A multi-disciplinary shared scientific research activity management and application platform

Abstract

The application relates to a multi-disciplinary shared scientific research activity management and application platform. The platform comprises a scientific research activity management module deployed in the cloud and configured to enable users to manage scientific research activities by running the scientific research activity management module without the need for additional software installation. The scientific research activity management module comprises a scientific research unit design environment and scientific research unit code packages of different disciplines designed and generated by users based on the scientific research unit design environment and a multi-disciplinary shared scientific research unit syntax. Each scientific research unit code package contains at least a scientific research protocol of the corresponding discipline, which is used to define the basic information, multi-modal information and data fields of the scientific research experiment of the discipline. The application can meet the diversified and customized needs of scientific research data recording and scientific research activity management of different disciplines, thereby realizing efficient accumulation of multi-disciplinary expert knowledge and providing rich resources for AI applications based on cross-disciplinary expert knowledge.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of scientific research project management, and particularly relates to a scientific research activity management and application platform supporting multi-disciplinary sharing. BACKGROUND

[0002] In recent years, artificial intelligence (AI) has shown increasing potential in supporting and promoting scientific innovation in the field of natural sciences. From protein structure prediction to drug discovery, from identification of new chemical reactions to synthesis of new materials, artificial intelligence has shown great potential to assist or even replace human scientists. Strategically using artificial intelligence technology can accelerate the progress of natural science research and enhance the technological competitiveness of universities and enterprises.

[0003] Data is the foundation of science and the cornerstone of artificial intelligence. To widely realize AI-driven progress in natural science research, the first step is to promote the electronicization of multi-disciplinary scientific research data. This is because, first of all, existing AI models have strict requirements for high-quality and large amounts of training data. Therefore, current AI-assisted scientific research is usually focused on specific field problems with relatively rich public data, such as protein structure, compound structure, medical images, and scientific literature. In contrast, in those natural science fields that are not covered by public data sets, AI-enabled research is still limited due to the difficulty of obtaining relevant scientific research data (including experimental protocols, experimental data, etc.). Secondly, cross-disciplinary research is increasingly important in promoting scientific discovery. On the one hand, the number of cross-disciplinary research is growing; on the other hand, cross-disciplinary research is gradually becoming a source of major scientific breakthroughs. Therefore, creating a multi-disciplinary scientific research data electronicization platform is an important way to promote scientific research data electronicization and meet the future development of science. In order to promote the electronic recording of laboratory scientific research data, many functional electronic laboratory notebooks (ELNs) have been developed. However, these ELNs are usually designed for a single discipline, and lack the generality required to achieve multi-disciplinary scientific research data electronicization, for example, due to financial and human costs, and significant professional barriers between different disciplines, existing platforms usually adopt a centralized design concept, providing users with pre-defined functions and data recording templates, and cannot support custom research protocols, so different types of scientific research data required by different protocols cannot be recorded.

[0004] Therefore, the prior art has not found a scientific research management and application platform that can well meet the diversification and fine-grained customization needs in different discipline scientific research data records, and therefore, the high-precision and high-quality expert knowledge of multi-disciplinary front-line scientific researchers has not been efficiently accumulated through such a platform, thereby providing key resources for cross-disciplinary AI research model training. SUMMARY

[0005] The present application is provided to solve the above-mentioned defects in the prior art. A multi-disciplinary shared scientific research activity management and application platform is needed, which can enable users to perform scientific research activity management by running a scientific research activity management module deployed in the cloud without the need for additional software installation, and design and generate scientific research protocols of different disciplines using a unified scientific research unit syntax. The platform of the present application can efficiently accumulate high-precision and high-quality expert knowledge of multi-disciplinary front-line scientific researchers, thereby providing key and rich resources for AI applications based on cross-disciplinary expert knowledge.

[0006] According to a first aspect of the present application, a multi-disciplinary shared scientific research activity management and application platform is provided, which includes a scientific research activity management module deployed in the cloud and configured to enable users to perform scientific research activity management by running the scientific research activity management module without the need for additional software installation; the scientific research activity management module includes a scientific research unit design environment and scientific research unit code packages of different disciplines designed and generated by users based on a multi-disciplinary shared scientific research unit syntax defined by the users in the scientific research unit design environment, wherein each scientific research unit code package at least contains a scientific research protocol of the corresponding discipline, and the scientific research protocol is used to define the basic information, multi-modal information and data fields of the scientific research experiment of the discipline.

[0007] The multi-disciplinary shared scientific research activity management and application platform provided by each embodiment of the present application provides a scientific research unit design environment deployed in the cloud for users, enabling users to design and generate scientific research unit code packages of different disciplines based on a multi-disciplinary shared scientific research unit syntax, breaking through the limitations of professional barriers between different disciplines, and defining the basic information, multi-modal information and data fields of the scientific research experiment of the discipline by only designing and describing the scientific research protocol of the corresponding discipline. With lower platform development and maintenance costs, the platform well meets the diversification and fine-grained customization needs in different discipline scientific research data records. In addition, users do not need to install additional software to conveniently perform scientific research activity management, thereby providing the greatest convenience for users of the platform in different disciplines, facilitating the use and promotion of the platform, thereby enabling efficient accumulation of high-precision and high-quality expert knowledge of multi-disciplinary front-line scientific researchers, and thereby providing key and rich resources for AI applications based on cross-disciplinary expert knowledge.

[0008] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clearly understood and implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0009] It should be understood that the foregoing general description and the following detailed description are only illustrative and explanatory, and are not limiting of the claimed application. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0011] Figure 1(a) shows a composition schematic diagram of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0012] Figure 1(b) shows another composition schematic diagram of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0013] Figure 1(c) shows another composition schematic diagram of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0014] Figure 1(d) shows another composition schematic diagram of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0015] Figure 1(e) shows another composition schematic diagram of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0016] Figure 2 Figure 2 shows a composition structure schematic diagram of a scientific research unit code package according to an embodiment of the present application.

[0017] Figure 3(a) shows a composition structure schematic diagram of a scientific research activity management module according to an embodiment of the present application.

[0018] Figure 3(b) shows another composition structure schematic diagram of a scientific research activity management module according to an embodiment of the present application.

[0019] Figure 4 Figure 4 shows a step schematic diagram of recording a scientific research unit by a scientific research unit record environment according to an embodiment of the present application.

[0020] FIG. 5(a) shows a schematic diagram of an AI system tool chat interface according to an embodiment of the present application.

[0021] FIG. 5(b) shows a schematic diagram of customizing scientific unit grammar checking using an AI system tool according to an embodiment of the present application.

[0022] FIG. 5(c) shows a schematic diagram of injecting relevant information of a scientific unit as a situational context into an AI system tool chat dialogue according to an embodiment of the present application.

[0023] FIG. 5(d) shows an answer given by GPT-4o to a question asked by a user in an embodiment of the present application.

[0024] FIG. 5(e) shows a schematic diagram of analyzing scientific records using an AI system tool according to an embodiment of the present application.

[0025] FIG. 6(a) shows a schematic diagram of a scientific unit workflow according to an embodiment of the present application.

[0026] FIG. 6(b) shows a research path schematic diagram according to an embodiment of the present application.

[0027] FIG. 6(c) shows a schematic diagram of scientific data generated by performing a research path according to an embodiment of the present application.

[0028] Figure 7 FIG. 7 shows a process schematic diagram of generating an automatically executable scientific unit workflow using an AI system tool according to an embodiment of the present application.

[0029] FIG. 8(a) shows a schematic diagram of a form-style scientific unit record interface according to an embodiment of the present application.

[0030] FIG. 8(b) shows a schematic diagram of a WYSIWYG-style scientific unit record interface according to an embodiment of the present application.

[0031] Figure 9 FIG. 9 shows a knowledge sharing block schematic diagram according to an embodiment of the present application.

[0032] Figure 10 FIG. 10 shows a schematic diagram of a scientific conclusion automatically generated by an AI system tool according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the described embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without any creative effort fall within the scope of protection of the present application.

[0034] Unless otherwise defined, technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of "including", "comprising" or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

[0035] The terms "first", "second", and similar terms used herein do not denote any order, quantity, or importance, but are used to identify one of the elements described. The terms "including", "comprising", and similar terms used herein mean that the elements listed after the terms encompass the elements listed and do not preclude the presence of other elements. The order of execution or performance of the steps of the methods disclosed herein is not limited to the order given in the embodiments unless otherwise specified. Steps can be performed in an order other than that specifically disclosed or claimed unless otherwise specified. Steps can be performed in an order other than that specifically disclosed or claimed unless otherwise specified. Steps can be performed concurrently in a parallel process unless otherwise specified. Unless otherwise specified, steps can be performed in an order other than that specifically disclosed or claimed unless otherwise specified.

[0036] It should also be understood that the term "and / or" as used herein merely means that there are three possible relationships between the associated objects, for example, A and / or B can mean that A exists alone, A and B exist together, or B exists alone. In addition, the character " / " in this application generally represents an "or" relationship between the associated objects.

