Proteomics and Machine Learning Approaches Reveal a Set of Prognostic Markers for COVID-19 Severity With Drug Repurposing Potential
Problem
Predicting and managing COVID-19 disease severity was difficult, and mechanisms underlying severity were unclear.
Why it matters
Severity prognosis could guide treatment decisions and save lives during the pandemic.
What the AI did
Machine learning combined with mass-spectrometry-based deep proteomics and metabolomics of 186 patients' plasma to reveal prognostic markers for COVID-19 severity with drug repurposing potential.
Results
Of 1200 proteins detected, 27 differentially regulated in COVID-positive group and 38 differentially expressed between non-severe and severe groups (186-patient cohort).
Impact
Identified candidate prognostic markers and repurposing targets in one 186-patient cohort; clinical validation not stated.
Record
| Venue | Frontiers in Physiology |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Kruthi Suvarna, Deeptarup Biswas, Medha Gayathri J. Pai, Arup Acharjee, Renuka Bankar, Viswanthram Palanivel, Akanksha Salkar, Ayushi Verma et al. |
| Citations | 76 (OpenAlex, at import time) |
| References | 114 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3389/fphys.2021.652799 |
| Open access | gold |
| Keywords | COVID-19, proteomics, machine learning, prognostic markers, drug repurposing |
Importance rating
existential / global scale
Abstract frames COVID-19 severity prognosis as a challenge during the ongoing pandemic; rubric assigns pandemic-scale problems the top score.