Pathformer: Multi-Omics Data for Disease Diagnosis and Prognosis

Pathformer is a multi-omics integration method developed by Xiao Fan Liu at Tsinghua University for disease diagnosis and prognosis. The associated GitHub repository contains preprocessed data from the TCGA dataset and a liquid biopsy dataset, along with results from the Pathformer model and comparison methods. The model's network architecture was implemented using the PyTorch package in Python v3.6.9.

Use Cases

• Training diagnostic models based on integrated multi-omics data.
• Benchmarking new multi-omics integration methods against Pathformer results.
• Developing prognostic tools for diseases like cancer using tissue and liquid biopsy data.
• Studying gene regulation patterns for disease understanding using preprocessed TCGA data.

Strengths

• Includes preprocessed data from the authoritative TCGA dataset.
• Provides results from the Pathformer model and comparison methods for benchmarking.
• Covers both tissue and liquid biopsy data types for multi-omics integration.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

GitHub repository (https://github.com/lulab/Pathformer) associated with Tsinghua University.

Collection Method

Preprocessed from TCGA and a liquid biopsy dataset.