[0037] In order to keep the following description of the embodiments of the present application clear and concise, detailed descriptions of known functions and known components are omitted.

[0038] Figure 1(a) shows a schematic diagram of a composition of a scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0039] As shown in Figure 1(a), the scientific research activity management and application platform 10 supporting multi-disciplinary sharing according to an embodiment of the present application at least includes a scientific research activity management module 11, and the scientific research activity management module 11 further includes a scientific research unit design environment 111, and scientific research unit code packages 121, 122 corresponding to different disciplines, etc.

[0040] According to embodiments of the present application, in the research unit design environment 111, a new research unit syntax is defined for multi-disciplinary use, so that one or more users 12 of different disciplines can design a research protocol (hereinafter also referred to as a research unit protocol, and the two are not distinguished) of the corresponding discipline based on the research unit syntax defined by the research unit design environment 111, including defining the basic information, multi-modal information and data fields of the scientific experiment of the discipline, and the files defining these information are packaged in a folder, that is, a research unit code package of the discipline is generated, that is, a new research unit (hereinafter also referred to as Research Unit, or RU for short, and the three are not distinguished) is established, for example, the research unit code package 121 in FIG. 1(a) corresponds to the research unit related to protein purification of the life science discipline, and the research unit code package 122 corresponds to the research unit related to the self-dispersion of carbon nanotubes of the material science discipline. These research unit code packages can be parsed and run by the research activity management and application platform 10, so that researchers of different disciplines can carry out research experiments of the discipline by establishing research units while complying with the research protocol of the corresponding discipline.

[0041] For example only, in the research protocol, the basic information of the research experiment may, for example, include the user using text to describe the detailed content of the research plan related to the research activity (such as an experiment), including introduction, method, step, etc.; the multi-modal information may, for example, include the multi-modal files inserted by the user, including images, videos, audio, documents (such as pdf), etc., to provide more intuitive and detailed research activity description; the data field may, for example, include various data fields required in the research plan defined by the user, which may include research variables, steps and checkpoints related to the research activity, etc. For example, if the user needs to add a data field for the solvent volume in the research plan to record the amount of solvent used in the experiment, a data field with ID solvent_volume can be defined to record the solvent volume in the data field.

[0042] The scientific unit syntax (hereinafter also referred to as scientific unit Markdown, and the two are not distinguished) in the embodiments of the present application is a newly defined lightweight markup language that can be shared by multiple disciplines. Users can create formatted text based on the scientific unit syntax, for example, through a simple text editor, to customize the content of basic information, multi-modal information and data fields in the scientific protocol. It is similar to the standard Markdown language, has a clear syntax, is compatible with the syntax of the standard Markdown specification (CommonMark), and at the same time has syntax extensions suitable for the content requirements of scientific protocols, models, assigners, etc. in the scientific units of the present application. Compared with the standard Markdown syntax, its unique feature is that the scientific unit Markdown provides template syntax for users. Using template syntax, data fields (hereinafter also referred to as scientific unit data fields, and the two are not distinguished) can be inserted and customized. Users can generate scientific unit variables, scientific unit steps, or scientific unit checkpoints based on the template syntax of the data fields. In the scientific unit Markdown, templates are represented using {{...}}. Commonly used templates for defining scientific unit variables, scientific unit steps, and scientific unit checkpoints, etc. data fields are shown in Table 1.

[0043] Table 1 Template examples for defining data fields in scientific protocols

[0044]

[0045] The scientific activity management and application platform 10 and the scientific activity management module 11 according to the embodiments of the present application can be deployed locally or in the cloud. From the perspective of facilitating multi-terminal sharing, it is generally preferred to be deployed in the cloud. In this way, users do not need to install additional software locally, and can only run the scientific activity management module 11 in the cloud to manage scientific activities.

[0046] Figure 2 The composition structure of the scientific unit code package according to the embodiments of the present application is shown.

[0047] Figure 2 Taking the scientific unit code package 121 as an example, it can be seen that when designing the scientific unit code package of the corresponding discipline based on the scientific unit design environment, the user can not only design the scientific protocol, but also design and generate the model (Model) of the scientific unit. The model is used to define the type constraint and / or numerical verification relationship of the data field, as well as the type and / or numerical combination verification relationship between each data field.

[0048] More specifically, the type constraint includes a constraint that the corresponding data field uses predefined multi-modal information at data entry, where the predefined multi-modal information includes one or more of text, image, video, audio, and file. In some embodiments, the user can further define the data type of a data field previously defined in the scientific protocol (e.g., various numerical types, time types, etc.), for example, define the solvent_volume data field as a floating point number, then when the user enters data in this field that is not a floating point number (e.g., a string of letters), the system will indicate a type error.

[0049] The numerical check constraint includes a constraint that the corresponding data field follows a specified pattern and / or does not exceed a preset value range at data entry. The user can further define a validation rule for a data field defined in the scientific protocol, for example, add a validation rule for the solvent_volume data field and its type constraint (floating point number) to ensure that the floating point number filled in this data field must be greater than zero, in which case if the user enters a negative floating point number, the system will display a numerical check error.

[0050] The combination check constraint includes a constraint that the types and / or numerical values of various data fields satisfy a predetermined constraint relationship. By way of example only, the so-called following a specified pattern may, for example, be using a RegExp regular expression RE (Regular Expression) to constrain the composition rule and pattern of the string in the field, for example, constraining the value recorded in a field related to an email to contain and only contain 1 “@” symbol, etc. In other embodiments, other patterns and value range constraints can also be set, which are not listed one by one here. In this way, rapid identification of scientific abnormal data can be achieved. If any check constraint fails, the reason for the check failure will be displayed, and the user can correct the invalid data field value one by one according to the error prompt until the valid record is modified. Through the above check process, the platform can ensure that the data entered by the user meets the requirements.

[0051] As Figure 2As shown, in some embodiments, if a user defines multiple data fields with dependencies and assignment relationships in a research agreement, these relationships can be further customized using an assigner. Specifically, when a user designs a research unit code package for a corresponding discipline based on a research unit design environment, it can also include an assigner for the data fields. This assigner is used by the user to assign values ​​to the data fields based on the data field dependency graph and assignment rules. In some embodiments, the data field dependency graph is a single-level or multi-level directed acyclic graph, the assignment relationships between the defined data fields are single or multiple dependencies, and each data field is assigned a value by at most one assigner. An assigner can have one or more upstream data fields as dependencies, and a data field can actually be a dependency of one or more data fields, but for a specific data field, the way its field value is determined should be unique. As an example, if two data fields, solvent_volume and solvent_volume_2, are defined in a research protocol, and the user uses an assigner to ensure that solvent_volume_2 is always twice the value of solvent_volume, then whenever a new value is entered for solvent_volume, the system will automatically set solvent_volume_2 to twice the solvent volume. For example, if the value of solvent_volume is 5, the system will automatically assign the value 10 to solvent_volume_2. This can greatly improve the efficiency and accuracy of research data recording.

[0052] By defining models of research units, user input can be dynamically verified to ensure the accuracy of data entry. By defining assigners, multi-level and multi-dependent field dependencies can be automatically calculated based on the value of a certain input field, thereby ensuring that even if there are complex dependencies between multiple data fields, they can be efficiently entered and run correctly. This can significantly promote the electronic management of laboratory research data, including data from outsourced experiments and orders, as well as the efficient retrieval of research plans, research data, and other related content.

[0053] Figure 3(a) shows a schematic diagram of the composition structure of a scientific research activity management module according to an embodiment of the present application. As shown in Figure 3(a), in some embodiments, the scientific research activity management module 11 may include a scientific research unit recording environment 112 in addition to a scientific research unit design environment 111, which provides users with an environment for recording and storing data based on a pre-designed scientific research unit.

[0054] Figure 4A schematic diagram showing the steps of recording a research unit by a research unit recording environment according to an embodiment of the present application is shown. As previously described, in the case where the user-designed research unit code package contains the research protocol corresponding to the discipline, the model of the research unit, and the assigner of the data fields, the research unit recording environment 112 can perform the operation steps shown as Figure 4 when recording the research unit related content.

[0055] First, in step 401, the model of the research unit can be converted into a data field JSON Schema.

[0056] Then, in step 402, based on the data field JSON Schema, a structured storage scheme is automatically generated for the research unit, so that in the case where the user submits the storage of the research record based on the research unit, and each data field meets the constraint and checking relationship of the model, the research record is stored as a JSON that conforms to the structured storage scheme corresponding to the research unit. In the embodiments of the present application, no matter whether the user-defined data field is simple text or more complex multi-modal data, the research unit recording environment 112 can automatically generate the corresponding data structure for it and ensure the consistency and integrity of the data during the entry and storage process. Since the automatically generated data structure is based on standardized protocols and definitions, research data can be easily shared globally. The automation of such structured storage not only simplifies the work of researchers, but also improves the reproducibility of research data and the collaboration ability across laboratories. In addition, through the automatically generated data structure, researchers can focus on experiments and data recording without worrying about the underlying data management problems, which greatly reduces the data management burden of researchers and improves the efficiency of research work.

