In silico experiment system for testing hypothesis on gene functions using three condition specific biological...

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Determining gene functions traditionally involves time-consuming and expensive experimental methods.
• Access to literature information and biological networks can significantly expedite the process of gene function discovery.

Purpose of the Study:

• To present a web-based information system for performing in silico experiments to computationally test hypotheses on gene function.
• To accelerate the process of gene function determination for researchers.

Main Methods:

• A literature and knowledge mining system (BEST) converts user-specified hypotheses into gene sets.
• Condition-specific transcription factor (TF), microRNA (miRNA), and protein-protein interaction (PPI) networks are generated using gene and miRNA expression data.
• In silico experiments test the connectivity between knockout genes and target genes within these condition-specific networks.

Main Results:

• The system visualizes paths from knockout genes to target genes across TF, miRNA, and PPI networks.
• Statistical and information-theoretic scores are provided to aid in hypothesis validation.
• The system successfully reproduced known gene functions for E2f1, Lrrk2, and Dicer1 knockout datasets.
• Comprehensive testing using MalaCards disease names demonstrated the system's effectiveness in identifying potential biological mechanisms.

Conclusions:

• The developed in silico experiment system effectively aids in computationally testing gene function hypotheses.
• The system provides valuable insights into biological mechanisms, guiding further in vivo or in vitro experimental validation.
• This approach offers a significant advancement in accelerating biological research and discovery.