Multi-Omics Joint Analysis: Proteomics and Metabolomics Help Classify Severe COVID-19 Patients

Coronavirus disease COVID-19 is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and continues to spread worldwide. Approximately 80% of patients infected with SARS-CoV-2 exhibit mild symptoms and have a good prognosis. However, about 20% of patients develop respiratory distress and require immediate oxygen therapy or other hospital interventions. These patients are classified and diagnosed primarily based on a set of clinical features, such as respiratory rate (≥30 breaths/min), average oxygen saturation (at rest ≤93%), or arterial oxygen partial pressure/fraction of inspired oxygen (≤300 mmHg). However, patients exhibiting these clinical features have already progressed to a clinically severe stage, necessitating specialized intensive care immediately; otherwise, they may rapidly die. Therefore, developing new methods to assess which cases may become clinically severe at an early stage is crucial.

In this study, the researchers hypothesized that characteristic molecular changes induced by SARS-CoV-2 could be detected in the serum of severe patients. They employed proteomics and metabolomics techniques to analyze the serum of COVID-19 patients and several control groups to validate this hypothesis.

s7

Journal: Cell

Impact Factor: 38.637

Publication Date: May 2020

Link: https://www.cell.com/cell/fulltext/S0092-8674(20)30627-9

Abstract

Early detection and effective treatment of severe COVID-19 patients remain major challenges. The researchers conducted proteomics and metabolomics analyses on the serum of 46 COVID-19 and 53 control individuals. They then trained a machine learning model using proteomics and metabolomics data from 18 non-severe and 13 severe patients. The model was validated with 10 independent patients, where 7 were correctly classified. In a second test group with 19 COVID-19 patients, targeted proteomics and metabolomics approaches were used to further validate the molecular classifier, resulting in 16 correct classifications. The study identified molecular changes in the serum of COVID-19 patients involving macrophage dysregulation, platelet degranulation, complement system pathways, and significant metabolic inhibition compared to other groups.

Main Results

The researchers first used TMT-based proteomics and untargeted metabolomics to analyze the serum of patients and control individuals, identifying and quantifying a total of 894 proteins and 941 metabolites (including 36 drugs and their metabolites). Without molecular selection, the omics data from the serum of SARS-CoV-2 infected patients could be well distinguished from healthy individuals, while other groups showed some degree of separation. Based on the proteomics and metabolomics data from 18 non-severe and 13 severe patients, the researchers established a random machine learning model, prioritizing 29 important variables, including 22 proteins and 7 metabolites. The model was tested on an independent test group of 10 patients, correctly identifying all severe patients except one. To further validate the classifier, the researchers developed a targeted mass spectrometry analysis for the 22 proteins and 7 metabolites, testing it on 19 patients with 16 correct classifications. The study found 105 differentially expressed proteins in the serum of COVID-19 patients, with 93 proteins specifically regulated in severe patients. Pathway and network enrichment analyses indicated that 50 of the 93 proteins belonged to three major pathways: complement system activation, macrophage function, and platelet degranulation. The study found significant changes in 373 metabolites in COVID-19 patients, with changes in 204 metabolites associated with disease severity assessed by mFuzz. Eighty significantly altered metabolites were linked to the three biological processes revealed by proteomics analysis.

Distinguishing Severe and Non-Severe COVID-19 Patients Using Proteomic and Metabolomic Features with Machine Learning

Dysregulated Proteins in COVID-19 Serum

Dysregulated Metabolites in COVID-19 Serum

Biotech company Biotyper offers multi-omics integration analysis services, including combined proteomics and metabolomics analysis.

We welcome interested researchers to contact us!

Related Services

Integrated Analysis of Proteomics and Metabolomics
Proteomics
Metabolomics
Multi-Omics Integration Analysis