[0057] In other embodiments, the research unit recording environment 112 can be further configured to automatically generate a corresponding research unit recording interface for the research unit based on the research protocol and the data field JSON Schema, the research unit recording interface having a stylized style based on the parsing of the custom research unit syntax and interactive controls corresponding to the data types of each data field based on the parsing of the data field JSON Schema. For example, font settings associated with heading levels can be provided according to the basic style of the research unit Markdown, and the recording interface of multi-modal information with different data types can be set to a form style, an embedded WYSIWYG style, etc. as needed, so as to adapt to the diversified needs of research data and effectively improve the operation convenience and accuracy of researchers during the recording process.

[0058] Figures 8(a) and 8(b) respectively illustrate schematic diagrams of the form-style and WYSIWYG style research unit recording interfaces according to embodiments of this application. Because in Figure 4 In step 401, the research unit model has been converted into a data field JSON Schema. For example, the user has defined different data types (such as strings, integers, floating-point numbers, booleans, dates, enumerations, etc.) for different fields in the model. In this case, appropriate interactive controls can be generated for each data field in the research unit recording interface 800 shown in Figures 8(a) and 8(b). More specifically, the research unit recording environment 112 can parse the data types of each field in the data field JSON Schema that are specially annotated, and thereby generate the corresponding interactive controls. As an example, in Figures 8(a) and 8(b), interactive controls 801 and 801' are interactive controls generated based on data fields annotated as positive integers. Therefore, it can be seen that the automatically generated interactive controls 801 and 801' in the research unit recording interface 800 have buttons for increasing and decreasing the value; while interactive controls 802 and 802' are interactive controls generated based on data fields of the time data type, and therefore use a clock face as a prompt icon. The Research Unit Recording Interface 800 helps scientists focus their time and energy on defining and developing the substantive content of their research plans, such as research protocols, models, and data fields, without having to worry about the research recording interface, research data storage structure, and methods. This allows scientists to efficiently design high-quality research plans that meet actual research needs in a user-friendly manner during their daily research activities and use them for research data recording.

[0059] FIG. 3(b) shows another component structure diagram of the scientific activity management module according to an embodiment of the present application. As shown in FIG. 3(b), the scientific activity management module 11 can include a scientific unit design environment 111, a scientific unit record environment 112, and a scientific report generation and reader 113, which is configured to at least respond to a first operation of a user on a scientific unit, based on the historical scientific records of the scientific unit, locally run and automatically generate a formatted scientific report of the scientific unit, and enable the user to locally read the generated formatted scientific report. In addition, the scientific report generation and reader 113 can support the formatted output of the report content, such as PDF format, etc., which are not listed one by one here. Users can directly use these reports for literature publication, internal discussion or printing, greatly simplifying the generation process of scientific records and reports. It is worth noting that when rendering the scientific report, only static text files related to the scientific protocol, data field JSON Schema, JSON of the scientific records of the scientific unit code package, etc. are used, and the data field JSON Schema is automatically generated and saved when the platform first loads the user-customized scientific unit code package, and the user does not need to run the Python file of the scientific unit model or execute the Python code again when generating the scientific report in the future. Only the generated data field JSON Schema file needs to be called, which eliminates the security risk of the user reading and generating the scientific report locally. For this reason, the scientific unit can be used as a standard format for research data exchange in the future.

[0060] FIG. 1(b) shows another composition diagram of the scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application. As shown in FIG. 1(b), the scientific research activity management and application platform 10 can further include a scientific research unit sharing and application tool 13 in addition to the scientific research activity management module 11 and the scientific research unit code packages of different disciplines. The scientific research unit sharing and application tool 13 can be configured to, for example, generate a unique ID number for each scientific research record of a scientific research unit and associate the unique ID number with the record time so that each scientific research record is non-tamperable; when a user wants to modify based on a submitted scientific research record, generate a copy of the scientific research record for the user and generate a new unique ID number for the copy of the scientific research record and associate the new unique ID number with the modification time so that the user modifies on the copy with the new unique ID number. Due to the above method of making data non-tamperable, the scientific research activity management and application platform according to an embodiment of the present application can actually serve as a kind of evidence platform for scientific research activity history / records, for example, as evidence when scientific research evidence is needed for certain matters. For example, disputes often occur in the scientific research field as to who first discovered a certain phenomenon, and if the relevant scientific research personnel recorded the phenomenon in the platform of the present application and the relevant scientific research data left a trace in the platform, these scientific research records with non-tamperability can serve as evidence of their scientific discovery.

[0061] In other embodiments, the scientific research unit sharing and application tool 13 is further configured to, in response to a second operation on a scientific research unit by a user with corresponding permissions, mark the scientific research unit with a hierarchy, the hierarchy being divided into at least a laboratory layer and a project layer from high to low, and provide a user access control mechanism corresponding to a disclosure level for each layer and / or each scientific research unit; and set a hierarchy mark corresponding to the hierarchy division of the scientific research unit for each user.

[0062] In some embodiments, providing a corresponding user access control mechanism for each layer and / or each scientific research unit can specifically include the following: setting accessible permissions for each laboratory layer that are open to all users; setting accessible permissions for each project layer that are for users with a specific hierarchy mark; in the case that a specific user has accessible permissions for a specific project, the specific user has accessible permissions for all scientific research units in the specific project.

[0063] For example, setting accessible permissions for each project layer that are for users with a specific hierarchy mark can specifically include: setting accessible permissions for all users of the project layer; or only users with the same laboratory layer mark have accessible permissions for the project layer; or only users with the same project layer mark have accessible permissions for the project layer.

[0064] By way of example only, providing corresponding user access control mechanisms for each level and / or each research unit can further include setting a group tag for users with the same laboratory level tag and correspondingly setting accessible permissions for each group tagged user to a specific project level in response to a third operation of a user with corresponding permissions.

[0065] Other user access control mechanisms can be set according to the specific characteristics and requirements of the disciplines and institutions, which are not listed one by one in this application, and the principle of maximizing sharing under the premise of ensuring the safety and privacy management requirements of experimental schemes and experimental data is taken as the reference.

[0066] FIG. 1(c) shows another component diagram of a research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application.

[0067] As shown in FIG. 1(c), based on the component structure of the research activity management and application platform 10 shown in FIG. 1(a), a knowledge sharing block corresponding to each research unit can be further included. Figure 9 A knowledge sharing block diagram according to an embodiment of the present application is shown. Figure 9 The knowledge sharing block 900 shown for the protein purification research unit can support real-time discussion by multiple people, provide a rich text editor, support comment and reply functions, and can provide a voting mechanism, etc., which are not described here. Different research units usually correspond to different disciplines, and therefore, by setting a knowledge sharing block for each research unit, the maintainer or other interested users of the research unit can more conveniently provide or obtain more focused professional knowledge in the field related to the research activities of the research unit in the form of questions and answers in the knowledge sharing block.

[0068] In an embodiment of the present application, the contents in the research agreement and the knowledge sharing block can come from the expert experience and knowledge of community users, and the contents in the knowledge sharing block can be updated in real time and dynamically on the research activity management and application platform, without modifying the research unit code package, which gives the knowledge sharing block more flexible updating characteristics.

[0069] In other embodiments, the research activity management and application platform 10 can be further configured to use the contents in the knowledge sharing block to update the research agreement of the corresponding research unit. By way of example only, the update of the research agreement described above can be triggered periodically or manually by a user, so that the latest field knowledge in the knowledge sharing block can be incorporated into the research unit code package, so that the newly generated research unit using the same research unit code package has more advanced features in the field.

[0070] In the case that the research activity management and application platform includes the knowledge sharing block corresponding to each research unit, the research unit sharing and application tool 13 shown in FIG. 1(b) can be further configured to support the sharing and application of the relevant information and data of the research unit across laboratories or projects, wherein the relevant information and data include the code package, research records and knowledge sharing block of the research unit, and specifically include the following steps:

[0071] In response to a fourth operation (not shown) of the user, a copy of the code package of the upstream research unit is established and applied to the downstream research unit with the same or lower disclosure level, and in the case that the user selects to synchronize the research records, the historical research records of the upstream research unit are linked to the downstream research unit. In this way, even if the upstream project is deleted for some reason, the corresponding research unit in the downstream project can still continue to run without being affected.

