xCell2

xCell2 performs cell type enrichment analysis from bulk gene expression data using cell type signatures, with a training function that enables utilization of any reference dataset and improved methodology for generating robust signatures with automated handling of cell type dependencies.

Key Features:

• Training Function: Permits utilization of any reference dataset for generating custom cell type gene signatures.
• Automated Dependency Handling: Includes automated handling of cell type dependencies in signature generation.
• Robust Signature Generation: Employs an improved methodology for generating more robust cell type gene signatures.
• Spillover Minimization: Demonstrates best performance in minimizing spillover effects between related cell types.
• Pre-trained References: Includes a large collection of pre-trained cell type signatures for human and mouse research.

Scientific Applications:

• Cell Type Proportion Estimation: Estimates cell type proportions from bulk gene expression data for understanding cellular heterogeneity in complex tissues and diseases.
• Immune Checkpoint Blockade Response Prediction: Predicts response to immune checkpoint blockade, with xCell 2.0-derived tumor microenvironment features significantly improving prediction accuracy compared to models using only cancer type and treatment information.
• Pan-cancer Immune Analysis: Applied in pan-cancer immune cell checkpoint blockage response prediction studies.

Methodology:

xCell 2.0 generates cell type gene signatures using an improved methodology with automated handling of cell type dependencies and more robust signature generation. The method was benchmarked against eleven popular deconvolution methods using nine human and mouse reference sets, 26 validation datasets encompassing 1711 samples and 67 cell types, and the independent Deconvolution DREAM Challenge dataset.