Comprehensive analysis of pathways in Coronavirus 2019 (COVID-19) using an unsupervised machine learning method

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning

Background:

• The COVID-19 pandemic, caused by SARS-CoV-2, necessitates rapid identification of effective treatments.
• Artificial intelligence and bioinformatics offer powerful tools for analyzing complex biological data related to the virus.
• Pathway enrichment analysis is crucial for discovering host cell targets of SARS-CoV-2.

Purpose of the Study:

• To propose and evaluate a two-stage machine learning approach for pathway analysis in COVID-19 research.
• To identify and rank significant biological pathways and signaling networks affected by SARS-CoV-2.
• To uncover potential therapeutic targets for developing COVID-19 treatments.

Main Methods:

• A two-stage machine learning pipeline was developed for pathway analysis.
• Stage one involved selecting informative gene sets representing COVID-19 pathology and constructing signaling and disease networks.
• Stage two ranked pathways using unsupervised scoring based on defined features.

Main Results:

• Identified and ranked key COVID-19 related signaling and disease pathways.
• Presented a comprehensive analysis of the most important pathways implicated in SARS-CoV-2 infection.
• The approach facilitates the discovery of critical host-pathogen interactions.

Conclusions:

• The proposed machine learning approach effectively identifies significant pathways for COVID-19 research.
• This method aids in discovering potential therapeutic targets and understanding disease mechanisms.
• The findings contribute to the ongoing efforts to combat the COVID-19 pandemic.