UnitedNet
An interpretable multi-task deep neural network for single-cell multi-omics integration and cross-modal analysis.
Overview
UnitedNet is an interpretable multi-task deep neural network designed to analyze single-cell multi-modality data, such as transcriptomics, chromatin accessibility, and proteomics. It provides a comprehensive end-to-end framework for multi-modal integration and cross-modal prediction, enabling researchers to uncover cell-type-specific regulatory relationships across different biological layers.
Key Features
- Multi-modal Integration — Combines data from different omics layers (e.g., RNA, ATAC-seq, proteins) into a unified representation.
- Cross-modal Prediction — Predicts missing modalities from available data, facilitating analysis of incomplete datasets.
- Explainable AI Integration — Incorporates SHAP (SHapley Additive exPlanations) to interpret model decisions and quantify feature relevance.
- Multi-task Learning — Simultaneously handles tasks like clustering, classification, and data fusion within a single model.
- Cell-type-specific Insights — Identifies modality relationships specific to cell types, aiding biological discovery.
Philosophy
UnitedNet is built on the principle that a unified, interpretable model can provide a more complete and actionable understanding of complex multi-modal biological data than single-task methods.
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