[0072] Next, in the case that the disclosure level of the downstream research unit is the same as that of the upstream research unit, the knowledge sharing block of the upstream research unit and the knowledge sharing block of the downstream research unit are synchronized bidirectionally; in the case that the disclosure level of the downstream research unit is lower than that of the upstream research unit, the knowledge sharing block of the upstream research unit is synchronized unidirectionally to the knowledge sharing block of the downstream research unit, and since one upstream research unit can be applied to multiple downstream projects, a radial research unit network can be formed around one research unit in the platform, and the central node is the upstream research unit. Through the bidirectional synchronization of the knowledge sharing block, the application of the research unit gradually evolves from individual behavior to social collaboration behavior. This research unit network naturally evolves into a community with a theme (the theme is the content of the research unit). With the continuous application of the upstream and downstream, the problems about the research unit can be gradually discovered, and the research unit can be continuously improved in actual application through the continuous feedback and optimization of the community. This community-driven mode effectively converts individual experience accumulation into collective wisdom, laying a solid foundation for the long-term development of the research unit.

[0073] Through the above steps, the research unit designed by the user can be applied to different laboratories or different projects in one key in the platform of the embodiment of the present application. In this way, the applicators of the platform can conveniently apply the research units designed and provided by others without having any experience of writing research units, which promotes the global sharing of research units and research schemes. On the other hand, researchers can share their research schemes and data records through the platform, so that other laboratories can easily reproduce these experiments, improve the repeatability of research schemes and research results, and further promote global scientific cooperation through the sharing of scientific experience across laboratories, projects and research units.

[0074] FIG. 1(d) shows another component diagram of the research activity management and application platform supporting multi-disciplinary collaboration according to an embodiment of the present application. The research activity management and application platform 10 can further include an AI system tool 14 based on a large language model on the basis of the component structure shown in FIG. 1(a). FIG. 5(a) shows a chat interface of the AI system tool according to an embodiment of the present application.

[0075] As shown in FIG. 5(a), the AI system tool 14 is configured to include a chat interface 141, which is named, for example, “AI Masterbrain”, and the chat interface 141 can automatically generate a research unit code package for a user upon receiving content and instructions in a user chat conversation, and can further receive relevant materials introduced through “Add context”, and the research unit code package can be revised by the user interacting with the AI system tool 14 in a chat interface 141.

[0076] FIG. 5(b) shows a schematic diagram of customizing research unit syntax checking using the AI system tool according to an embodiment of the present application.

[0077] As shown in FIG. 5(b), the AI system tool 14 further includes a research unit syntax checker 142 for checking whether the code conforms to the custom research unit syntax. Thus, the automatic generation of the research unit code package for the user and the revision of the research unit code package by the user interacting with the AI system tool 14 in the chat interface 141 further includes: in the case where the user provides a protocol document to be converted into a research protocol to the AI system tool 14 in the chat interface 141, injecting the custom research unit syntax and examples of converting reference documents into research protocols as contextual information into the chat conversation, and causing the large language model to perform the following operations: generating a research protocol in the research unit code package based on the chat conversation with contextual information; generating a model of a research unit in the research unit code package based on the generated research protocol and the chat conversation with contextual information; and generating an assigner of a data field in the research unit code package based on the generated research protocol, model and chat conversation with contextual information. In this way, the chat conversation with contextual information is provided to the AI system tool 14, which can help it more accurately determine the data type of the data field, especially certain specific data types that need to be directly read from the original document provided by the user to accurately determine, and on this basis, the syntax part corresponding to the model and the assigner in the research unit syntax of the present application is used to generate the model of the research unit and the assigner of the data field in the research unit code package in turn and more accurately.

[0078] On this basis, the scientific research unit syntax checker 142 is used to perform scientific research unit syntax checking on the scientific research unit code package generated by the large language model, and the syntax checking result is fed back to the large language model, so that the large language model generates a scientific research unit code package again based on the syntax checking result, until the syntax in the generated scientific research unit code package is completely correct.

[0079] Notably, since the platform according to the embodiments of the present application is a framework that is being actively developed, the scientific research unit syntax commonly used in multiple disciplines is also constantly updated. Therefore, in order to ensure that the AI system tool can continue to be compatible with the latest scientific research unit syntax, in the embodiments of the present application, the fine-tuning model strategy is not adopted, but as previously described, the current version of the scientific research unit syntax being used is injected as a situational context into the dialogue, taking advantage of the significant enhancement of the context length of the large language model (such as qwen-long supporting 10,000,000 tokens context, gpt-4o supporting 128,000 tokens context). This method greatly improves the generation efficiency of the scientific research unit, and ensures that the large language model generates content that conforms to the latest scientific research unit syntax on the basis of a comprehensive understanding of the latest scientific research unit syntax.

[0080] In some other embodiments, in the case where the user asks questions or gives modification suggestions for a specified part of the scientific research unit code package in the chat interface 141, the large language model provides an explanation for the user's question or provides a revision scheme that meets the user's modification suggestions for the specified part, realizing the directional optimization of the scientific research scheme. For example, the user can ask questions about the syntax of the scientific research unit, and the system tool can provide detailed answers.

[0081] Through AI-driven scientific research agreement automatic generation, researchers no longer need to manually write complex scientific research agreements, reducing the learning threshold of the scientific research unit framework, and also simplifying the startup process of the scientific research project, significantly reducing the difficulty of getting started with the scientific research unit design. In addition, AI can understand the scientific research needs of different disciplines and generate appropriate scientific research agreements for interdisciplinary projects, which is particularly important for projects involving cooperation of multiple disciplines, improving the coordination and implementation efficiency of interdisciplinary projects.

[0082] In some embodiments, in the case where the user opens the chat interface 141 under a specific scientific research unit and asks questions, the relevant information of the specific scientific research unit is injected as a situational context into the chat dialogue, so that the AI system tool 14 generates an answer associated with the specific scientific research unit in the chat 141, wherein the relevant information of the scientific research unit at least includes the scientific research agreement in the scientific research unit code package, and the content in each knowledge sharing board corresponding to the scientific research unit set according to the public level that can be obtained according to the scientific research unit.

[0083] Unlike general chatbots (such as ChatGPT), the AI-assisted QA function provided by the AI ​​system tool for research units has two unique characteristics: First, it can fully utilize research unit information relevant to the dialogue (such as research agreements); second, it can utilize content from knowledge-sharing sections related to the research unit, as well as knowledge resources across the entire platform. Therefore, the deeply integrated AI system tool can more accurately answer user questions about specific research activity scenarios, enhancing the intelligent support for research data recording and analysis.

[0084] Figure 5(c) illustrates a schematic diagram of injecting relevant information of a research unit as context into a chat dialogue of an AI system tool according to an embodiment of this application. As shown in Figure 5(c), a user raises the following question in a research unit related to protein purification (Purification of Lipoylated H Protein):

[0085] How can I improve the yield?

[0086] Since the AI ​​system tools inject relevant information about the research unit into the dialogue—in this embodiment, the relevant information being the research unit's research protocol—the following response will be given based on this specific scenario:

[0087] To improve the yield in the protein purification experiment, consider the following strategies:

[0088] 1. Optimize Lysis Conditions: Adjust buffer composition, temperature, and duration to maximize protein release.

[0089] 2. Minimize Protein Loss: Use gentle techniques, keep the process cold, and reduce the number of purification steps.

[0090] 3. Optimize Affinity Chromatography: Fine-tune binding and elution conditions specific to your protein’s affinity tag.

[0091] 4. Prevent Proteolysis: Add protease inhibitors and work quickly at cold temperatures to prevent protein degradation.

[0092] 5. Carefully Concentrate and Refold Protein: Use appropriate methods to avoid loss and aggregation during concentration and refolding.

[0093] As can be seen, the above answer is more specific, targeted and in-depth than the general ChatGPT answer shown in Figure 5 (d). This is because the AI system tool injects the research protocol of the research unit into the conversation, which provides clear experimental background information, enabling it to understand that the conversation takes place in the context of a protein purification experiment and infer that "the yield" refers to "the yield in the protein purification experiment", thus giving a more professional and targeted answer. As a result, even if the user has no knowledge of AI background, by simply customizing the research unit, without any secondary development of AI, the AI system tool can automatically provide answers tailored to the scene of the research unit, greatly improving the practicality and convenience of AI in scientific research applications.

[0094] In some embodiments, for example, when the user asks the same question again, the AI system tool can also search the service, dynamically obtain the content in the knowledge sharing board available in the research unit, and apply it to the answer, thereby not only achieving efficient retrieval of research protocols, research plans and research data, but also achieving cross-disciplinary research Q&A, improving the shareability of research plans, data and expert experience within and between laboratories and even globally, weaving the research knowledge carried by the platform into an organic whole, enabling efficient dissemination and application of knowledge in a wide range of research communities, and promoting cross-project, cross-laboratory research cooperation and knowledge sharing.

