High-performance single-cell gene regulatory network inference at scale: the Inferelator 3.0

Area of Science:

• Computational biology
• Genomics
• Systems biology

Background:

• Gene regulatory networks are crucial for understanding cellular processes.
• Reconstructing these networks is essential for biological insights.
• Advances in sequencing and machine learning have driven progress in network inference.

Purpose of the Study:

• To present Inferelator 3.0, an updated tool for reconstructing gene regulatory networks.
• To enable integration of distinct cell types for context-specific network learning.
• To aggregate cell-type-specific networks into a shared regulatory network.

Main Methods:

• Utilized Inferelator 3.0 to integrate large-scale single-cell gene expression and chromatin accessibility data.
• Applied the tool to learn cell-type-specific gene regulatory networks from a 1.3 million single-cell dataset.
• Demonstrated scalability by analyzing neuronal and glial cell types from developing Mus musculus brain.

Main Results:

• Inferelator 3.0 successfully integrated diverse cell types to learn context-specific regulatory networks.
• The tool demonstrated the ability to learn cell-type-specific networks from large single-cell datasets.
• New and informative Saccharomyces cerevisiae networks were inferred, validated against a known gold standard.

Conclusions:

• Inferelator 3.0 is a scalable tool capable of reconstructing context-specific gene regulatory networks from large, multi-modal single-cell data.
• The software facilitates the aggregation of individual cell-type networks into a comprehensive shared regulatory network.
• The updated Inferelator enhances the understanding of cellular function and regulation by providing robust network inference capabilities.