Data integration and quality control system
Inventors
Guo, Jun • Chalkia, Dimitra • Robertson, Colin John
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12387013-B1
Publication Date
2025-08-12
Expiration Date
Abstract
A data management system, method and computer-readable medium for data integration and quality control are described. In some implementations, a computer-implemented method comprises steps of receiving datasets from one or more data sources, storing the datasets belonging to a domain, checking data integrity of the datasets, performing a quality control check on the datasets, receiving selections from the domain on the datasets that are selected to be processed based on one or more reference libraries, processing one or more selected datasets based on the one or more reference libraries to create mapped data, integrating the mapped data with metadata to provide an integrated analysis, and causing to display, at a graphical user interface (GUI), real-time processing status for each of the one or more selected datasets.
Core Innovation
A data management system and a data management platform centrally manage datasets received from one or more data sources for a domain. The platform includes a data store configured to store datasets belonging to the domain and a data import platform that receives the datasets, followed by a centralized data management process. The process includes checking data integrity of the datasets and performing a quality control check on the datasets.
The quality control check provides analysis on whether a region of a dataset passes one or more quality control thresholds. The one or more quality control thresholds are determined using a machine learning model based on historical data quality metrics and user feedback. The process also receives selections from the domain on datasets selected to be processed based on one or more reference libraries, and processes the selected datasets by mapping the data to the reference libraries to create mapped data.
Mapped data is integrated with metadata to provide an integrated analysis. The integrated analysis includes performing pathway enrichment analysis on the mapped data to identify significant pathways and visualizing interaction networks to assist in identifying potential markers or targets. A graphical user interface is configured to display real-time processing status for each of the one or more selected datasets, enabling management of the datasets on the domain and the integrated analysis.
Claims Coverage
Independent claims cover a centralized domain-specific omics/workflow that includes data import, data integrity checking, machine-learning-determined region-level quality control thresholds based on historical data quality metrics and user feedback, reference-library-based dataset selection and mapping, metadata-integrated analysis with pathway enrichment and interaction-network visualization, and a GUI that displays real-time processing status. The same core inventive workflow is implemented as a system, a computer-implemented method, and a non-transitory computer-readable medium.
Centralized domain dataset management with data import platform
A data management system that centrally manages datasets received from one or more data sources for a domain by receiving datasets via a data import platform, checking data integrity, and performing a data management process on the received datasets.
Machine-learning-determined region quality control thresholds with user feedback
A quality control check that analyzes whether a region of a dataset passes one or more quality control thresholds, where the one or more quality control thresholds are determined using a machine learning model based on historical data quality metrics and user feedback.
Reference-library-based dataset selection and mapping to create mapped data
Receiving selections from the domain on datasets selected to be processed based on one or more reference libraries, and processing the selected datasets by mapping the data to the one or more reference libraries to create mapped data.
Metadata-integrated pathway enrichment and interaction-network visualization for markers or targets
Integrating the mapped data with metadata to provide an integrated analysis that includes performing pathway enrichment analysis on the mapped data to identify significant pathways and visualizing interaction networks to assist in identifying potential markers or targets.
GUI for real-time processing status display for selected datasets
A graphical user interface configured to display real-time processing status for each of the one or more selected datasets, allowing management of the datasets on the domain and the integrated analysis.
Computer-implemented method executing the integrated workflow on processors
A computer-implemented method that receives via a network datasets from one or more data sources, stores datasets belonging to a domain, checks data integrity, performs a machine-learning-driven quality control check with thresholds based on historical data quality metrics and user feedback, receives domain selections based on reference libraries, maps selected datasets to create mapped data, integrates mapped data with metadata for integrated analysis including pathway enrichment and interaction-network visualization, and causes display at a GUI of real-time processing status.
Non-transitory computer-readable medium storing executable instructions for the workflow
A non-transitory computer-readable medium configured to store code comprising instructions which, when executed, cause processors to receive datasets via network, store datasets belonging to a domain, check data integrity, perform the quality control check with machine learning-determined thresholds based on historical data quality metrics and user feedback, receive domain selections based on reference libraries, map selected datasets to create mapped data, integrate mapped data with metadata for integrated analysis including pathway enrichment and interaction-network visualization, and cause display at a GUI of real-time processing status.
Across the independent claims, the core claim coverage is directed to a centralized domain dataset workflow that combines integrity checking, region-level quality control thresholds determined by a machine learning model using historical data quality metrics and user feedback, domain-driven selection of datasets to be processed against reference libraries and mapping to create mapped data, metadata-integrated pathway enrichment and interaction-network visualization to assist in identifying potential markers or targets, and a GUI that displays real-time processing status for each selected dataset.
Stated Advantages
Provides pathway enrichment analysis on the mapped data to identify significant pathways.
Visualizes interaction networks to assist in identifying potential markers or targets.
Displays real-time processing status for each of the one or more selected datasets on the GUI.
Documented Applications
No documented applications found
Interested in licensing this patent?