the execution computing system can also access an experiment command database 418 to retrieve automation commands and other parameters that make up an experiment.
  • an execution computing system may be implemented using other computing architectures, for example, as a virtual machine, as a cluster of computers, or using a cloud computing service.
  • Processes for automating biological experiments in accordance with embodiments of the invention may utilize systems as described above including an interpreter device, an execution device, a database, and at least one piece of laboratory equipment to be controlled.
  • a process in accordance with several embodiments of the invention is illustrated in Fig. 5.
  • the process 500 includes capturing (502) user input that describes a biological or other laboratory experiment to conduct.
  • the user input may be captured, for example, by entry on a computer display screen or by captured audio (e.g., dictated into a microphone and written to an audio fde). If captured by audio, the user input can be transcribed to text by speech-to-text or other types of algorithms that use machine learning.
  • the experiment can be described at different levels of specificity.
  • a user can state that the experiment is given a number of cells of a certain type, and a certain number is desired at a certain time (e.g., given these cells, produce 10 million cells by Wednesday).
  • the user can request that the cells be monitored and split every day at 3 o’clock for the next month.
  • the description of the experiment is provided to a biological and automation interpreter.
  • the interpreter may be implemented as two components for each of its two functions (a biological interpreter and an automation interpreter), or as one component performing both responsibilities.
  • the biological interpreter portion converts (504) the description provided by the user into a list of biological commands (e.g., transfect, split cells, image cells, passaging, etc.), as well as an experiment schedule.
  • the automation interpreter portion converts the list of biological commands (which can be thought of as higher level) into automation commands that can be directly issued to the hardware (e.g., individual pieces of laboratory equipment).
  • an object-oriented programming or scripting language such as Python may be used to construct automation commands and interact with each piece of laboratory equipment.
  • Libraries can include a class for each equipment, with the available actions the equipment is capable of defined as functions within the class that is associated with the equipment.
  • an instance of the associated equipment class can be instantiated for that piece of equipment. It can then be directed to perform actions by calling functions in that instance.
  • the biological and automation interpreter creates (504) a set of experiment instructions as an experiment schedule, i.e., a list of automation commands to be performed and timing information associated with each command.
  • the timing information can be expressed, for example, as a frequency at which to performing an action, a defined start time and end time, etc.
  • the schedule and commands within the schedule can include additional information concerning the laboratory materials (e g., specimens) being used (e g., quantities of plates, metadata (what is in the plate, which wells, what is being done, why is it being done). For example, a command may specify to change media every 6 hours for next 3 days. Another command may take a plate, move the plate to a machine, and execute another action on the plate.
  • each command is addressed to a corresponding piece of laboratory equipment to perform the command.
  • the actions may be performed on one or more sample holders containing samples.
  • the interpreter utilizes biological meaning in constructing the set of commands.
  • the automation commands may be implemented in an object-oriented programming or scripting language such as Python as described further above.
  • Libraries may be utilized in accordance with embodiments of the invention that include classes for each piece of laboratory equipment.
  • the actions that each piece of equipment may perform can be defined as functions within the class.
  • the automation commands can include calls to these functions, as well as directions to instantiate a class for a piece of equipment if it has not already been done and also enable the transfer of data from and to the piece of equipment.
  • the biological and automation interpreter is implemented using a machine learning language model (LM).
  • LM machine learning language model
  • LLM large learning model
  • an interpreter application can perform the biological and automation interpreter duties.
  • the output of the biological and automation interpreter is provided in a standard format (e.g., extensible markup language or XML) as a set of equipment instructions.
  • the output of the interpreter e.g., XML fde of commands and timing information
  • the database is part of another server or machine and the output is sent from the interpreter to the database server.
  • the experiment schedule need not be stored in a database and can be stored in another type of data structure or computing system.
  • An execution service runs (506) to carry out commands from the experiment schedule (e.g., as read from a database) in conducting the experiment.
  • the execution service runs on the same machine as the interpreter.
  • the execution service is on a different machine from the interpreter.
  • the execution service is implemented as a daemon process in Python.
  • a daemon can be implemented as an independent background process that reviews the database for automation commands to be run according to the schedule or a dependent process that automatically terminated with the main process. It may execute the automation commands to control the associated piece of laboratory equipment with the parameters saved to each command when the timing is appropriate according to the schedule and system time of the machine that it is running on.
  • a set of related commands that act upon the same samples can be thought of as a workflow.
  • the execution service may update the schedule so that the command can be performed when the equipment becomes available and any sequence dependency with other commands are adjusted.
  • the database may also keep a record of actions and progress of the experiment (e.g., a plate in incubator shelf 3, position 20 was completed, certain wells were already used on a plate, etc.) as well as associated data such as confluency of cells.
  • a data structure called a worklist can be used.
  • the worklist is a list of entries that, when provided to a piece of equipment, the equipment will operate on each entry. For example, a subset of wells can be listed for being moved to a larger container or a liquid handler can be instructed to ingest what is listed in a worklist.
  • a piece of equipment may provide output information that can be stored back into the database. For example, it may give numerical data such as density in a well, absorption value of bacteria, or time to completion of an action. It may indicate if a particular action succeeded or failed. It may alarm and may provide an image when an alarm goes off.
  • an image may be taken by a microscope or other imager hardware.
  • the image or a link to the image can be stored in the database.
  • the execution service directs a review of the output information.
  • the image may be sent to a user device and shown on a user interface for review by a user.
  • the user may indicate approval or disapproval of the image, or may indicate what next action to take.
  • the new action can be provided back to the equipment for to perform the next action.
  • the image from a microscope may be sent to a machine that runs a TensorFlow or other model to review if the cells are good or not.
  • a message may be sent to a user to request further instruction, e.g., whether to replenish media.
  • an equipment may provide an error message.
  • the execution service may respond to the error message in any of a variety of ways, such as, but not limited to, sending a message to a user to request further instruction, reproducing the plates that were in the machine in another machine, or otherwise seeking a solution to the error.
  • the execution service may direct some pieces of equipment to clean up tasks, e.g., put plates away to a storage location (e.g., incubator), put media into refrigeration, etc.
  • the execution service may direct equipment to hibernate or shut down if there are no further commands.
  • an experiment may be modified (510) by receiving user instructions in real-time while it is being conducted.
  • the user instructions may be expressed as a biological command.
  • a gradient may be changed, which updates the timing of events in the next iteration.
  • a sample can be harvested earlier if target confluence is reached, or a specific well or plates within a run can be terminated.
  • the modifications may be made in any of a variety of ways: through a user interface (e.g., entering in a command line or in a graphical representation of the schedule), through an LLM (e.g., prompt with the changes desired), or by editing values in the database.
  • the user instruction is received by an interpreter device, which updates the experiment schedule in the database in real-time while the biological experiment is being conducted.
  • the experiment schedule can be updated by adapting the automation commands and timing information to accommodate the described biological command modifications of the user instruction while avoiding conflicts.
  • the change may affect some samples in a workflow and not other workflows.
  • the automated laboratory equipment can be directed by the execution device with modified operations on one or more active samples based on the updated experiment schedule, while leaving the remaining samples of the experiment to their original workflow so they are not affected.
  • Fig. 6 illustrates a process for an execution service in accordance with several embodiments of the invention.
  • the process 600 includes checking (602) if a piece of equipment is available that a command should be given to. If the equipment is not available, a similar piece of an equipment that is appropriate for the task to be performed by the command is identified (604). If a class for the equipment is not yet instantiated, it can be instantiated (606) so that its functions may be called.
  • the execution service retrieves instructions from the database and sends (608) one or more commands to the equipment using the class functions. If user input is received to change the experiment as discussed further above with respect to Fig. 5, the experiment instructions and schedule can be updated (610) in the database and the execution service follows the updated instructions.
  • the equipment may collect data in performance of the instructed actions and the data may be written back to the database. As discussed further above with respect to Fig. 5, an image or other data may be sent to a user or another machine for review and to obtain further instructions. [00130] Although specific processes are discussed above with respect to Figs. 5 and 6, one skilled in the art will recognize that any of a variety of processes may be utilized in accordance with embodiments of the invention.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Operations Research (AREA)
  • Biochemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Epidemiology (AREA)
  • Medical Informatics (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Analytical Chemistry (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Child & Adolescent Psychology (AREA)
  • Automatic Analysis And Handling Materials Therefor (AREA)

