Targeted plasma metabolomics combined with machine learning for the diagnosis of severe acute respiratory syndrome virus type 2

Anthony T Le; Manhong Wu; Afraz Khan; Nicholas Phillips; Pranav Rajpurkar; Megan Garland; Kayla Magid; Mamdouh Sibai; ChunHong Huang; Malaya K Sahoo; Raffick Bowen; Tina M Cowan; Benjamin A Pinsky; Catherine A Hogan

doi:10.3389/fmicb.2022.1059289

Targeted plasma metabolomics combined with machine learning for the diagnosis of severe acute respiratory syndrome virus type 2

Front Microbiol. 2023 Mar 29:13:1059289. doi: 10.3389/fmicb.2022.1059289. eCollection 2022.

Authors

Anthony T Le¹, Manhong Wu², Afraz Khan³, Nicholas Phillips⁴, Pranav Rajpurkar⁴, Megan Garland¹, Kayla Magid¹, Mamdouh Sibai¹, ChunHong Huang¹, Malaya K Sahoo¹, Raffick Bowen^{1

5}, Tina M Cowan^{1

6}, Benjamin A Pinsky^{1

7

8}, Catherine A Hogan^{1

3

7

9}

Affiliations

¹ Department of Pathology, Stanford University School of Medicine, Stanford, CA, United States.
² Department of Anesthesiology, Stanford University School of Medicine, Stanford, CA, United States.
³ British Columbia Center for Disease Control Public Health Laboratory, Vancouver, BC, Canada.
⁴ Stanford Computer Science Department, Stanford University, Stanford, CA, United States.
⁵ Stanford Biochemical Genetics Laboratory, Stanford Health Care, Palo Alto, CA, United States.
⁶ Clinical Chemistry and Immunology Laboratory, Stanford Health Care, Palo Alto, CA, United States.
⁷ Stanford Clinical Virology Laboratory, Stanford Health Care, Palo Alto, CA, United States.
⁸ Division of Infectious Diseases and Geographic Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, CA, United States.
⁹ Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada.

Abstract

Introduction: The routine clinical diagnosis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is largely restricted to real-time reverse transcription quantitative PCR (RT-qPCR), and tests that detect SARS-CoV-2 nucleocapsid antigen. Given the diagnostic delay and suboptimal sensitivity associated with these respective methods, alternative diagnostic strategies are needed for acute infection.

Methods: We studied the use of a clinically validated liquid chromatography triple quadrupole method (LC/MS-MS) for detection of amino acids from plasma specimens. We applied machine learning models to distinguish between SARS-CoV-2-positive and negative samples and analyzed amino acid feature importance.

Results: A total of 200 samples were tested, including 70 from individuals with COVID-19, and 130 from negative controls. The top performing model overall allowed discrimination between SARS-CoV-2-positive and negative control samples with an area under the receiver operating characteristic curve (AUC) of 0.96 (95%CI 0.91, 1.00), overall sensitivity of 0.99 (95%CI 0.92, 1.00), and specificity of 0.92 (95%CI 0.85, 0.95).

Discussion: This approach holds potential as an alternative to existing methods for the rapid and accurate diagnosis of acute SARS-CoV-2 infection.

Keywords: COVID-19; SARS-CoV-2; amino acids; metabolomics; plasma.