TGS-Lite

TGS-Lite reconstructs time-varying gene regulatory networks from time-series gene expression data to enable dynamic analysis of gene regulation.

Key Features:

• Time efficiency: Implements algorithms that reduce computational time for reconstructing GRNs from time-series expression data.
• Memory efficiency: Operates with low main memory requirements to support reconstruction for large gene sets.
• No smoothly time-varying assumption: Reconstructs networks without requiring the 'smoothly time-varying' assumption for temporal changes in interactions.
• Candidate regulator shortlisting: Employs a shortlisting approach for candidate regulators to reduce the search space and improve computational performance.
• Reconstruction correctness: Has been validated on three benchmark datasets, demonstrating improved accuracy of network reconstruction relative to existing methods.

Scientific Applications:

• Dynamic interaction inference: Infers time-resolved gene interactions from time-series expression data.
• Regulatory gene identification: Identifies key regulatory genes involved in disease progression or response to treatment.
• Temporal modeling of biological processes: Supports modeling of complex biological processes that require temporal resolution.

Methodology:

TGS-Lite and TGS-Lite+ implement algorithms that shortlist candidate regulators to reduce the search space, optimizing time- and memory-efficient reconstruction while not assuming smoothly time-varying interactions.