A high-throughput screening experiment system and method based on multi-robot arm cooperation
The high-throughput screening experimental system with multiple robotic arms has achieved full automation of chemical experiments, solving the problems of cumbersome operation and low efficiency in traditional chemical experiments, and improving experimental accuracy and safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DALIAN UNIV OF TECH
- Filing Date
- 2024-04-15
- Publication Date
- 2026-08-04
AI Technical Summary
Existing robotic arms for chemical experiments suffer from cumbersome manual operation, low efficiency of a single robotic arm, complex data processing, and susceptibility to human error, making it impossible to achieve full automation and high-throughput chemical experiments.
A high-throughput screening experimental system employing multi-robotic arm collaboration includes an intelligent control terminal, a multi-robotic arm collaborative high-throughput experimental workstation, a product analysis workstation, and an intelligent data analysis unit. It completes the entire chemical experiment process through multi-robotic arm collaboration and achieves automated data analysis by combining machine learning and Bayesian optimization.
It improves the automation and precision of chemical experiments, reduces human error, increases work efficiency, frees up the time of experimental staff so that they can focus their energy on chemical research, and reduces the risk of personal injury.
Smart Images
Figure CN118219268B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of high-throughput chemical experimental technology, and more particularly to a high-throughput screening experimental system and method based on multi-robotic arm collaboration. Background Technology
[0002] With the rapid development of intelligent manufacturing technology, high-throughput experiments have entered a new development model. In traditional chemical experiments, chemists need to perform complex operations such as reagent preparation, mixing, heating reactions, and detection and analysis. These processes are easily affected by the instability of human operation, thus affecting the accuracy of experimental results. In addition, repetitive operations also consume a lot of chemists' time and energy, which is not conducive to the progress of chemical research.
[0003] Currently, robotic arm technology has advanced to the point where it can replace humans in performing a large number of repetitive and simple tasks, which is highly compatible with the repetitive procedures in chemical experiments. However, relying solely on robotic arms to complete a series of operations in chemical experiments and to switch operating modes according to the experimental scenario remains a significant challenge in building smart laboratories.
[0004] In addition, some related invention patents have also involved the application of robotic arms in chemical experiments. Chinese invention patent (CN 111659483 A) discloses an automated chemical experiment system based on a six-axis robotic arm, which allows the robotic arm to move within its working range to complete chemical experiment tasks in batches. However, this invention does not include the detection and analysis of experimental results. Chinese invention patent (CN 115660572A) discloses an automated chemical experiment scheduling method and system, which can automatically schedule automated equipment to perform chemical experiments and detections according to the content of the chemical experiment. Chinese invention patent (CN 107782885A) discloses a high-throughput screening system based on a single robotic arm, which realizes a production line operation for high-throughput screening of microorganisms through the rotation of the robotic arm. Chinese invention patent (CN 116651342 A) discloses an automated device and method for high-throughput screening of drug synthesis conditions, which realizes automated weighing, sample injection, sampling, and capping operations.
[0005] Existing robotic arms for chemical experiments have certain shortcomings: 1. Traditional experiments often require tedious manual operations, including weighing chemicals, preparing reagents, sealing well plates, conducting reactions, and collecting and processing data. These involve a large number of repetitive operations, which not only consume valuable time and energy of chemists but also easily introduce human error, limiting the accuracy and reproducibility of the experiments. 2. Current technologies do not address the coordinated scheduling of multiple robotic arms, resulting in low efficiency for individual robotic arms. 3. Traditional experiments generate large and complex datasets, requiring time-consuming data processing and analysis by chemists. Summary of the Invention
[0006] To address the aforementioned technical problem that existing robotic arms in chemical experiments cannot collaboratively handle the entire experimental process, this invention provides a high-throughput screening experimental system and method based on multi-robotic arm collaboration. The main purpose of this invention is to achieve automated, end-to-end high-throughput chemical experiments by setting up a high-throughput screening experimental system based on multi-robotic arm collaboration.
[0007] The technical means employed in this invention are as follows:
[0008] A high-throughput screening experimental system based on multi-robotic arm collaboration includes:
[0009] The intelligent control terminal is used to input experimental instructions and retrieve several tasks from the task set to form an experimental workflow according to the experimental instructions. The intelligent control terminal sends corresponding experimental signals to the multi-robotic arm collaborative high-throughput experimental workstation according to the experimental workflow. The task is edited from several actions in the action set.
[0010] A multi-robotic arm collaborative high-throughput experimental workstation is used to execute experimental signals sent by an intelligent control terminal and transport the experimental products generated to a product analysis workstation.
[0011] The product analysis workstation is used to analyze experimental products and obtain analysis results. The product analysis workstation sends the analysis results to the intelligent data analysis unit.
