CellFreeGMF - identify cfRNA biomarkers, and analyze the alterations in their originating cells

• Installation for CellFreeGMF package
  • Software dependencies
  • Installation
• Tutorial 1: Disease diagnosis and biomarker identification
  • Preparation
  • Hyperparameter Configuration
  • Load Data
  • Diagnostic classification using machine learning was performed.
  • Performance comparison of diagnostic classification algorithms to screen for the algorithm optimally suited to PDAC.
  • The optimal logistic regression model was selected and subsequently retrained on the full dataset; interpretability analysis was then performed to identify candidate target cfRNAs.
  • SHAP-based interpretability analysis
  • Save the differential expression analysis results and the SHAP results.
• Tutorial 2: inferring a sample–cell matrix and reconstructing sample-specific single-cell transcriptomic profiles
  • Preparation
  • Hyperparameter Configuration
  • Load the cfRNA transcriptome matrix
  • Load the references single-cell transcriptomic data
  • Visualization of scRNA-seq Data
  • Target information was loaded for subsequent analysis.
  • inferring a sample–cell matrix and reconstructing sample-specific single-cell transcriptomic profiles
• Tutorial 3: Compare cell-cell communication difference between the PDAC group and the normal group
  • Preparation
  • Hyperparameter Configuration
  • Load predicted sample-specific single-cell transcriptomic
  • Cell-Cell communications
• Tutorial 4: Compare the predicted sample-cfRNA between the disease group and the normal group
  • Preparation
  • Hyperparameter Configuration
  • Load predicted data

Abstract

Understanding the cellular origins of cell-free RNA (cfRNA) and their alterations is essential for elucidating disease-related molecular processes. Here, we propose CellFreeGMF, a tool designed to enable diagnosis classification of samples, identify cfRNA biomarkers, and analyze the alterations in their originating cells based on graph matrix factorization. Furthermore, by utilizing cell–cell communication analysis, CellFreeGMF investigates the functional alterations occurring in the cfRNA originating cells under disease conditions. We validate CellFreeGMF on diverse cell-free RNA transcriptome datasets. In the case of pancreatic ductal adenocarcinoma (PDAC), CellFreeGMF not only identified cfRNA biomarkers but also traced their cellular origins to myeloid and T-cell populations. Further analysis revealed significant transcriptomic differences in these cell populations between pathological and normal control groups. Our user-friendly toolkit, CellFreeGMF, enables identifying cfRNA biomarkers and elucidating pathophysiological changes in their source cells.

Citation

Wenxiang Zhang, et al. “From Cell-Free Transcriptomes to Single-Cell Landscapes: Biomarker Discovery and Originating Cell Alteration Analysis via Graph Matrix Factorization”, Submitted.