Abstract

Dans un mode de réalisation de l'invention, un système d'automatisation d'équipement de laboratoire comprend un dispositif d'interprétation pour recevoir une première entrée utilisateur qui décrit une expérience biologique, et générer un programme d'expérience comprenant des commandes d'automatisation et des informations temporelles, chaque commande d'automatisation étant déterminée en fonction d'une signification biologique interprétée à partir de l'entrée utilisateur et adressée à un équipement de laboratoire automatisé pour opérer sur au moins un support d'échantillons contenant des échantillons, et un dispositif d'exécution lisant le programme d'expérience et dirigeant des commandes d'automatisation vers l'équipement de laboratoire en fonction des informations temporelles dans l'exécution des flux de travail, le dispositif d'interprétation étant également destiné à recevoir une deuxième entrée uilisateur qui décrit au moins une commande biologique, mettre à jour le programme d'expérience en temps réel tout en réalisant l'expérience biologique par adaptation des commandes d'automatisation et des informations temporelles pour s'adapter à la commande biologique décrite, diriger l'équipement de laboratoire avec des opérations modifiées sur des échantillons actifs.
PCT/US2024/048706 2023-09-26 2024-09-26 Systèmes et procédés de commande d'expérience biologique dynamique et en temps réel Pending WO2025072554A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202363585417P 2023-09-26 2023-09-26
US63/585,417 2023-09-26

Publications (1)