[0012] The intelligent data analysis unit is used to analyze the spectrum sequentially based on the analysis results to obtain the yield results. The relevant data obtained from the analysis are used for machine learning model training to predict the reaction yield. At the same time, the experimental conditions are optimized using Bayesian methods based on these data to obtain the optimal conditions for the reaction.
[0013] Furthermore, the multi-robotic arm collaborative high-throughput experimental workstation includes a weighing robot, a pipetting robot, a capping robot, a temperature-controlled vibration reaction module, and a collaborative linkage scheduling module;
[0014] The weighing robot includes a first six-axis robotic arm, a reagent storage management module, and an analytical balance; the six-axis robotic arm is used to perform gripping functions, and the reagent storage management module is used to manage solid and liquid chemicals in the drug warehouse;
[0015] The pipetting robot includes a three-axis robotic arm with a gripper, a consumable storage module, a multi-specification liquid storage module, and multi-specification high-precision pipettes; the multi-specification liquid storage module is used to store pipette tips of different specifications, glass containers of different specifications, and deep-hole plates of different specifications.
[0016] The collaborative scheduling module includes a second six-axis robotic arm, a mobile trolley, and a wireless charging station; three grippers for the second six-axis robotic arm are mounted on the mobile trolley, and one gripper for the second six-axis robotic arm is mounted next to the high-performance liquid chromatography-mass spectrometry instrument.
[0017] Furthermore, the product analysis workstation includes a high-performance liquid chromatography-mass spectrometry (HPLC-MS), a gas chromatography-mass spectrometry (GC-MS), and a liquid chromatograph.
[0018] Furthermore, the intelligent control terminal includes a central control unit, a terminal application load workstation, and a dedicated wireless router; the terminal application load workstation is equipped with an application that controls the entire automated experimental system, a corresponding UI interface, a 3D digital twin module, and an automated experimental database; the central control unit is used to control communication between the various modules; the dedicated wireless router is used to generate a local area network to connect the terminal application load workstation with each user terminal accessing the local area network.
[0019] Furthermore, the automated experimental database includes a material management database, an experimental task workflow database, and a high-throughput experimental database.
[0020] Furthermore, the intelligent data analysis unit includes:
[0021] The spectral analysis module is used to analyze the spectra generated by various analytical devices in the product analysis workstation, obtain the corresponding data of the target product, including the content data of each component of the raw material and the product, and transmit the data to the machine learning-based intelligent prediction module and the experimental condition intelligent optimization module.
[0022] The machine learning-based intelligent prediction module is used to train machine learning models on the data generated by the spectral analysis module, thereby predicting reaction yields.
[0023] The intelligent experimental condition optimization module is used to perform Bayesian optimization on the data generated by the spectral analysis module in combination with the corresponding experimental conditions. The optimized condition data is then used to form experimental instructions, which are transmitted to the intelligent control terminal to form new experimental signals. The reaction experiment with optimized experimental conditions is then executed by a high-throughput experimental workstation with multiple robotic arms. The data generated by the product analysis workstation is then transmitted back to the intelligent data analysis unit. After analysis by the spectral analysis module, a second intelligent optimization of experimental conditions is performed. After multiple iterations, an experimental parameter configuration close to the optimal solution is obtained.
[0024] Furthermore, the action set includes a weighing robot action set, a pipetting robot action set, a capping robot action set, a mobile cart robot action set, and a chromatograph action set;
[0025] The weighing robot's action set includes actions such as gripping the balance with a specified sample head value, gripping a specified container onto the balance, weighing the target mass, and returning to the original position.
[0026] The action set of the pipetting robot includes actions such as picking up a specified pipette, transferring liquid to a specified container, filtering, and returning to the original position;
[0027] The capping robot motion set includes actions such as tightening chromatographic vial caps, opening chromatographic vial caps, tightening centrifuge tube caps, opening centrifuge tube caps, fixing metal 96-well plate covers, and opening metal 96-well plate covers.
[0028] The motion set of the mobile robot includes actions such as changing the designated gripper, transferring the sample plate, pressing the analyzer button, and discarding waste materials;
[0029] The reaction zone action set includes actions for setting the reaction time, setting the reaction temperature, setting the vibration frequency, and initialization.
[0030] The chromatograph action set includes actions for initialization, injection, cleaning, and resetting position;
[0031] The task set includes sample weighing tasks, reaction solution preparation tasks, sample vial sealing tasks, sample transfer tasks, product dilution tasks, and chromatographic analysis tasks. The designed tasks are saved in the task set according to the specific needs of the experiment, and the task set is expanded in real time.
[0032] This invention also provides a high-throughput screening experimental method based on multi-robotic arm collaboration, implemented using any of the above-mentioned high-throughput screening experimental devices based on multi-robotic arm collaboration, comprising the following steps:
[0033] S1. Develop high-throughput experimental procedures;
[0034] The user terminal is connected to the local area network. The user terminal inputs experimental instructions through the UI interface of the terminal application load workstation and retrieves several tasks from the task set to form an experimental workflow according to the experimental instructions. The intelligent control terminal sends the corresponding experimental signals to the multi-robotic arm collaborative high-throughput experimental workstation according to the experimental workflow.
