Protocol to identify biomarkers in patients with post-COVID condition using multi-omics and machine learning analysis of human plasma

Mobin Khoramjoo; Karthik Srinivasan; Kaiming Wang; David Wishart; Vinay Prasad; Gavin Y Oudit

doi:10.1016/j.xpro.2024.103041

Protocol to identify biomarkers in patients with post-COVID condition using multi-omics and machine learning analysis of human plasma

STAR Protoc. 2024 Apr 27;5(2):103041. doi: 10.1016/j.xpro.2024.103041. Online ahead of print.

Authors

Mobin Khoramjoo¹, Karthik Srinivasan², Kaiming Wang³, David Wishart⁴, Vinay Prasad², Gavin Y Oudit⁵

Affiliations

¹ Department of Physiology, University of Alberta, Edmonton, AB T6G 2H7, Canada; Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB T6G 2S2, Canada.
² Department of Chemical and Materials Engineering, University of Alberta, Edmonton, AB T6G 1H9, Canada.
³ Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB T6G 2S2, Canada; Division of Cardiology, Department of Medicine, University of Alberta, Edmonton, AB T6G 2G3, Canada.
⁴ The Metabolomics Innovation Center, University of Alberta, Edmonton, AB T6G 2E9, Canada.
⁵ Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB T6G 2S2, Canada; Division of Cardiology, Department of Medicine, University of Alberta, Edmonton, AB T6G 2G3, Canada. Electronic address: gavin.oudit@ualberta.ca.

Abstract

Here, we present a workflow for analyzing multi-omics data of plasma samples in patients with post-COVID condition (PCC). Applicable to various diseases, we outline steps for data preprocessing and integrating diverse assay datasets. Then, we detail statistical analysis to unveil plasma profile changes and identify biomarker-clinical variable associations. The last two steps discuss machine learning techniques for unsupervised clustering of patients based on their inherent molecular similarities and feature selection to identify predictive biomarkers. For complete details on the use and execution of this protocol, please refer to Wang et al.¹.

Keywords: Bioinformatics; Clinical Protocol; Metabolomics; Proteomics.