Publication Number Publication Date
WO2025072554A1 true WO2025072554A1 (fr) 2025-04-03

Family

ID=95202116

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/048706 Pending WO2025072554A1 (fr) 2023-09-26 2024-09-26 Systèmes et procédés de commande d'expérience biologique dynamique et en temps réel

Country Status (1)

Country Link
WO (1) WO2025072554A1 (fr)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060168515A1 (en) * 2005-01-27 2006-07-27 Symyx Technologies, Inc. Parser for generating structured data
US20140095007A1 (en) * 2005-09-30 2014-04-03 Colin Angle Companion robot for personal interaction
US20170069009A1 (en) * 2007-09-28 2017-03-09 Great-Circle Technologies, Inc. Contextual execution of automated workflows
US20190228269A1 (en) * 2018-01-04 2019-07-25 IAS Machine, LLC Procedural language and content generation environment for use in augmented reality/mixed reality systems to support laboratory and related operations
US20210374370A1 (en) * 2015-10-23 2021-12-02 Gen-Probe Incorporated Systems And Methods For Reading Machine-Readable Marks On Racks And Receptacles
US20230004886A1 (en) * 2021-07-02 2023-01-05 Strateos, Inc. Systems and methods for performing experiments at remote laboratories
US20230120417A1 (en) * 2008-09-24 2023-04-20 First Light Diagnostics, Inc. Kits And Devices For Detecting Analytes
US20230259705A1 (en) * 2021-08-24 2023-08-17 Unlikely Artificial Intelligence Limited Computer implemented methods for the automated analysis or use of data, including use of a large language model

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060168515A1 (en) * 2005-01-27 2006-07-27 Symyx Technologies, Inc. Parser for generating structured data
US20140095007A1 (en) * 2005-09-30 2014-04-03 Colin Angle Companion robot for personal interaction
US20170069009A1 (en) * 2007-09-28 2017-03-09 Great-Circle Technologies, Inc. Contextual execution of automated workflows
US20230120417A1 (en) * 2008-09-24 2023-04-20 First Light Diagnostics, Inc. Kits And Devices For Detecting Analytes
US20210374370A1 (en) * 2015-10-23 2021-12-02 Gen-Probe Incorporated Systems And Methods For Reading Machine-Readable Marks On Racks And Receptacles
US20190228269A1 (en) * 2018-01-04 2019-07-25 IAS Machine, LLC Procedural language and content generation environment for use in augmented reality/mixed reality systems to support laboratory and related operations
US20230004886A1 (en) * 2021-07-02 2023-01-05 Strateos, Inc. Systems and methods for performing experiments at remote laboratories
US20230259705A1 (en) * 2021-08-24 2023-08-17 Unlikely Artificial Intelligence Limited Computer implemented methods for the automated analysis or use of data, including use of a large language model

Similar Documents

Publication Publication Date Title
JP7498016B2 (ja) デジタルツインによるプロセス制御
JP6877051B2 (ja) ネットワーク化されたインキュベータ動作
Winn An overview of the CCP4 project in protein crystallography: an example of a collaborative project
Wassermann et al. Sedna: A BPEL-based environment for visual scientific workflow modeling
Akram et al. Evaluation of BPEL to scientific workflows
CN100538630C (zh) 在为客户提供用户接口信息的存取系统中使用的系统
US7519917B2 (en) Method and apparatus for graphically displaying compatible workflow steps
US20060045348A1 (en) Method and apparatus for automated cellular bioinformatics
JP7532073B2 (ja) デジタルツインによるモデリング
WO2003102854A2 (fr) Systeme et procede de commande en circuit ouvert et de suivi d'instruments biologiques
CN111798216A (zh) 一种机器人流程自动化多工具集成方法和系统
US20080126987A1 (en) Graphical representation of compatible workflow steps
Souza et al. Towards lightweight data integration using multi-workflow provenance and data observability
Matzke et al. Bayesian inference for psychology, part III: Parameter estimation in nonstandard models
Peterka et al. ASCR workshop on in situ data management: Enabling scientific discovery from diverse data sources
CN101561850B (zh) 海量多道神经信息综合处理系统及其控制方法
US20070156878A1 (en) System and method for managing workflow execution in a distributed system
Casanova et al. The virtual instrument: Support for grid-enabled mcell simulations
Horchidan et al. Evaluating model serving strategies over streaming data
KR101279392B1 (ko) 생명 정보 분석 파이프라인 처리 시스템 및 방법
WO2025072554A1 (fr) Systèmes et procédés de commande d'expérience biologique dynamique et en temps réel
Maccagnan et al. Combining ontologies and workflows to design formal protocols for biological laboratories
Arzymatov et al. SANgo: a storage infrastructure simulator with reinforcement learning support
EP1493096B1 (fr) Automatisation centree sur les donnees
García et al. Data-intensive analysis for scientific experiments at the large scale data facility

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24873611

Country of ref document: EP

Kind code of ref document: A1