Dataset Information

Comparative performance of two automated machine learning platforms for COVID-19 detection by MALDI-TOF-MS.

ABSTRACT: The 2019 novel coronavirus infectious disease (COVID-19) pandemic has resulted in an unsustainable need for diagnostic tests. Currently, molecular tests are the accepted standard for the detection of SARS-CoV-2. Mass spectrometry (MS) enhanced by machine learning (ML) has recently been postulated to serve as a rapid, high-throughput, and low-cost alternative to molecular methods. Automated ML is a novel approach that could move mass spectrometry techniques beyond the confines of traditional laboratory settings. However, it remains unknown how different automated ML platforms perform for COVID-19 MS analysis. To this end, the goal of our study is to compare algorithms produced by two commercial automated ML platforms (Platforms A and B). Our study consisted of MS data derived from 361 subjects with molecular confirmation of COVID-19 status including SARS-CoV-2 variants. The top optimized ML model with respect to positive percent agreement (PPA) within Platforms A and B exhibited an accuracy of 94.9%, PPA of 100%, negative percent agreement (NPA) of 93%, and an accuracy of 91.8%, PPA of 100%, and NPA of 89%, respectively. These results illustrate the MS method's robustness against SARS-CoV-2 variants and highlight similarities and differences in automated ML platforms in producing optimal predictive algorithms for a given dataset.

SUBMITTER: Rashidi HH

PROVIDER: S-EPMC9337631 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comparative performance of two automated machine learning platforms for COVID-19 detection by MALDI-TOF-MS.

Rashidi Hooman H HH Pepper John J Howard Taylor T Klein Karina K May Larissa L Albahra Samer S Phinney Brett B Salemi Michelle R MR Tran Nam K NK

PloS one 20220729 7

The 2019 novel coronavirus infectious disease (COVID-19) pandemic has resulted in an unsustainable need for diagnostic tests. Currently, molecular tests are the accepted standard for the detection of SARS-CoV-2. Mass spectrometry (MS) enhanced by machine learning (ML) has recently been postulated to serve as a rapid, high-throughput, and low-cost alternative to molecular methods. Automated ML is a novel approach that could move mass spectrometry techniques beyond the confines of traditional labo ...[more]

PMID: 35905092

Similar Datasets

Project description:Burkholderia pseudomallei is the causative agent of melioidosis, a disease highly endemic to Southeast Asia and northern Australia, though the area of endemicity is expanding. Cases may occur in returning travelers or, rarely, from imported contaminated products. Identification of B. pseudomallei is challenging for laboratories that do not see this organism frequently, and misidentifications by matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) and automated biochemical testing have been reported. The in vitro diagnostic database for use with the Vitek MS has recently been updated to include B. pseudomallei and we aimed to validate the performance for identification in comparison to automated biochemical testing with the Vitek 2 GN card, quantitative real-time polymerase chain reaction (qPCR) targeting the type III secretion system, and capsular polysaccharide antigen detection using a lateral flow immunoassay (LFA). We tested a "derivation" cohort including geographically diverse B. pseudomallei and a range of closely related Burkholderia species, and a prospective "validation" cohort of B. pseudomallei and B. cepacia complex clinical isolates. MALDI-TOF MS had a sensitivity of 1.0 and specificity of 1.0 for the identification and differentiation of B. pseudomallei from related Burkholderia species when a certainty cutoff of 99.9% was used. In contrast, automated biochemical testing for B. pseudomallei identification had a sensitivity of 0.83 and specificity of 0.88. Both qPCR and LFA correctly identified all B. pseudomallei isolates with no false positives. Due to the high level of accuracy, we have now incorporated MALDI-TOF MS into our laboratory's B. pseudomallei identification workflow.IMPORTANCEBurkholderia pseudomallei causes melioidosis, a disease associated with high morbidity and mortality that disproportionately affects rural areas in Southeast Asia and northern Australia. The known area of endemicity is expanding and now includes the continental United States. Laboratory identification can be challenging which may result in missed or delayed diagnoses and poor patient outcomes. In this study, we compared mass spectrometry using an updated spectral database with multiple other methods for B. pseudomallei identification and found mass spectrometry highly accurate. We have therefore incorporated this fast and cost-effective method into our laboratory's workflow for B. pseudomallei identification.

Dataset Information

Comparative performance of two automated machine learning platforms for COVID-19 detection by MALDI-TOF-MS.

Publications

Comparative performance of two automated machine learning platforms for COVID-19 detection by MALDI-TOF-MS.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets