Cross-domain shared biological information metadata integration and management platform

Abstract

The invention provides a cross-domain shared biological information metadata integration and management platform, and belongs to the technical field of bioinformatics. Comprising a metadata acquisition module used for supporting uploading of various metadata file formats, recording data change history and providing field preview and condition prompt; the metadata standardization and semantic mapping module is used for realizing field semantic analysis, standardized mapping and user-defined mapping management through a built-in biological ontology library; the universal metadata model building module is used for providing a dual-interface definition metadata structure template and realizing model version control and field verification; the authority control and sharing mechanism module is used for setting a data visibility authority and distributing a metadata unique reference identifier; the advanced retrieval and semantic search module is used for supporting condition combination retrieval and natural language semantic association retrieval; and a database. According to the method, standardized integration, safe sharing and semantic retrieval of the multi-source heterogeneous metadata are realized, the cross-laboratory cooperation efficiency is improved, and the scientific research cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of bioinformatics, and in particular to a cross-domain shared biological information metadata integration and management platform. Background Art

[0002] In the field of life science research, the scale and complexity of bioinformatics data are growing at an unprecedented rate. Metadata, which describes key information such as the characteristics, sources, and processing of biological data, plays an indispensable role in scientific research. However, bioinformatics metadata currently suffers from numerous problems that seriously hinder the development of scientific research, including:

[0003] 1. Data dispersion and heterogeneity: Different laboratories operate independently in recording sample information, designing and planning experiments, and setting sequencing conditions, lacking unified standards and specifications. This results in a high degree of heterogeneity in the format, structure, and semantic expression of bioinformatics metadata. For example, for the same experimental sample, some laboratories may record relevant information in simple text form, while others may use complex tabular formats. Different laboratories also use different terminology when describing the sample tissue source, making it difficult to directly integrate and reuse data.

[0004] 2. Limitations of existing platforms: While existing public databases provide some support for the storage and sharing of biological data, they primarily focus on uploading and archiving structured data. In daily scientific research, the large amount of semi-structured metadata, such as experimental descriptions and research plans recorded in Excel spreadsheets, is difficult to effectively share and standardize on these platforms. This is especially true for comprehensive research across laboratories, projects, and species. Existing data management methods significantly restrict the efficiency of collaboration among researchers, prolonging research cycles and increasing research costs.

[0005] Therefore, it is of great practical significance to develop a cross-domain shared bioinformation metadata integration and management platform that can solve the problems of unified modeling and efficient reuse of multi-source, multi-format, and multi-structure bioinformation metadata. Summary of the Invention

[0006] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide a cross-domain shared bioinformatics metadata integration and management platform, which realizes the standardized integration, secure sharing and semantic retrieval of multi-source heterogeneous metadata, improves the efficiency of cross-laboratory collaboration and reduces scientific research costs.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A cross-domain shared biological information metadata integration and management platform, including:

[0009] a metadata collection module for supporting uploading of multiple metadata file formats, recording data change history, and providing field preview and condition tips;

[0010] a metadata standardization and semantic mapping module for realizing field semantic analysis, standardization mapping, and custom mapping management through an embedded biological ontology library;

[0011] a general metadata model construction module for providing graphical and tabular interface definition of metadata structure templates, realizing model version control, and field verification;

[0012] a permission control and sharing mechanism module for setting data visibility permissions and assigning metadata unique reference identifiers;

[0013] an advanced search and semantic search module for supporting conditional combination search and natural language semantic association search;

[0014] a database for storing semi-structured metadata using MongoDB or Elasticsearch.

[0015] Preferably, the metadata collection module comprises:

[0016] a file format support unit for supporting uploading of multiple common cloud data file formats such as CSV, Excel, and JSON;

[0017] a data version management unit for assigning a unique version number to each uploaded file to record change history;

[0018] a field preview and condition tips unit for providing field preview function before uploading metadata files and informing users of filling conditions and restriction rules for each field.

[0019] Preferably, the metadata standardization and semantic mapping module comprises:

[0020] an embedded biological ontology library unit for mapping non-standard fields uploaded by users to standard biological concepts and identifying aliases to provide semantic support for field standardization and semantic matching;

[0021] a field recommended mapping unit for automatically identifying structured fields from user-uploaded files and recommending standard terms through semantic comparison with the embedded biological ontology library unit;

[0022] a user custom mapping unit for supporting custom mapping of private fields to public standards.

[0023] Preferably, the general metadata model construction module comprises:

[0024] The metadata model definition unit is used to provide two intuitive interfaces, graphical and tabular, to facilitate users to define metadata structure templates;

[0025] Model version control unit, which is used to generate a new version number each time the field structure is modified and associate it with the metadata record;

[0026] The model validation unit is used to check field missing, data type matching, and ontology association accuracy when uploading data.

[0027] Preferably, the metadata model definition unit supports a tree-structured setting of field levels, allowing the definition of an inclusion relationship from sample to tissue to cell type.

[0028] Preferably, the authority control and sharing mechanism module includes:

[0029] A data permission granularity control unit is used to set data visibility to be visible only to oneself, shared within the organization, or open to the entire platform;

[0030] The reference identifier allocation unit is used to generate a unique reference identifier in the form of DOI or UUID for each metadata record.

[0031] Preferably, the authority control and sharing mechanism module further includes a tag management unit for adding classification tags to metadata to achieve efficient retrieval and collaborative research of data.

[0032] Preferably, the advanced search and semantic search module includes:

[0033] Conditional search unit, used to support combined search based on multiple dimensions of conditions such as species, tissue, disease, platform, and experimental type;

[0034] The natural language retrieval unit is used to introduce natural language processing technology to realize natural language query and semantic association functions.

[0035] Preferably, the database utilizes the scalability and natural language retrieval support capabilities of MongoDB or Elasticsearch to achieve efficient storage and indexing of semi-structured data.

[0036] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0037] (1) The present invention supports uploading metadata files in multiple formats, such as CSV / Excel / JSON, and combines the semantic mapping and field standardization mechanism of built-in biological ontology libraries such as Uberon / NCIT to achieve unified conversion of multi-source heterogeneous metadata into a universal bioinformatics model. Compared with the existing technology with the problems of inconsistent data formats and serious semantic ambiguity, the integration efficiency and reuse capability of cross-laboratory metadata are significantly improved.

[0038] (2) The present invention supports users to customize metadata structure templates that meet scientific research needs through a graphical metadata model construction interface and field level setting function, and ensures the traceability of the model through a version control mechanism. Compared with the existing platform's lack of a unified modeling tool, the present invention realizes the standardized definition and flexible evolution of the metadata structure, and improves the standardization of data migration.

[0039] (3) The present invention realizes the secure sharing and traceability management of metadata through the three-level permission granularity control of "visible only to oneself / shared within the organization / open to all platforms" combined with the DOI / UUID unique reference identification and labeling mechanism. Compared with the problem that traditional metadata is limited to internal use and lacks reference standards, it breaks the barriers to cross-institutional collaboration and improves the sharing efficiency and credibility of scientific research data.

[0040] (4) The present invention uses a search engine that combines natural language processing technology with ontology reasoning to support conditional combination retrieval and semantic association in dimensions such as "species / tissue / disease". Compared with the limitation of traditional keyword matching retrieval that does not support semantic extension, it realizes accurate and intelligent discovery of metadata and improves the retrieval efficiency and association mining capabilities of scientific research data.

[0041] (5) The present invention adopts MongoDB / Elasticsearch as a semi-structured data storage solution, and takes advantage of its scalability and natural language retrieval support features to achieve efficient storage and indexing of large-scale biological metadata. Compared with the problem that traditional databases lack support for unstructured data, it meets the storage requirements of large-volume and diverse-structured biological data and improves the scalability and retrieval performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 This is a schematic diagram of a cross-domain shared biological information metadata integration and management platform of the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] like Figure 1 As shown, the present invention provides a cross-domain shared biological information metadata integration and management platform, including:

[0047] Metadata collection module, which supports uploading metadata files in various formats, records data change history, and provides field previews and condition prompts;

[0048] Metadata standardization and semantic mapping module, which is used to implement field semantic parsing, standardized mapping, and custom mapping management through the built-in biological ontology library;

[0049] A general metadata model building module that provides graphical and tabular interfaces for defining metadata structure templates, implementing model version control, and field validation.

[0050] The permission control and sharing mechanism module is used to set data visibility permissions and assign unique metadata reference identifiers;

[0051] Advanced search and semantic search module, used to support condition combination search and natural language semantic association search;

[0052] A database that uses MongoDB or Elasticsearch to store semi-structured metadata.

[0053] The metadata collection module includes:

[0054] The file format support unit is used to support the upload of multiple common cloud data file formats such as CSV, Excel, and JSON; it fully takes into account the differences in data recording and storage habits among different laboratories, and minimizes the user's data upload threshold.

[0055] The data version management unit assigns a unique version number to each uploaded file to record the change history. It also automatically assigns a unique version number to each uploaded metadata file, detailing the data change history. This not only helps users trace the evolution of data but also ensures that the original data is not lost when it is updated or modified, facilitating data comparison and analysis for researchers.

[0056] The Field Preview and Condition Prompts section provides a preview of fields before uploading metadata files and informs users of the entry conditions and restrictions for each field. For example, it provides information about whether a field can be left blank, data type requirements (such as structured data or text data), and other information. This helps users check and prepare data in advance, reducing upload errors.

[0057] Secondly, the metadata standardization and semantic mapping module includes:

[0058] The built-in biological ontology library unit is used to map non-standard fields uploaded by users to standard biological concepts and identify aliases, providing semantic support for field standardization and semantic matching. The built-in biological ontology library covers a wide range of concepts and terms in the biomedical field, providing a solid foundation for field standardization and semantic matching. The platform uses these ontologies to deeply analyze and match user-uploaded fields, converting ambiguous and non-standard fields into standard biological concepts to ensure data consistency and understandability.

[0059] The field recommendation mapping unit is used to automatically identify structured fields and recommend standard terms from user-uploaded files by performing semantic comparisons with built-in biological ontology library units. Specifically, it has an intelligent field automatic recognition function that can accurately determine structured fields from user-uploaded files and infer their meanings. By performing semantic comparisons with the ontology library, the most appropriate standard term is recommended for each field, generating detailed field mapping suggestions. For example, when the "tumor_type" field is detected, the platform automatically recommends mapping it to "Disease"; for the "platform" field, it recommends mapping it to "Sequencing Platform", improving the efficiency and accuracy of data standardization.

[0060] The user-defined mapping unit supports customizing templates for mapping private fields to public standards. Taking into account the diversity and specificity of scientific research, users can define templates for mapping private fields to public standards based on their needs. This feature allows users to flexibly handle special data fields while adhering to common standards to meet personalized research needs.

[0061] In this embodiment, the general metadata model construction module includes:

[0062] The metadata model definition unit provides intuitive graphical and tabular interfaces to facilitate user definition of metadata structure templates. During template definition, users can set detailed attributes such as field name, field type (e.g., string, enumeration, etc.), whether it is required, whether it supports multiple values, whether it is associated with a controlled dictionary, and whether it is structured data. At the same time, hierarchical settings of fields are supported, such as the hierarchical structure from sample to organization to cell type, ensuring that the data model accurately reflects complex biological information relationships. The metadata model definition unit supports tree structure settings of field hierarchy, allowing the definition of inclusion relationships from sample to organization to cell type.