[0035] S2, Multi-robotic arm collaborative experiment;
[0036] The multi-robotic arm collaborative high-throughput experimental workstation executes the corresponding commands of the experimental signals according to the time nodes of the experimental signals, and feeds back the status to the 3D digital twin module in real time to form a digital twin laboratory.
[0037] The exchange of medicines, reactants, and experimental containers between the collaborative scheduling module and the consumables storage module;
[0038] The temperature-controlled vibration reaction module is used to control the temperature and vibration to enable the reactor to react. After the reaction is completed, the reactor is sent to the pipetting robot for sample dilution and filtration. At the same time, it exchanges items with the consumable storage module. After filtration, the sample is sent to different analytical module instruments for detection and analysis. The mobile cart returns the sample bottle to the fixed charging pile to wait for the next operation, completing the multi-robotic arm collaborative experiment.
[0039] S3. High-throughput experimental sample analysis;
[0040] High-throughput experimental sample analysis was performed using a chromatograph in the product analysis workstation to obtain analytical results.
[0041] S4, Intelligent Data Processing;
[0042] After the analysis results are processed by the intelligent data analysis unit, a high-throughput screening test report is generated. The test report and feedback information are uploaded to the central control computer, and a report is generated on the intelligent control terminal and stored in the database.
[0043] Furthermore, S4 specifically includes the following steps:
[0044] Based on the requirements, the analysis results are integrated with previous data, or input separately. The intelligent prediction module based on machine learning uses various machine learning methods to train the model, obtain the weights of each factor in the experiment and the corresponding model prediction performance map. The experimental conditions are optimized using Bayesian optimization in the experimental condition intelligent optimization module to obtain the recommended optimal experimental conditions. The experimental report is integrated and stored in the database.
[0045] Compared with the prior art, the present invention has the following advantages:
[0046] Compared with existing technologies, this system adopts the collaborative cooperation of multiple robotic arms. The first six-axis robotic arm on the weighing robot, the three-axis robotic arm on the pipetting robot, and the second six-axis robotic arm on the collaborative linkage scheduling module work together to make the automated chemical experiment system highly definable to the working environment. At the same time, the reagents can be transferred smoothly and efficiently between devices. With the cooperation of various devices, high-throughput chemical experiments and analyses can be automated.
[0047] This invention utilizes a robotic arm and various devices to perform tasks such as sample loading, reagent addition, reaction control, and result analysis, avoiding the instability caused by human operation and improving work efficiency and experimental accuracy.
[0048] This invention can also liberate experimental personnel, freeing them from repetitive experimental operations, allowing them to devote more energy to other aspects of chemical research and promoting efficient research.
[0049] This invention offers enhanced safety and stability, freeing people from chemical experiments and reducing the likelihood of personal injury. Attached Figure Description
[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0051] Figure 1 This is a system architecture diagram of the present invention.
[0052] Figure 2 This is a plan view of the system structure of the present invention.
[0053] Figure 3 This is a three-dimensional view of the system structure of the present invention.
[0054] Figure 4 This is the system scheduling diagram for the present invention.
[0055] In the diagram: 1. Multi-arm collaborative high-throughput experimental workstation; 11. Weighing robot; 111. Reagent storage management module; 112. First six-axis robotic arm; 113. Analytical balance; 12. Pipetting robot; 121. Three-axis robotic arm; 122. Consumable storage module; 123. Multi-specification liquid storage module; 124. Multi-specification high-precision pipette; 13. Capping robot; 14. Temperature-controlled vibration reaction module; 15. Collaborative linkage scheduling module; 151. Second six-axis robotic arm; 152. Mobile cart; 153. Wireless charging station; 16. Waste bin; 2. Product analysis workstation; 21. High-performance liquid chromatography-mass spectrometry; 22. Liquid chromatograph; 23. Gas chromatography-mass spectrometry; 3. Intelligent control terminal. Detailed Implementation
[0056] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0058] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.
[0059] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.
[0060] like Figure 1-4 As shown, the present invention provides a high-throughput screening experimental system based on multi-robotic arm collaboration, including a multi-robotic arm collaborative high-throughput experimental workstation 1, a product analysis workstation 2, an intelligent control terminal 3, and an intelligent data analysis system.
