Integrative COVID-19 biological network inference with probabilistic core decomposition.

Area of Science:

• Computational Biology
• Bioinformatics
• Virology

Background:

• Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes significant global mortality.
• Understanding SARS-CoV-2 and human protein-protein interactions (PPIs) is crucial for developing effective treatments.
• Existing knowledge on SARS-CoV-2 PPIs requires expansion to generate new research hypotheses.

Purpose of the Study:

• To extend the known SARS-CoV-2-human PPI network using the Biomine database.
• To generate novel hypotheses regarding virus-host interactions and potential therapeutic targets.
• To identify potential COVID-19 comorbidities and drug repurposing opportunities.

Main Methods:

• In silico extension of the SARS-CoV-2-human PPI network by integrating Biomine data with experimental results.
• Application of a data analysis pipeline for core decomposition to identify dense subgraphs.
• Evaluation of identified subgraphs and hypotheses through literature validation, gene function enrichment, and drug repurposing analysis.

Main Results:

• An extended SARS-CoV-2-human PPI network was generated, revealing high-connectivity sub-communities.
• Novel hypotheses were proposed, including potential virus-targeting genes and proteins, ranked by their connection to validated nodes.
• A comprehensive list of novel genes, proteins, and potential COVID-19 comorbidities was compiled.

Conclusions:

• The study provides a novel computational framework for generating hypotheses in virus-host interactions.
• The generated hypotheses offer new insights into SARS-CoV-2 pathogenesis and potential therapeutic strategies.
• The findings contribute valuable knowledge for further experimental validation and drug development for COVID-19.