[0095] In other embodiments, the AI system tool allows users to use buttons such as Add context to select the background knowledge that the AI system tool can apply when answering, for example, when opening the AI system tool in the research protocol interface, the AI can default to injecting all protocol information about the research protocol; In other cases, when the user opens the AI system tool in the knowledge sharing board interface, the default is to inject the research protocol information and knowledge of the knowledge sharing board contained in the research unit. Of course, at this time the user can also configure the scope of context, for example, whether to use all the knowledge of the knowledge sharing board under the project, or under the laboratory, or the knowledge of the publicly accessible knowledge sharing board of the entire platform, etc. In addition, the user can also inject the history into the AI system tool as context, then at this time the AI system tool will be able to consider the history when intelligently analyzing the question.

[0096] FIG. 5(e) shows a schematic diagram of analyzing scientific records by using the AI system tool according to an embodiment of the present application. As shown in FIG. 5(e), in the case that the user opens the chat interface 141 under a specific scientific unit and requests to analyze the scientific records 143 (the numbers 219, 218, 217, etc. shown in the figure are exemplary simplified scientific record IDs, corresponding to 5 scientific records with different unique IDs, which correspond to the No. 5, No. 4, No. 3 scientific records shown in 144), the scientific protocol and the scientific unit data field JSON Schema corresponding to the scientific unit are taken as the situational context, so that the AI system tool 14 generates an analysis report of the scientific records of the specific scientific unit based on the historical scientific records 144 of the specific scientific unit and the situational context. Since the separation of the scientific unit code package and the scientific record is realized in the embodiment according to the present application, that is, if you want to understand any scientific record generated based on the same scientific unit code package, you only need to provide the scientific unit code package and its corresponding data field JSON Schema, therefore, even if multiple scientific records are to be analyzed, as long as they correspond to the same scientific unit code package, the scientific protocol and the scientific unit data field JSON Schema only need to be provided once as the context, without the need for repeated provision, thereby saving resources and being more efficient without losing any effective information. Of course, in the case that a more in-depth analysis and understanding of each scientific record is required, the model and the assigner part of the scientific record can be selectively injected into the AI system tool 14 as the situational context, which is not limited by the present application.

[0097] In other embodiments, the user can also open the chat interface 141 under a specific scientific unit and request to analyze the scientific records with a specified analysis intent, in which case the scientific protocol, the scientific unit data field JSON Schema and the specified analysis intent corresponding to the scientific unit are taken as the situational context, so that the AI system tool 14 generates an answer or an analysis report of the specified analysis intent of the scientific records of the specific scientific unit based on the historical scientific records of the specific scientific unit and the situational context. In addition, the embodiment according to the present application also supports multi-turn dialogue, and when the user proposes further or completely different analysis intents in different turns of dialogue, the analysis results of the previous turn can be provided to the AI system tool 14 as the situational context, so that the AI system tool 14 can perform more in-depth and comprehensive analysis.

[0098] The AI system tool according to the embodiments of the present application can build a bridge between different fields by understanding scientific research protocols and data from different disciplines, and researchers can quickly understand and apply experimental data and methods of other disciplines through the tool, greatly reducing the knowledge barrier between disciplines. In addition, the AI system tool can not only answer questions related to protocols, but also provide suggestions based on existing data to help researchers optimize experimental results, and is a powerful assistant / AI tutor for scientists to improve research efficiency, and can also promote one-key analysis of scientific research data, generation of heuristic suggestions based on AI, and generation of scientific research reports, etc.

[0099] FIG. 1 (e) shows another constituent diagram of the scientific research activity management and application platform supporting multi-disciplinary sharing according to an embodiment of the present application. As shown in FIG. 1 (e), on the basis of FIG. 1 (d), the scientific research activity management and application platform 10 can further include a scientific research process integration tool 15.

[0100] In some embodiments, the scientific research process integration tool 15 can be configured to generate an automatically executable scientific research unit workflow (hereinafter also referred to as RUW) based on a user-given scientific research unit workflow graph (hereinafter also referred to as RUWG), or based on a user-given scientific research unit workflow graph and a workflow graph logic (hereinafter also referred to as RUWGL), using the AI system tool 14, wherein the scientific research unit workflow graph is a directed graph and contains a plurality of scientific research units, and each scientific research unit in the plurality of scientific research units corresponds to the same or different disciplines.

[0101] It is worth noting that the scientific research unit workflow graph in the embodiments of the present application can be a combination of scientific research units in any same or different fields / disciplines in any manner, because the scientific research activity management and application platform according to the embodiments of the present application can include scientific research units from any same or different fields. In other embodiments, it is not necessarily required to design a perfect workflow graph in advance, but rather, according to the characteristics of the target to be achieved, a plurality of scientific research units of interest are selected as nodes in the same scientific research unit workflow graph on demand, and only when necessary, the workflow graph logic is used for constraint and limitation, which is flexible in form. By way of example only, the above-mentioned workflow method can be applied to carbon nanotube material self-dispersion research, new drug discovery, protein engineering, bioengineering fermentation research, chemical synthesis of gold nanoparticles, and single-cell sequencing research, etc. For example, each scientific research unit can represent a step in the same discipline or similar discipline scientific research activity, and a plurality of scientific research units can be combined according to experimental requirements to construct a complex experimental workflow, thereby helping users to realize the automatic management of experimental steps or multiple stages of research projects. By way of example only, the periodic maintenance of a complete set of scientific research instruments can be designed as a workflow, and each scientific research unit in the workflow represents the maintenance work of different scientific research instruments. Through automatic design and automatic execution of the workflow, the electronicization and intelligentization of scientific research instrument maintenance can be realized.

[0102] In other embodiments, each scientific research unit can represent an experimental step in a different discipline, and the above-mentioned flexible scientific research unit collaborative working mode can also empower cross-research of different disciplines, and greatly promote the efficient generation of multi-disciplinary cross / fusion innovation achievements. In addition, the above-mentioned workflow mode can also provide methodological and implementation means support for industrialized scientific research relying on scientific research automation equipment.

[0103] Taking the ultrasonic dispersion research of carbon nanotubes as an example, FIG. 6(a) shows a schematic diagram of a scientific research unit workflow graph according to an embodiment of the present application, and FIG. 6(b) shows a research path (also referred to as a scientific research unit path, hereinafter referred to as RUP) according to an embodiment of the present application.

[0104] In FIG. 6(a) and FIG. 6(b), RU1-RU4 represent four research units respectively, RU1 represents preparing a dispersion liquid from carbon nanotube powder, RU2 represents ultrasonic dispersion, RU3 represents preparing a low-concentration dispersion liquid from a high-concentration carbon nanotube dispersion liquid, and RU4 represents characterization of dispersion. In actual research of ultrasonic dispersion of carbon nanotubes, each of the above research units can be carried out in sequence to achieve a specific research goal. As shown in FIG. 6(b), in research path 1, a high-concentration dispersion liquid is first prepared from carbon nanotube powder (RU1), then the dispersion liquid is dispersed by using ultrasonic technology (RU2), and then the dispersion result is characterized to check the effect of the current dispersion (RU4). The characterization result shows that the current dispersion structure has not yet reached the expected result, so the high-concentration dispersion liquid is prepared into a low-concentration dispersion liquid (RU3), then ultrasonic dispersion is carried out again (RU2), and finally characterization is carried out (RU4). The characterization result reaches the expected result, and the process can be terminated at this time (End).

[0105] However, in actual research, other research paths similar to research path 1 cannot be enumerated. For example, according to the characterization result of the dispersion, it can be determined whether it is necessary to re-disperse by ultrasonic and then characterize (meaning that the process of RU2→RU4 needs to be repeated), or to further dilute the dispersion liquid, then disperse by ultrasonic and then characterize (meaning that the process of RU3→RU2→RU4 needs to be repeated). The above process can be repeated until a satisfactory dispersion result is obtained. Research paths 2-6 show some possible research paths, however, all possible paths cannot be enumerated. Therefore, FIG. 6(a) shows a workflow diagram composed of four research units RU1, RU2, RU3 and RU4. It can be seen that this is a directed graph that can constitute multiple research paths. A user can define such a directed graph to describe a research process, in which each directed edge represents a logical relationship between each RU. Such a directed graph can well handle logical relationships with a cyclic topological structure. The workflow diagram can be regarded as a "path set" composed of all reasonable research paths. It can be seen that research paths 1-6 in FIG. 6(b) all conform to the topological structure of the workflow diagram in FIG. 6(a).

[0106] In order to make the research paths generated based on the workflow diagram conform to the logic of research in this field, the workflow diagram can be supplemented with corresponding workflow diagram logic. In the above embodiment, the workflow diagram logic can include the following four:

[0107] 1. The entire dispersion process must be carried out in a solution system. Preparing a dispersion from a solid powder can only be the first step of an experiment: RU1 must be the starting point of the research path.

[0108] 2. Each dispersion system must go through the preparation, sonication and characterization stages: one research path must include at least one (RU1→ RU2→ RU4) instance, and this order is irreversible.

[0109] 3. Based on the characterization results, it is determined whether: 1) the sample needs to be sonicated again (RU4→ RU2), or, 2) the dispersion solution needs to be further diluted before sonication (RU4→ RU3→ RU2). After repeating either of the two paths, the characterization (RU4) must be performed again to confirm the subsequent results. Based on the results of RU4, either of the two paths can be followed alternately.