[0061] The multi-robotic arm collaborative high-throughput experimental workstation 1 includes a weighing robot 11, a pipetting robot 12, a capping robot 13, a temperature-controlled vibration reaction module 14, and a collaborative linkage scheduling module 15. The weighing robot 11 is primarily a six-axis robotic arm 112, and is also equipped with a reagent storage management module 111 and a high-precision analytical balance 113 with a precision of 1 / 200,000. The reagent storage management module 111 manages solid and liquid chemicals in the drug library. The sample storage containers have chips containing drug information, including drug name, CAS number, relative molecular weight, density, concentration, storage location, and remaining stock. The reagent storage management module 111 can store up to 32 types of solid samples and 18 types of liquid samples. The pipetting robot 12 is primarily a three-axis robotic arm 121 with a gripper, and is equipped with a consumable storage module 122, a multi-specification liquid storage module 123, and multi-specification high-precision pipettes 124. The multi-specification liquid storage module 123 can store pipette tips of different sizes, glass containers of different sizes ranging from 1 to 250 ml, and deep-well plates of different sizes. The consumable storage module 122 includes a self-conveying guide rail and various consumables to replenish the consumables used by the pipetting robot 12. The collaborative linkage scheduling module 15 includes a second six-axis robotic arm 151, a mobile cart 152, and a wireless charging station 153. Of the four specially designed grippers, grippers one, two, and three are mounted on the mobile cart 152, while gripper four is fixed next to the liquid chromatography-mass spectrometry (LC-MS) high-performance liquid chromatograph 22. The wireless charging station 153 can charge the mobile cart 152. A waste bin 16 is provided on the experimental table for collecting waste.
[0062] Product analysis workstation 2 includes a high performance liquid chromatography-mass spectrometry system 21, a gas chromatography-mass spectrometry system 23, and a liquid chromatograph 22.
[0063] The intelligent control terminal 3 includes a central control unit, a terminal application load workstation, and a dedicated wireless router. The terminal application load workstation is equipped with the application controlling the entire automated experimental system, a corresponding UI interface, a 3D digital twin module, and an automated experimental database. The central control unit connects all modules in the system to maintain communication. The dedicated wireless router creates a local area network connecting the terminal application load workstation with the electronic devices of various users accessing the local area network. The database includes a material management database, an experimental task workflow database, and a high-throughput experimental database.
[0064] The intelligent data analysis system includes a graph analysis module, a machine learning-based intelligent prediction module, and an experimental condition intelligent optimization module.
[0065] A fully automated experimental system for high-throughput screening based on multi-robotic arm collaboration is mainly divided into four parts: action set, task set, workflow, and database. For example... Figure 4 Scheduling diagram of the fully automated system for high-throughput screening experiments.
[0066] The motion set includes motion sets for weighing robot 11, pipetting robot 12, capping robot 13, mobile cart robot 152, and the chromatograph. Specifically, the weighing robot 11 motion set includes actions such as gripping a balance with a specified sample tip value, gripping a specified container onto the balance, weighing the target mass, and returning to the correct position. The pipetting robot 12 motion set includes actions such as gripping a specified pipette, transferring liquid to a specified container, filtering, and returning to the correct position. The capping robot 13 motion set includes actions such as tightening and opening chromatograph vials, tightening and opening centrifuge tube caps, fixing and opening metal 96-well plate covers. The mobile cart robot 152 motion set includes actions such as changing specified grippers, transferring sample plates, pressing analytical instrument buttons, and discarding waste materials. The reaction zone motion set includes actions such as setting reaction time, setting reaction temperature, setting vibration frequency, and initialization. The chromatograph motion set includes actions such as initialization, sample injection, cleaning, and returning to the correct position.
[0067] Based on experimental requirements, several actions from the action set are retrieved, edited together to form a task, and saved to the task set. Tasks include sample weighing, reaction solution preparation, sample vial sealing, sample transfer, product dilution, and chromatographic analysis. The designed tasks can be saved in the task set according to specific experimental needs, and the task set can be expanded in real time.
[0068] Tasks from the task set are retrieved based on experimental requirements and integrated into an experimental workflow. This mainly consists of four steps: high-throughput experimental workflow design, multi-robotic arm collaborative experimentation, high-throughput experimental sample analysis, and intelligent data processing. Workflows can also be designed according to specific experimental needs, and experimental plans can be saved and expanded.
[0069] Experimental protocols are stored in the experimental task and workflow library within the database. Simultaneously, the material management library in the database stores data on all reagents and consumables used in the entire automated high-throughput screening experimental process. Before use, reagents are registered in the material management library with their name, CAS number, and quantity. Usage is recorded and deducted after each experiment. After the experiment, the analyzed data is processed by an intelligent data analysis system, sequentially analyzing graphs to obtain yield results, using machine learning to train a model to predict responses, and optimizing experimental conditions using Bayesian methods to obtain optimal conditions. The integrated and analyzed data is then uploaded to the high-throughput experimental database for management.
[0070] A fully automated high-throughput screening experiment based on multi-robotic arm collaboration is mainly divided into four steps: high-throughput experimental process design, multi-robotic arm collaborative experiment, high-throughput experimental sample analysis, and intelligent data processing.
