Inference of Gene Co-expression Networks from Single-Cell RNA-Sequencing Data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) offers higher resolution than bulk RNA sequencing.
• Gene co-expression networks are crucial for inferring gene function via the "guilt-by-association" heuristic.
• Existing methods for bulk RNA data often require adaptation for scRNA-seq.

Purpose of the Study:

• To review and explore established gene co-expression network construction methods for their applicability to scRNA-seq data.
• To discuss novel approaches for incorporating temporal information from scRNA-seq data.
• To enable the identification of directed gene associations within cellular functions.

Main Methods:

• Adaptation of established gene co-expression network algorithms for scRNA-seq.
• Application of "guilt-by-association" principles to single-cell gene expression matrices.
• Development and evaluation of methods for leveraging time-series scRNA-seq data.

Main Results:

• Many standard gene co-expression network methods are transferable to scRNA-seq.
• Novel methods allow for the construction of directed gene co-expression networks using temporal scRNA-seq data.
• These approaches enhance the identification of functionally related genes.

Conclusions:

• Gene co-expression network analysis is a valuable tool for interpreting scRNA-seq data.
• Incorporating temporal information significantly improves the inference of gene regulatory relationships.
• The discussed methods provide a framework for advancing functional genomics studies using scRNA-seq.