[0110] 4. The characterization process (RU4), as the only quality control step in the experiment, can appear in the middle of the steps, but must always be the last step in the research path.

[0111] 5. The research path can be terminated when the characterization results (RU4) meet the research purpose.

[0112] In the process of carrying out scientific research according to the research path generated based on the work flow diagram, the scientific research data obtained can be considered as the scientific research records generated by the scientific research units in the order of connection when carrying out research on the path. Figure 6(c) shows a schematic diagram of scientific research data generated by executing a research path according to an embodiment of the present application. Figure 6(c) shows the scientific research records generated in order by each scientific research unit when executing research path 1 in Figure 6(b), where RUR i represents the scientific research record generated in the i-th step, and the upper index RU j represents that the scientific research record is generated by RU j The above scientific research data can also be represented as a list of scientific research records as shown in Figure 6(c).

[0113] Figure 7 Figure 7 shows a schematic diagram of the process of generating an automatically executable scientific research unit workflow using an AI system tool according to an embodiment of the present application. As shown in Figure 7, the specific process of generating an automatically executable scientific research unit workflow using an AI system tool is as follows. Figure 7

[0114] In step 701, based on the scientific research unit workflow diagram given by the user and the current scientific research intention (hereinafter also referred to as RP) generated by the user given / AI system tool, in combination with the research progress of the previous scientific research unit in the scientific research unit workflow, the AI system tool is used to determine whether there is a feasible research strategy for the current scientific research intention. In the case of "yes" in step 701, go to step 702.

[0115] ​Suppose a user gives any one of the research unit workflow (RUW) defined by the research unit workflow graph (RUWG) and the workflow graph logic (RUWGL), the RUW can be expressed as the following formula (1) - formula (3):

[0116] RUW = (RUWG, RUWGL)(1)

[0117] RUWG = (RUs, RUEs)(2)

[0118] RUs = {RU1, RU2,..., RU m}(3)

[0119] Wherein, RUs in formula (3) represents the set of research units that make up RUWG, and RUEs in formula (2) represents the set of edges in RUWG.

[0120] The work to be completed by the AI system tool can be expressed as the following formula (4):

[0121] (RUW, RP) → automatic process? → RC(4)

[0122] That is, given RUW and RP, can the RUs in the RUW be applied to generate a RUP, and obtain the RUR corresponding to each RU on the RUP, and finally obtain the RC (research conclusion) corresponding to the RP. The above problem can be described as the following formula (5) - formula (8):

[0123] (5)

[0124] (6)

[0125] End(7)

[0126] (8)

[0127] Wherein, n represents the total number of RUs on the research path RUP. represents the i-th RU passed on the path, note that i only represents the position number, not the number of RU. represents the RUR generated by the i-th RU on the path (i.e. ). In this way, a sequence of RURs is obtained, which contains the RURs generated by the RUs corresponding to the first to n steps on the RUP. Thus, as shown in formula (5), by comprehensively analyzing RUW and The information contained in the RUP can be automatically analyzed by AI1 to obtain the RC corresponding to the RP, where AI1 refers to a certain artificial intelligence algorithm trained to solve the above problems, which can be a single deep learning network or composed of multiple sub-networks with different functions, and the present application does not limit this.

[0128] The above process can be further decomposed as follows in equation (9):

[0129] (9)

[0130] In the above process, when a RP is given, we can let the AI system tool target the RP, combine various information of the RUW, and design a feasible research strategy (RS) for the RP. The core purpose of this step is to let the AI system tool comprehensively consider the RUW and the corresponding RP, and carry out the following two key steps: 1. Determine whether the RP can be appropriately solved by some reasonable application of the RUW; 2a. If not, the user should be directly informed that the RP is not suitable for applying the RUW to solve; 2b. If yes, what is the specific RS that should be applied to achieve the RP, and give the specific RS, thereby guiding how to select the appropriate RU for subsequent scientific research. As an example, for the carbon nanotube self-dispersion research related RUW shown in FIG. 6(a) and FIG. 6(b), it is obviously not suitable for the RP of “studying how cells mitosis”, so in the embodiments of the present application, the rationality of the RP can be first judged, and the RP that is not reasonable will be given the conclusion that the RUW cannot support the current RP, i.e. there is no feasible research strategy for the current scientific research intention. In other embodiments, different RPs actually require different research strategies. For example, for the carbon nanotube self-dispersion research related RUW shown in FIG. 6(a) and FIG. 6(b), the two different RPs of “studying how to use m-cresol to disperse carbon nanotubes to an average tube diameter of 20-30 nm under the condition of as little ultrasonic as possible” and “studying how to use m-cresol to disperse carbon nanotubes to an average tube diameter of 20-30 nm under the condition of as little dilution as possible” require different RSs. The generation of the above RS can be achieved by the method represented by equation (10) as follows:

[0131] (10)

[0132] The above formula (10) indicates that the AI method 2 (AI2) is used to automatically determine whether a given RUW can support the RP, and if yes, the RS is given, and if no, the RUP is terminated (End). The AI2 can be an AI method under the same framework as the AI1, or can be an independent AI method dedicated to RS generation, which is not limited in the present application.

[0133] Next, in the case where the AI system tool determines that there is a feasible research strategy for the current research intention, in step 702, the next research unit for carrying out the research is automatically selected in the research unit workflow graph, or the current research path is terminated according to the research progress of each research unit. It can be understood that the current research path will be terminated only in the following two cases: 1. each research unit has completed its own research, and the required research intention has been achieved and the corresponding research conclusion can be obtained; 2. although the RUs involved in the RUW are tried for many times, the target RP cannot be achieved.

[0134] After obtaining the RS, the first RU can be further selected in an automated manner according to the following formula (11): ):

[0135] (11)

[0136] Similarly, the AI3 can be an AI method under the same framework as the AI1 or AI2, or can be an independent AI method dedicated to RU selection, which is not limited in the present application.

[0137] Then, by applying and executing , that is, when the research is completed, the corresponding RUR ( ) can be obtained, and on this basis, the AI system tool can be used to analyze to obtain a stage conclusion ( ). The stage conclusion can include rich meaning dimensions beyond the simple literal meaning "summary", such as: 1) what stage conclusion can the current RUR provide for the RP; 2) whether it is sufficient to respond to the RP; 3) if it is not sufficient, whether there is any further planning for the subsequent research; 4) due to the great unknown and uncertainty of the research, the AI system tool can also reflect the rationality of the RS at this time, and if it is not reasonable, how to make appropriate adjustments; 5) whether abnormal / special phenomena and / or abnormal / special data are found in the research process, etc., which are not listed here. The process can be represented by the following formula (12):

[0138] (12)

[0139] Similarly, AI4 can be an AI method under the same framework as AI1, AI2 or AI3, or an independent AI method specially used for generating RC, which is not limited in the present application.

[0140] In the subsequent steps, the process of formula (11) - formula (12) can be repeated to select each RU on the RUP in turn until the end (End). The general representation of the process is shown in formula (13) - formula (15):

[0141] (13)

[0142] (14)

[0143] (15)

[0144] In formula (13) - formula (15), when i = 1, , , formula (14) and formula (11) are consistent, and formula (15) and formula (12) are consistent.

[0145] In the embodiments of the present application, when there is a feasible research strategy, after the scientific research unit at each step in the research path completes the scientific research, the AI system tool can be used to automatically generate a stage conclusion based on the historical scientific research records of the research path, and automatically select the next scientific research unit to carry out scientific research in the scientific research unit workflow graph considering the stage conclusion.

[0146] In some embodiments, the stage conclusion may, for example, include whether the historical scientific research record data can meet the current scientific research intention, whether the current scientific research strategy is effective, whether it is necessary to correct or optimize according to the real scientific research data, and the suggestion of the next step, etc., or other results based on certain evaluation standards to evaluate the previous scientific research process or scientific research results, which are not limited in the present application. That is, the scientific research unit to carry out scientific research in the next step is not selected based on a static strategy, but is associated with the stage conclusion formed by the actual scientific research records of the current scientific research unit and even the previous scientific research process. In this way, not only can a scientific research unit workflow graph designed or given by any user be automatically generated in a general way to form a research path that can be automatically executed, but also the scientific research process can be dynamically adjusted to make the entire scientific research process close to optimal, which can promote the efficient execution of iterative optimization type scientific research.

[0147] Next, in step 703, after the current research path ends, the AI system tool automatically generates a scientific research conclusion for the scientific research unit workflow graph and the current scientific research intention given by the user based on the historical scientific research records of the current research path.