[0063] The model version control unit generates a new version number for each modification of the field structure and associates it with the metadata record. Specifically, each time the field structure is modified, the platform automatically generates a new version number. Old data still retains its original structure information, and newly uploaded data is processed according to the new template. Each metadata record is accompanied by the version number of the associated model, facilitating accurate identification and processing during data parsing and migration, ensuring data integrity and traceability.

[0064] The model verification unit checks for missing fields, data type matching, and ontology association accuracy during data upload. Specifically, during metadata upload or filling, the platform strictly verifies the fields based on the current metadata model. The check includes whether the fields are missing, whether the data types match, whether the multi-value format is correct, and whether the ontology association is accurate. Once problems are found, the user is prompted to repair or complete them in a timely manner, ensuring the quality and standardization of uploaded data.

[0065] In this embodiment, the permission control and sharing mechanism module includes:

[0066] The data permission granularity control unit sets the data visibility to be visible only to oneself, shared within the institution, or open to the entire platform. Specifically, flexible permission setting options are provided, and users can set the data visibility to be visible only to oneself, shared within the institution, or open to the entire platform according to their needs. This granular permission control method can protect users' private data and sensitive information while promoting scientific research collaboration within the institution and across the platform.

[0067] The reference identifier allocation unit generates a unique reference identifier in the form of DOI or UUID for each metadata record. For example, DOI (Digital Object Identifier) or UUID (Universally Unique Identifier). These identifiers enable other researchers to conveniently reference and trace the source of the metadata, enhancing the credibility and traceability of the data, and also helping to standardize the citation behavior of scientific research data.

[0068] In addition, the authority control and sharing mechanism module also includes a tag management unit for adding classification tags to metadata to achieve efficient data retrieval and collaborative research.

[0069] In this embodiment, the advanced search and semantic search modules include:

[0070] The conditional search unit is used to support combined searches based on multiple dimensions of conditions such as species, tissue, disease, platform, and experimental type. Researchers can quickly filter out metadata that meets specific conditions based on research needs, thereby improving the accuracy and efficiency of data retrieval.

[0071] The natural language retrieval unit uses natural language processing technology to implement natural language queries and semantic associations. For example, when a user enters "cancer tissue sample," the platform automatically matches tumor-related tags and metadata records, providing a more convenient and intelligent search experience.

[0072] In this example, the database leverages the scalability and natural language search support of MongoDB or Elasticsearch to efficiently store and index semi-structured data. These two databases offer excellent scalability and support for natural language search, enabling efficient storage and management of biometric metadata within the platform, meeting the needs of large-scale data storage and rapid retrieval.

[0073] The cross-domain shared bioinformation metadata integration and management platform operates on the principle of multi-module collaboration to achieve standardized management and secure sharing of multi-source heterogeneous data. The platform first accepts file uploads in formats such as CSV, Excel, and JSON through the metadata collection module. It automatically assigns a unique version number to each upload to record change history. Before uploading, a field preview function prompts users to check field filling conditions, such as whether to allow nulls and data type requirements, to ensure the standardization of raw data collection.

[0074] After the data enters the platform, the metadata standardization and semantic mapping module uses built-in biological ontology libraries such as Uberon, NCIT, and DOID to perform in-depth semantic analysis of the fields uploaded by users. It converts non-standard fields into standard biological concepts through semantic comparison algorithms. For example, it automatically recommends mapping "tumor_type" to "Disease" and "platform" to "Sequencing Platform". It also supports users to customize mapping templates from private fields to public standards to solve the problem of data semantic heterogeneity. The general metadata model construction module provides both graphical and tabular interfaces. Users can define metadata templates that include field name, type, mandatory or not, multi-value support, controlled dictionary association, and hierarchical structure (such as sample → tissue → cell type). Each time the field structure is modified, a new version number is automatically generated and associated with the metadata record. During upload, the model verification unit checks for field missing, data type matching, and ontology association accuracy to ensure the standardization and traceability of the data structure.