[0071] The specific system scheduling process involves connecting users to the system's dedicated local area network, designing and editing modular experimental commands or command combinations through the UI interface to invoke actions in the action set, forming new tasks, and storing them in the task set. The task set is integrated into the system to form an experimental workflow, and the workflow instructions are distributed to various systems. Each module executes corresponding commands according to time nodes and provides real-time status feedback, forming a digital twin laboratory for convenient online viewing by users. After receiving the execution command, each workstation completes tasks such as weighing reagents, preparing solutions, capping reagent bottles, and opening bottle caps according to instructions in the intelligent control terminal 3. The exchange of reagents, reactants, and experimental containers in the consumables storage module is completed through the collaborative scheduling module 15. The reactor is delivered to the intelligent temperature-controlled vibration reaction zone for reaction, where temperature and vibration can be controlled to meet reaction requirements. After the reaction is complete, the reactor is sent to the pipetting robot 12 for sample dilution and filtration, while simultaneously exchanging items with those in the consumables storage module; reaction container caps, etc., are handled by the mobile cart 152. After filtration, the samples are sent to different analytical module instruments for detection and analysis. Mobile trolley 152 returns the sample vials to the fixed charging station to await further instructions, completing the multi-robotic arm collaborative experiment. High-throughput sample analysis is performed using the chromatograph in product analysis workstation 2. The analysis results are then processed by the intelligent data analysis system to generate a high-throughput screening test report. After each module completes its phased execution, it uploads the experimental report and feedback information to the central control unit, where a report is generated and stored in the database for user access at any time.
[0072] To elaborate further:
[0073] Task motion design: Users access the system via a dedicated local area network, retrieve motions from the robot motion set, and generate tasks.
[0074] Experimental workflow design: Select task combinations from the task set to design an experimental plan, generate a workflow, and store it in the database.
[0075] Weighing reagents: The system issues instructions to search for the required reaction raw materials from the storage management module and weigh the reagents, including both solid and liquid reagents. After weighing, the mobile cart 152 delivers the high-throughput reactor to the designated area.
[0076] Pipette Preparation: Before pipetting, the moving carriage 152 grips the required solvents and consumables, such as pipette tips and 96-well plates, and places them in the corresponding areas. The consumables storage module replenishes consumables automatically via a self-replenishing rail. For bottles requiring capping, the moving carriage 152 opens and tightens the caps on the capping machine.
[0077] Solution preparation: The high-throughput reactor is fed to the reactor area, where a pipetting procedure is performed to complete the solution preparation. If no pipetting instruction is given, the reactor will be sent directly to the reaction area.
[0078] Post-pipette processing: The reactor containing the reactants is transported to the reaction area, where it is capped or screwed on according to instructions. Used consumables are picked up by the mobile cart 152 and transported to the waste area. Solvent bottles are tightened and returned to the consumables storage module.
[0079] Start the reaction: According to the instructions, place the reactor at the corresponding reaction site to carry out the reaction, which may be vibration only, heating only, or heating and vibration.
[0080] Sample preparation: After the reaction is complete, the reactor containing the product is transferred to the pipetting robot 12 for sample preparation, including dilution and filtration. The dispatching system's mobile cart 152 removes the filtration plate.
[0081] Sample analysis: The prepared sample plate is sent to the capping machine by the moving cart 152, the corresponding cap is added, and then it is sent to different analytical instruments, such as gas mass spectrometry, high performance liquid chromatography 22 and liquid mass spectrometry, high performance liquid chromatography 22, and the analysis program is started according to the instructions.
[0082] Feedback Report: After the analysis program is completed, the system terminal receives the analysis data, the experiment ends, and the dispatch system mobile vehicle 152 returns to the fixed charging pile to stand by.
[0083] Data Analysis: The system terminal collects and uploads data, and the chromatogram analysis module in the intelligent data analysis system performs data analysis on the chromatograms uploaded by the analytical instruments to draw a reaction yield heatmap.
[0084] Model Prediction: Based on requirements, the obtained experimental data is integrated with previous data, or input separately. The intelligent prediction module, based on machine learning, trains the model using various machine learning methods to obtain the weights of each factor in the experiment and the corresponding model prediction performance graph. The experimental conditions are then optimized using Bayesian optimization in the intelligent experimental condition optimization module to obtain recommended optimal experimental conditions. The experimental report is then integrated and stored in the database.
[0085] The specific implementation plan can be adjusted according to the actual experimental process.
[0086] Example 1
[0087]
[0088] This embodiment uses the screening of organic amines, carboxylic acids, organic solvents, and coupling agents used in carbon-nitrogen bond coupling reactions to illustrate the continuous screening capability of the system of the present invention.