[0148] Figure 10 This diagram illustrates scientific research conclusions automatically generated by AI system tools according to an embodiment of this application. Figure 10 As shown, research conclusions can include multiple aspects. For example, they can include whether the research conducted at each node in the workflow diagram has achieved the research intent given by the user. Alternatively, they can include whether any special phenomena, data, or process discoveries worthy of further research have been observed in the entire research path history. Or, they can include suggestions for potential new research directions and strategy optimizations discovered in the future, etc. This application does not limit these aspects.

[0149] In other words, during the generation and execution of research pathways, RC sequences are also generated. (n represents the total number of times the RU was applied in the RUP). These intermediate conclusions can also help generate the final scientific research conclusions. Therefore, equation (5) can be extended to the following equation (16):

[0150] (16)

[0151] In some embodiments, since an RU is essentially defined as a research scheme with adjustable space, this adjustable space actually comes from the data fields defined in the research protocol of the RU. For example, in the ultrasonic dispersion RU in Figures 6(a) and 6(b), a data field related to ultrasonic time is defined in its research protocol. That is, if you want to apply the RU, you need to know the value of the data field first, and then you can obtain the corresponding research record. Usually, this part of the data field can be called parameter data field (while others can be called feedback data field). Therefore, while automatically selecting the research unit to carry out the next research research in the research unit workflow diagram, it is also necessary to further use AI system tools to automatically design research parameters that meet the current research intention for the research unit that should carry out the next research research. The predicted research parameters will be assigned to the respective parameter data fields defined in the research protocol of the research unit, so that the research unit can carry out research when the parameters are determined. The research parameters that meet the current research intention can be automatically derived using AI system tools as follows (17):

[0152] (17)

[0153] in, Representative research record The collection of all data fields in ( In the above-mentioned formula, the set of parameter class data fields; AI5 can be an AI method under the same framework as AI1, AI2, AI3, or AI4, or an independent AI method dedicated to deriving the value of the parameter class data field, which is not limited in the present application.

[0154] Thus, in the embodiments according to the present application, it is first creatively implemented in a general way to utilize a series of AI automation methods such as AI1-AI5 in the AI system tool to cooperate with the application and development of each RU, and AI scientific research automation under any given workflow graph (or workflow graph + workflow graph logic) and scientific research intention.

[0155] In other embodiments, the scientific research process integration tool can be further configured to integrate the generated automatically executable scientific unit workflow into a single scientific unit code package. Specifically, for example, each scientific unit to which the scientific unit workflow will be applied can be defined in any scientific project first. Notably, these scientific units should be defined under the same scientific project in the same laboratory. Since, as previously described, the scientific unit workflow is essentially to obtain a list of scientific records related to the scientific unit workflow, the user can define the scientific unit workflow to a workflow type scientific unit variable with an ID such as cnt_dispersion in the scientific protocol. In this way, a scientific unit workflow can be integrated into a single scientific unit code package, which can be shared with all users worldwide in the same way as the scientific unit code package.

[0156] As the number of institutions and laboratories using the scientific activity management and application platform supporting multi-disciplinary sharing according to the embodiments of the present application increases, as many scientific activities as possible in each laboratory will be efficiently digitized and electronicized, which will provide a hand of data feed for future AI scientific models. These scientific data, whether applied by these laboratories themselves or cooperated with AI researchers / laboratories through data cooperation, will become the feed for the next generation of artificial intelligence. Thus, it will promote data-intelligence spiral progress, the formation of a global cross-disciplinary unified scientific community and social network, global scientific equality, rapid sharing of scientific data, circulation of scientific data as asset elements, global scientific cooperation and division of labor, and scientific automation.

[0157] In addition, the scientific research activity management and application platform according to the embodiments of the present application is also applicable to various scenarios with potential customized record needs, and can be directly applied to an electronic medical record scenario, for example. Similar to the scientific research described in other embodiments of the present application, in a medical / hospital / clinical scenario, each institution / department has a demand for medical data records (such as electronic medical records), and the content and type specifications of the data required to be recorded are different in different departments, different diseases, and different medical scenarios. It can be thought that if each department records in a self-defined manner, it is not only inefficient, but also not conducive to sharing and experience accumulation. With the aid of the scientific research activity management and application platform in the embodiments of the present application, users in the professional field can customize the protocol, model and assignor of the medical record according to the scientific research unit syntax, tailor the medical record mode of the professional field according to the specific needs, and the scientific research unit code package defined can be widely shared in different departments. In this way, once designed, it can be applied in the hospital / cross-hospital to realize the unification and standardization of specific medical records, so the present application has great application potential and significant economic benefits in multiple industries.

[0158] In addition, although the exemplary embodiments have been described herein, the scope of their protection is to be understood as including any and all equivalents based on the same basic concept embodied in the application. The language of the claims is to be interpreted in the broadest reasonable manner, considering the context of the patent to which that claim is attached and the context of prior art to which it may be relevant. The claims should be interpreted, and the specification deemed, to cover any adjacencies or equivalents of the elements described in the claims that would be apparent to one of ordinary skill in the art in view of the prior art and this specification. Accordingly, the claims should not be limited to the specific embodiments described herein, but should be given the full scope consistent with the patent to which they belong.

[0159] The above description is intended to be illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) can be used in combination with each other. Other embodiments will be apparent to those of ordinary skill in the art upon reviewing the above description. Additionally, in the specific description of embodiments above, various features can be grouped together or divided up so that the novelty disclosed herein can be leveraged independently. This should not be interpreted as a requirement that the claimed subject matter must claim all combinations of features. Rather, the subject matter claimed herein is a subset of the disclosed subject matter. Thus, the claims are to be interpreted not as including all possible combinations of features that can be leveraged independently, but rather as including only those combinations that are explicitly set forth in the claims. The scope of the disclosure is to be determined solely by the claims that follow, along with the full scope of equivalents to which such claims are entitled. The claims are to be interpreted broadly, in accordance with the principles of patent law.

[0160] The above examples are only exemplary embodiments of the present application, and are not intended to limit the present application, and the protection scope of the present application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements to the present application within the spirit and protection scope of the present application, and such modifications or equivalent replacements are also considered to fall within the protection scope of the present application.

Claims

1. A platform for supporting management and application of research activities shared by multiple disciplines, characterized in that, The scientific research activity management module deployed in the cloud is configured to enable a user to manage scientific research activities by running the scientific research activity management module without additional software installation; The scientific research activity management module includes a scientific research unit design environment and scientific research unit code packages of different disciplines designed and generated by a user based on the scientific research unit design environment and a multidisciplinary scientific research unit syntax defined by the user, wherein each scientific research unit code package at least contains a scientific research protocol of a corresponding discipline, and the scientific research protocol is used to define basic information, multi-modal information and data fields of a scientific research experiment of the discipline; The scientific research activity management and application platform further includes an AI system tool and a scientific research process integration tool configured to generate an automatically executable scientific research unit workflow based on a user-given scientific research unit workflow graph or a user-given scientific research unit workflow graph and workflow graph logic using the AI system tool, wherein the scientific research unit workflow graph is a directed graph and contains a plurality of scientific research units, and each scientific research unit in the plurality of scientific research units corresponds to the same or different disciplines; The AI system tool is used to generate an automatically executable scientific research unit workflow, which specifically includes: based on a user-given scientific research unit workflow graph and a user-given current scientific research intention, in combination with research progress of a previous scientific research unit in the scientific research unit workflow, the AI system tool is used to determine whether there is a feasible research strategy for the current scientific research intention; in the case of a feasible research strategy: automatically selecting a scientific research unit for next step of scientific research in the scientific research unit workflow graph, or terminating the current research path according to the research progress of each scientific research unit; after the current research path ends, the AI system tool is used to automatically generate a scientific research conclusion for the scientific research unit workflow graph and the user-given current scientific research intention based on historical scientific research records of the current research path.

2. The scientific research activity management and application platform according to claim 1, characterized in that, The scientific research unit design environment is further configured to enable the scientific research unit code package of the corresponding discipline designed by the user to contain a model of the scientific research unit, which is used to define type constraints and / or numerical verification relationships of the data fields, and combination verification relationships of types and / or numerical values between the data fields; wherein The type constraints include constraints on the use of pre-defined multi-modal information by the corresponding data field during data entry, and the pre-defined multi-modal information includes one or more of text, image, video, audio and file; The numerical verification relationship includes constraints on the specified mode followed by the corresponding data field during data entry and / or the pre-set value range not being exceeded; The combination verification relationship includes constraints on the types and / or numerical values of the data fields satisfying predetermined constraint relationships.

3. The platform for management and application of scientific research activities according to claim 2, characterized in that, The scientific research unit design environment is further configured to enable the scientific research unit code package of the corresponding discipline designed by the user to contain an assigner of the data fields, which is used to assign values to the data fields based on a data field dependency graph and an assignment rule, wherein The data field dependency graph is a single-level or multi-level directed acyclic graph, and the assignment relationship between each data field is single dependency or multi-dependency, and each data field is assigned by at most one assigner.

