xcore: an R package for inference of gene expression regulators

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Identifying transcription factors (TFs) driving gene expression changes is crucial in research.
• Current methods often rely on predicted Transcription Factor Binding Sites (TFBS), limiting analysis to TFs with known motifs and assuming universal TF binding profiles.
• These limitations hinder comprehensive understanding of gene regulation.

Purpose of the Study:

• To develop a novel approach for modeling TF activity directly from ChIP-seq data, bypassing traditional motif prediction.
• To introduce xcore, an R package designed for TF activity modeling using ChIP-seq signatures and gene expression data.
• To provide a user-friendly tool that leverages existing ChIP-seq resources for gene expression analysis.

Main Methods:

• Utilizing ChIP-seq "signatures" directly to model gene expression, circumventing the need for motif finding and TFBS prediction.
• Developing the xcore R package for TF activity modeling.
• Creating the companion xcoredata package with preprocessed ChIP-seq signatures.
• Validating the approach using diverse biological datasets, including TGF-beta induced EMT, rinderpest infection, and ESC differentiation.

Main Results:

• Demonstrated that xcore can effectively model TF activity using ChIP-seq signatures.
• Showcased biologically relevant predictions across multiple experimental models and cell types.
• Successfully integrated ChIP-seq data with gene expression profiles for enhanced analysis.

Conclusions:

• xcore offers a streamlined analytical framework for gene expression modeling using linear models.
• The package can be readily integrated into existing differential expression analysis pipelines.
• xcore effectively identifies significant molecular signatures and relevant ChIP-seq experiments by utilizing public ChIP-seq databases.