[0089] Among them, 13 organic amines were screened, including aniline, naphthylamine, o-toluidine, p-toluidine, p-methoxyaniline, p-nitroaniline, 4-aminotrifluorotoluene, m-chloroaniline, p-iodoaniline, p-chloroaniline, N-methylaniline, p-bromoaniline, and m-toluidine; 4 coupling reagents were screened, including HBTU, PyBOP, DPP-CI, and BOP-CI; 12 carboxylic acids were screened, including acetic acid, propionic acid, acrylic acid, butyric acid, hexanoic acid, levulinic acid, benzoic acid, p-chlorobenzoic acid, p-methoxybenzoic acid, nicotinic acid, isonicotinic acid, and acetylsalicylic acid; and 4 solvents were screened, including DMAC, DMF, MeCN, and DMSO, for a total of 1632 reactions.
[0090] Experimental steps:
[0091] 1. Weigh once;
[0092] Using a weighing robot, weigh the organic aromatic amine (0.25 mmol), organic carboxylic acid (0.25 mmol), coupling reagent (0.60 mmol), and triethylamine (0.60 mmol) into 2 ml vials. Weigh the solvent (10 ml) into 10 ml vials.
[0093] 2. Transfer solvent to dissolve the drug;
[0094] After weighing, the mobile robot transports the 48-well plate containing 2ml vials and the 24-well plate containing 10ml vials to the working area of the pipetting robot, where the pipetting robot transfers 1ml of solvent into each 2ml vial to dissolve the solute.
[0095] 3. Shake to aid dissolution;
[0096] After solvent transfer is complete, the mobile robot transports the 48-well plate containing 2ml vials to the encapsulation area, caps the 2ml vials, and then transports the 48-well plate to the shaker in the reaction area for shaking to aid dissolution. The 24-well plate containing 10ml vials is discarded to the waste area.
[0097] 4. Second weighing;
[0098] Using a weighing robot, 2 ml each of the organic solvent, organic acid, triethylamine, and coupling reagent selected for the experiment were weighed.
[0099] 5. Transfer the solution to a 96-well metal plate;
[0100] After the medicine has dissolved, the mobile robot moves the 48-well plate to the packaging area to open the 2ml vials, and then moves the 48-well plate to the working area of the pipetting robot.
[0101] The mobile robot transfers the reagents to be screened to the pipetting robot.
[0102] An organic solvent, organic acid, triethylamine, coupling reagent, and organic aromatic amine were sequentially added from 2 ml vials to 1 ml vials in a 96-well metal plate using a pipette. The equivalent ratio of organic acid:triethylamine:coupling reagent:organic aromatic amine was 1:1.2:1.2:1.
[0103] 6. Encapsulate a 96-hole metal plate and react;
[0104] The mobile robot transports the 96-hole metal plate to the encapsulation area for encapsulation. After encapsulation, the plate is transported to the heater in the reaction area for reaction. The reaction temperature is 25°C and the reaction time is 12 hours.
[0105] 7. Prepare test samples;
[0106] After the reaction is complete, the mobile robot transports the 96-well metal plate to the encapsulation area to open the cap, and then transports the 96-well metal plate to the working area of the pipetting robot, where the pipetting robot dilutes and filters the reaction solution, and then transfers it to the liquid chromatography vial of the 48-well plate.
[0107] 8. Testing;
[0108] The mobile robot transports the 48-well plate containing the test sample to the packaging area, caps the liquid chromatography vial, and then transfers the vial to a high-performance liquid chromatograph for detection and analysis.
[0109] 9. Results;
[0110] Ultimately, 1632 reactions were completed within 340 hours, and the optimal reaction conditions were selected.
[0111] Example 2
[0112]
[0113] This embodiment demonstrates the screening capability of the system of the present invention by screening the inorganic base, organic solvent, and catalyst ligand used in the Suzuki coupling reaction of 3-bromo-2-methylbenzoic acid and 4-(trifluoromethoxy)phenylboronic acid.
[0114] Among them, 8 inorganic bases were screened, including Na2CO3, K2CO3, Cs2CO3, K3PO4, CsF, NaOH, KOH, and KOAC; 4 organic solvents were screened, including DMSO, DMF, dioxane, and DME; and 9 catalyst ligands were screened, including P(tBu)3, P(Cy)3, PPh3, Xphos, SPhos, AmPhos, Xantphos, dtbpf, and dppf, for a total of 288 reactions.
[0115] Experimental steps:
[0116] 1. Weighing;
[0117] Using a weighing robot, weigh 3-bromo-2-methylbenzoic acid (0.5 mmol), 4-(trifluoromethoxy)phenylboronic acid (0.55 mmol), palladium acetate (0.04 mmol), inorganic base (2 mmol), and catalyst ligand (0.02 mmol) into 2 ml vials.
[0118] 2. Transfer solvent to dissolve the drug;
[0119] After weighing, the mobile robot transports the 48-well plate containing 2ml vials to the working area of the pipetting robot, where the pipetting robot transfers 1ml of solvent into each 2ml vial to dissolve the solute.