4. The scientific research activity management and application platform according to any one of claims 1-3, characterized in that, The scientific research unit syntax provides a template syntax of the data field for the user, so that the user generates a scientific research unit variable, or a scientific research unit step, or a scientific research unit checkpoint based on the template syntax of the data field.

5. The scientific research activity management and application platform according to claim 3, characterized in that, The scientific research activity management module further comprises a scientific research unit record environment, which is configured to: convert the model of the scientific research unit into a data field JSON Schema; based on the data field JSON Schema, automatically generate a structured storage scheme for the scientific research unit, so that when the user submits a scientific research record based on the scientific research unit, and each data field meets the constraint and check relationship of the model, the scientific research record is stored as a JSON that meets the corresponding structured storage scheme of the scientific research unit.

6. The scientific research activity management and application platform according to claim 5, characterized in that, The scientific research unit record environment is further configured to: based on the scientific research agreement and the data field JSON Schema, automatically generate a corresponding scientific research unit record interface for the scientific research unit, which has a stylized style based on the analysis of the custom scientific research unit syntax and interactive controls corresponding to the data types of each data field based on the analysis of the data field JSON Schema.

7. The scientific research activity management and application platform according to claim 1, characterized in that, The scientific research activity management module further comprises a scientific research report generation and reader, which is configured to: in response to a first operation of the user on the scientific research unit, based on the historical scientific research record of the scientific research unit, locally run and automatically generate a formatted scientific research report of the scientific research unit, and enable the user to locally read the generated formatted scientific research report.

8. The scientific research activity management and application platform according to any one of claims 1-3, characterized in that, The scientific research activity management and application platform further comprises a scientific research unit sharing and application tool, which is configured to: generate a unique ID number for each scientific research record of the scientific research unit, and associate the unique ID number with the record time, so that each scientific research record has non-tamperability; when the user wants to modify based on the submitted scientific research record, generate a copy of the scientific research record for the user, and generate a new unique ID number for the copy of the scientific research record and associate it with the modification time, so that the user modifies on the copy with the new unique ID number.

9. The scientific research activity management and application platform according to claim 8, characterized in that, The scientific research unit sharing and application tool is further configured to: in response to a second operation of a user with corresponding permissions on the scientific research unit, label the levels of the scientific research unit, the levels are at least divided into laboratory layers and project layers from high to low, and provide user access control mechanisms corresponding to public levels for each level and / or each scientific research unit; and set a level label corresponding to the level division of the scientific research unit for each user.

10. The scientific research activity management and application platform according to claim 9, characterized in that, Providing corresponding user access control mechanisms for each level and / or each scientific research unit specifically includes: setting the accessible permission of each laboratory layer to be open to all users; setting the accessible permission of each project layer to be for specific level marked users; In the case that a specific user has access rights to a specific project, the specific user has access rights to all research units in the specific project.

11. The scientific research activity management and application platform according to claim 10, characterized in that, Setting access rights for users of a specific level for each project level specifically includes: setting all users to have access rights to the project level; or, only users with the same laboratory level marker have access rights to the project level; or, only users with the same project level marker have access rights to the project level.

12. The scientific research activity management and application platform according to claim 9, characterized in that, Providing corresponding user access control mechanisms for each level and / or each research unit further includes: In response to a third operation of a user with corresponding rights, setting group markers for users with the same laboratory level marker, and correspondingly setting access rights to a specific project level for each group marker user.

13. The platform for management and application of scientific research activities according to any one of claims 1-3, characterized in that, The research activity management and application platform further includes a knowledge sharing block corresponding to each research unit, so that the maintainer or other interested users of the research unit can provide or obtain knowledge associated with the research activity of the research unit in the knowledge sharing block in the form of questions and answers.

14. The scientific research activity management and application platform according to claim 13, characterized in that, The research activity management and application platform is further configured to use the content in the knowledge sharing block to update the research agreement of the corresponding research unit.

15. The scientific research activity management and application platform of claim 9, wherein, The research activity management and application platform further includes a knowledge sharing block corresponding to each research unit, so that the maintainer or other interested users of the research unit can provide or obtain knowledge associated with the research activity of the research unit in the knowledge sharing block in the form of questions and answers. The research unit sharing and application tool is further configured to support cross-laboratory or cross-project sharing and application of relevant information and data of research units, wherein the relevant information and data include code packages, research records, and knowledge sharing blocks of research units, specifically including: In response to a fourth operation of a user, establishing a copy of the code package of the upstream research unit, and applying the copy of the code package of the upstream research unit to the downstream research unit with the same or lower disclosure level, and linking the historical research records of the upstream research unit to the downstream research unit if the user chooses to synchronize the research records; In the case that the disclosure level of the downstream research unit is the same as that of the upstream research unit, the knowledge sharing block of the upstream research unit and the knowledge sharing block of the downstream research unit are synchronized bidirectionally; in the case that the disclosure level of the downstream research unit is lower than that of the upstream research unit, the knowledge sharing block of the upstream research unit is synchronized to the knowledge sharing block of the downstream research unit unidirectionally.

16. The platform for management and application of scientific research activities according to any one of claims 1-3, characterized in that, The AI system tool takes a large language model as the core, and is configured to include a chat interface and can automatically generate a research unit code package for a user and revise the research unit code package through the user's interaction with the AI system tool in the chat interface.

17. The scientific research activity management and application platform according to claim 16, characterized in that, The AI system tool is further configured to include a research unit syntax checker for checking whether the code conforms to a custom research unit syntax; The AI system tool is further configured to include a research unit syntax checker for checking whether the code conforms to a custom research unit syntax; The AI system tool is further configured to include a research unit syntax checker for checking whether the code conforms to a custom research unit syntax; In the case that the user provides a protocol document to be converted into a scientific research agreement to the AI system tool in the chat interface, the custom scientific research unit syntax and examples of conversion from reference documents to scientific research agreements are injected as situational context into the chat dialogue, and the large language model is caused to: generate a scientific research agreement in the scientific research unit code package based on the chat dialogue with situational context; generate a model of the scientific research unit in the scientific research unit code package based on the generated scientific research agreement and the chat dialogue with situational context; and generate an assigner of data fields in the scientific research unit code package based on the generated scientific research agreement, model and chat dialogue with situational context; The scientific research unit syntax checker is used to check the scientific research unit syntax of the scientific research unit code package generated by the large language model, and the syntax checking result is fed back to the large language model, so that the large language model generates the scientific research unit code package again based on the syntax checking result until the syntax of the generated scientific research unit code package is completely correct; In the case that the user asks questions or gives modification opinions for a specified part in the scientific research unit code package in the chat interface, the large language model provides explanations for the user's questions or revision schemes for the specified part.

18. The scientific research activity management and application platform according to claim 16, characterized in that, The AI system tool is further configured to: In the case that the user opens the chat interface under a specific scientific research unit and asks questions, the relevant information of the specific scientific research unit is injected as situational context into the chat dialogue, so that the AI system tool generates answers associated with the specific scientific research unit in the chat interface, wherein the relevant information of the scientific research unit at least includes the scientific research agreement in the scientific research unit code package and the content in each knowledge sharing board corresponding to the scientific research unit that can be obtained according to the public level of the scientific research unit.

19. The scientific research activity management and application platform of claim 16, wherein, The AI system tool is further configured to: In the case that the user opens the chat interface under a specific scientific research unit and requires analysis of scientific research records, the scientific research agreement corresponding to the scientific research unit and the scientific research unit data field JSON Schema are taken as situational context, so that the AI system tool generates an analysis report of the scientific research records of the specific scientific research unit based on the historical scientific research records of the specific scientific research unit and the situational context; Or, In the case that the user opens the chat interface under a specific scientific research unit and requires analysis of scientific research records with a specified analysis intent, the scientific research agreement corresponding to the scientific research unit, the scientific research unit data field JSON Schema and the specified analysis intent are taken as situational context, so that the AI system tool generates an answer or an analysis report of the scientific research records of the specific scientific research unit with the specified analysis intent based on the historical scientific research records of the specific scientific research unit and the situational context.

20. The platform for management and application of scientific research activities according to claim 1, characterized in that, The automatic selection of the scientific research unit for which the next step of scientific research should be carried out in the scientific research unit workflow further includes automatically designing scientific research parameters that meet the current scientific research intent for the scientific research unit for which the next step of scientific research should be carried out.

21. The platform for management and application of scientific research activities according to claim 1, characterized in that, In the case that there is a feasible research strategy, the automatic selection of the scientific research unit for which the next step of scientific research should be carried out in the scientific research unit workflow further includes: In the presence of a feasible research strategy: after the completion of the current research unit research, using the AI system tool, based on the historical research records of the research path, automatically generate a phased conclusion, and automatically select the next research unit to carry out research in the research unit workflow graph considering the phased conclusion.

22. The platform for management and application of scientific research activities according to claim 1, characterized in that, The research process integration tool is further configured to: Integrate the generated automatically executable research unit workflow into a single research unit code package.

Images