[0075] The storage phase uses MongoDB or Elasticsearch databases, leveraging their scalability and natural language retrieval support to achieve efficient storage and indexing of semi-structured metadata. The permission control and sharing mechanism module implements three levels of permission granularity control: "visible only to the user, shared within the organization, and open to all platforms." Combined with DOI or UUID unique reference identifiers and tag management units, it enables secure data sharing and classified retrieval. The advanced retrieval module integrates conditional combination retrieval based on species, tissue, disease, and other dimensions with natural language processing technology. When a user enters "cancer tissue sample," the system expands the term through ontology reasoning and matches relevant tags, enabling semantic associative retrieval. This ultimately forms a closed-loop management process from data collection to retrieval application, improving the efficiency of cross-domain scientific research collaboration.

[0076] Therefore, the above-mentioned cross-domain shared bioinformatics metadata integration and management platform is adopted to achieve standardized integration, secure sharing and semantic retrieval of multi-source heterogeneous metadata, improve cross-laboratory collaboration efficiency and reduce scientific research costs.

[0077] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A cross-domain shared biological information metadata integration and management platform, characterized by: include: Metadata collection module, which supports uploading metadata files in various formats, records data change history, and provides field previews and condition prompts; Metadata standardization and semantic mapping module, which is used to implement field semantic parsing, standardized mapping, and custom mapping management through the built-in biological ontology library; A general metadata model building module that provides graphical and tabular interfaces for defining metadata structure templates, implementing model version control, and field validation. The permission control and sharing mechanism module is used to set data visibility permissions and assign unique metadata reference identifiers; Advanced search and semantic search module, used to support condition combination search and natural language semantic association search; A database that uses MongoDB or Elasticsearch to store semi-structured metadata.

2. A cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The metadata collection module includes: File format support unit, used to support uploading of multiple common cloud data file formats such as CSV, Excel, and JSON; Data version management unit, used to assign a unique version number to each uploaded file to record the change history; The field preview and condition prompt unit is used to provide a field preview function before uploading the metadata file and inform the user of the filling conditions and restriction rules for each field.

3. A cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The metadata standardization and semantic mapping module includes: Built-in biological ontology library unit, used to map non-standard fields uploaded by users to standard biological concepts and identify aliases, providing semantic support for field standardization and semantic matching; Field recommendation mapping unit, which is used to automatically identify structured fields from user-uploaded files and recommend standard terms by performing semantic comparison with built-in biological ontology library units; User-defined mapping unit, used to support customization of mapping templates from private fields to public standards.

4. A cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The general metadata model building module includes: The metadata model definition unit is used to provide two intuitive interfaces, graphical and tabular, to facilitate users to define metadata structure templates; Model version control unit, which is used to generate a new version number each time the field structure is modified and associate it with the metadata record; The model validation unit is used to check field missing, data type matching, and ontology association accuracy when uploading data.

5. A cross-domain shared biological information metadata integration and management platform according to claim 4, characterized in that: The metadata model definition unit supports a tree-structured setting of field levels, allowing the definition of inclusion relationships from samples to tissues to cell types.

6. A cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The authority control and sharing mechanism module includes: A data permission granularity control unit is used to set data visibility to be visible only to oneself, shared within the organization, or open to the entire platform; The reference identifier allocation unit is used to generate a unique reference identifier in the form of DOI or UUID for each metadata record.

7. A cross-domain shared biological information metadata integration and management platform according to claim 6, characterized in that: The authority control and sharing mechanism module also includes a tag management unit for adding classification tags to metadata to achieve efficient retrieval and collaborative research of data.

8. The cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The advanced search and semantic search modules include: Conditional search unit, used to support combined search based on multiple dimensions of conditions such as species, tissue, disease, platform, and experimental type; The natural language retrieval unit is used to introduce natural language processing technology to realize natural language query and semantic association functions.

9. The cross-domain shared biological information metadata integration and management platform according to claim 1, characterized in that: The database utilizes the scalability and natural language search support capabilities of MongoDB or Elasticsearch to achieve efficient storage and indexing of semi-structured data.

Images