[0120] 3. Shake to aid dissolution;
[0121] After the solvent transfer is complete, the mobile robot transports the 48-well plate containing 2ml vials to the encapsulation area to cap the 2ml vials, and then transports the 48-well plate to the shaker in the reaction area for shaking to aid dissolution.
[0122] 4. Transfer the solution to a 96-well metal plate;
[0123] After the medicine has dissolved, the mobile robot moves the 48-well plate to the packaging area to open the 2ml vials, and then moves the 48-well plate to the working area of the pipetting robot.
[0124] Using a pipette robot, 100 μL of 3-bromo-2-methylbenzoic acid solution (0.05 mmol), 100 μL of 4-(trifluoromethoxy)phenylboronic acid solution (0.055 mmol), 50 μL of palladium acetate solution (0.002 mmol), 50 μL of alkaline solution (0.1 mmol), and 200 μL of ligand solution (0.004 mmol) were transferred from 2 mL vials to 1 mL vials in a 96-well metal plate. The reaction volume of each vial was 500 μL.
[0125] 5. Encapsulate a 96-hole metal plate and react;
[0126] The mobile robot transports the 96-hole metal plate to the encapsulation area for encapsulation. After encapsulation, the plate is transported to the heater in the reaction area for reaction. The reaction temperature is 60℃ and the reaction time is 6 hours.
[0127] 6. Prepare test samples;
[0128] After the reaction is complete, the mobile robot transports the 96-well metal plate to the encapsulation area to open the cap, and then transports the 96-well metal plate to the working area of the pipetting robot, where the pipetting robot dilutes and filters the reaction solution, and then transfers it to the liquid chromatography vial of the 48-well plate.
[0129] 7. Testing;
[0130] The mobile robot transports the 48-well plate containing the test sample to the packaging area, caps the liquid chromatography vial, and then transfers the vial to a high-performance liquid chromatograph for detection and analysis.
[0131] 8. Results;
[0132] We ultimately completed 288 reactions within 60 hours and identified the optimal reaction conditions. Using DMF as the solvent, KOH as the inorganic base, and AmPhos as the catalyst ligand, the reaction yield reached 99.99%.
[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A high-throughput screening experimental system based on multi-robotic arm collaboration, characterized in that, include: The intelligent control terminal is used to input experimental instructions and retrieve several tasks from the task set to form an experimental workflow according to the experimental instructions. The intelligent control terminal sends corresponding experimental signals to the multi-robotic arm collaborative high-throughput experimental workstation according to the experimental workflow. The task is edited from several actions in the action set. A multi-robotic arm collaborative high-throughput experimental workstation is used to execute experimental signals sent by an intelligent control terminal and transport the experimental products generated to a product analysis workstation. The product analysis workstation is used to analyze experimental products and obtain analysis results. The product analysis workstation sends the analysis results to the intelligent data analysis unit. The intelligent data analysis unit is used to analyze the spectrum sequentially based on the analysis results to obtain the yield results. The relevant data obtained from the analysis are used for machine learning model training to predict the reaction yield. At the same time, the experimental conditions are optimized using Bayesian methods based on these data to obtain the optimal conditions for the reaction. The multi-robotic arm collaborative high-throughput experimental workstation includes a weighing robot, a pipetting robot, a capping robot, a temperature-controlled vibration reaction module, and a collaborative linkage scheduling module. The weighing robot includes a first six-axis robotic arm, a reagent storage management module, and an analytical balance; the first six-axis robotic arm is used to perform gripping functions, and the reagent storage management module is used to manage solid and liquid chemicals in the drug warehouse; The pipetting robot includes a three-axis robotic arm with a gripper, a consumable storage module, a multi-specification liquid storage module, and multi-specification high-precision pipettes; the multi-specification liquid storage module is used to store pipette tips of different specifications, glass containers of different specifications, and deep-hole plates of different specifications. The collaborative scheduling module includes a second six-axis robotic arm, a mobile trolley, and a wireless charging station; three grippers for the second six-axis robotic arm are mounted on the mobile trolley, and one gripper for the second six-axis robotic arm is mounted next to the high-performance liquid chromatography-mass spectrometry instrument. The product analysis workstation includes a high-performance liquid chromatography-mass spectrometry system, a gas chromatography-mass spectrometry system, and a liquid chromatograph. The intelligent control terminal includes a central control unit, a terminal application load workstation, and a dedicated wireless router; the terminal application load workstation is equipped with an application that controls the entire automated experimental system, a corresponding UI interface, a 3D digital twin module, and an automated experimental database; the central control unit is used to control the communication between the various modules; The dedicated wireless router is used to create a local area network to connect terminal application load workstations with various user terminals accessing the local area network. The intelligent data analysis unit includes: The spectral analysis module is used to analyze the spectra generated by various analytical devices in the product analysis workstation, obtain the corresponding data of the target product, including the content data of each component of the raw material and the product, and transmit the data to the machine learning-based intelligent prediction module and the experimental condition intelligent optimization module. The machine learning-based intelligent prediction module is used to train machine learning models on the data generated by the spectral analysis module, thereby predicting reaction yields. The intelligent experimental condition optimization module is used to perform Bayesian optimization on the data generated by the spectral analysis module in combination with the corresponding experimental conditions. The optimized condition data is then used to form experimental instructions, which are transmitted to the intelligent control terminal to form new experimental signals. The reaction experiment with optimized experimental conditions is then executed by a high-throughput experimental workstation with multiple robotic arms. The data generated by the product analysis workstation is then transmitted back to the intelligent data analysis unit. After analysis by the spectral analysis module, a second intelligent optimization of experimental conditions is performed. After multiple iterations, an experimental parameter configuration close to the optimal solution is obtained.
2. The high-throughput screening experimental system based on multi-robotic arm collaboration according to claim 1, characterized in that, The automated experimental database includes a material management database, an experimental task workflow database, and a high-throughput experimental database.
3. The high-throughput screening experimental system based on multi-robotic arm collaboration according to claim 1, characterized in that, The action set includes the weighing robot action set, the pipetting robot action set, the capping robot action set, the mobile cart robot action set, and the chromatograph action set; The weighing robot's action set includes actions such as gripping a designated sample feeding head onto the balance, gripping a designated container onto the balance, weighing the target mass, and returning to the original position; The action set of the pipetting robot includes actions such as picking up a specified pipette, transferring liquid to a specified container, filtering, and returning to the original position; The capping robot motion set includes actions such as tightening chromatographic vial caps, opening chromatographic vial caps, tightening centrifuge tube caps, opening centrifuge tube caps, fixing metal 96-well plate covers, and opening metal 96-well plate covers. The motion set of the mobile robot includes actions such as changing the designated gripper, transferring the sample plate, pressing the analyzer button, and discarding waste. The reaction zone action set includes actions for setting the reaction time, setting the reaction temperature, setting the vibration frequency, and initialization. The chromatograph action set includes actions for initialization, injection, cleaning, and resetting position; The task set includes sample weighing tasks, reaction solution preparation tasks, sample vial sealing tasks, sample transfer tasks, product dilution tasks, and chromatographic analysis tasks. The designed tasks are saved in the task set according to the specific needs of the experiment, and the task set is expanded in real time.
4. A high-throughput screening experimental method based on multi-robotic arm collaboration, implemented based on the high-throughput screening experimental system based on multi-robotic arm collaboration as described in any one of claims 1-3, characterized in that, Includes the following steps: S1. Develop high-throughput experimental procedures; The user terminal is connected to the local area network. The user terminal inputs experimental instructions through the UI interface of the terminal application load workstation and retrieves several tasks from the task set to form an experimental workflow according to the experimental instructions. The intelligent control terminal sends the corresponding experimental signals to the multi-robotic arm collaborative high-throughput experimental workstation according to the experimental workflow. S2, Multi-robotic arm collaborative experiment; The multi-robotic arm collaborative high-throughput experimental workstation executes the corresponding commands of the experimental signals according to the time nodes of the experimental signals, and feeds back the status to the 3D digital twin module in real time to form a digital twin laboratory. The collaborative scheduling module is used to realize the exchange of medicines, reactants and experimental containers in the consumables storage module; The temperature-controlled vibration reaction module is used to control the temperature and vibration to enable the reactor to react. After the reaction is completed, the reactor is sent to the pipetting robot for sample dilution and filtration. At the same time, it exchanges items with the consumable storage module. After filtration, the sample is sent to different analytical module instruments for detection and analysis. The mobile cart returns the sample bottle to the fixed charging pile to wait for the next operation, completing the multi-robotic arm collaborative experiment. S3. High-throughput experimental sample analysis; High-throughput experimental sample analysis was performed using a chromatograph in the product analysis workstation to obtain analytical results. S4, Intelligent Data Processing; After the analysis results are processed by the intelligent data analysis unit, a high-throughput screening test report is generated. The test report and feedback information are uploaded to the central control computer, and a report is generated on the intelligent control terminal and stored in the database.
5. The high-throughput screening experimental method based on multi-robotic arm collaboration according to claim 4, characterized in that, S4 specifically includes the following steps: Based on the requirements, the analysis results are integrated with previous data, or input separately. In the machine learning-based intelligent prediction module, various machine learning methods are used to train the model to obtain the weights of each factor in the experiment and the corresponding model prediction performance graph. In the experimental condition intelligent optimization module, Bayesian optimization is used to obtain the recommended optimal experimental conditions. The experimental report is integrated and stored in the database.