Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance

Abstract Molecular assays are critical tools for the diagnosis of infectious diseases. These assays have been extremely valuable during the COVID pandemic, used to guide both patient management and infection control strategies. Sustained transmission and unhindered proliferation of the virus during...

Full description

Saved in:
Bibliographic Details
Main Authors: Brittany Knight, Taylor Otwell, Michael P. Coryell, Jennifer Stone, Phillip Davis, Bryan Necciai, Paul E. Carlson, Shanmuga Sozhamannan, Alyxandria M. Schubert, Yi H. Yan
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-98444-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850190504352808960
author Brittany Knight
Taylor Otwell
Michael P. Coryell
Jennifer Stone
Phillip Davis
Bryan Necciai
Paul E. Carlson
Shanmuga Sozhamannan
Alyxandria M. Schubert
Yi H. Yan
author_facet Brittany Knight
Taylor Otwell
Michael P. Coryell
Jennifer Stone
Phillip Davis
Bryan Necciai
Paul E. Carlson
Shanmuga Sozhamannan
Alyxandria M. Schubert
Yi H. Yan
author_sort Brittany Knight
collection DOAJ
description Abstract Molecular assays are critical tools for the diagnosis of infectious diseases. These assays have been extremely valuable during the COVID pandemic, used to guide both patient management and infection control strategies. Sustained transmission and unhindered proliferation of the virus during the pandemic resulted in many variants with unique mutations. Some of these mutations could lead to signature erosion, where tests developed using the genetic sequence of an earlier version of the pathogen may produce false negative results when used to detect novel variants. In this study, we assessed the performance changes of 15 molecular assay designs when challenged with a variety of mutations that fall within the targeted region. Using data generated from this study, we trained and assessed the performance of seven different machine learning models to predict whether a specific set of mutations will result in significant change in the performance for a specific test design. The best performing model demonstrated acceptable performance with sensitivity of 82% and specificity of 87% when assessed using tenfold cross validation. Our findings highlighted the potential of using machine learning models to predict the impact of emerging mutations on the performance of specific molecular test designs.
format Article
id doaj-art-a4e938c6bb2b4c5ea60d6467e6c05011
institution OA Journals
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-a4e938c6bb2b4c5ea60d6467e6c050112025-08-20T02:15:16ZengNature PortfolioScientific Reports2045-23222025-05-0115111110.1038/s41598-025-98444-8Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performanceBrittany Knight0Taylor Otwell1Michael P. Coryell2Jennifer Stone3Phillip Davis4Bryan Necciai5Paul E. Carlson6Shanmuga Sozhamannan7Alyxandria M. Schubert8Yi H. Yan9MRIGlobalMRIGlobalLaboratory of Mucosal Pathogens and Cellular Immunology, Division of Bacterial, Parasitic and Allergenic Products, Office of Vaccines Research and Review, Biologics Evaluation and Research, U.S. Food and Drug AdministrationMRIGlobalMRIGlobalDefense Biological Product Assurance Office (DBPAO), Joint Program Executive Office for Chemical, Biological, Radiological and Nuclear Defense (JPEO-CBRND), Joint Project Lead, Enabling Biotechnologies JPL-EBLaboratory of Mucosal Pathogens and Cellular Immunology, Division of Bacterial, Parasitic and Allergenic Products, Office of Vaccines Research and Review, Biologics Evaluation and Research, U.S. Food and Drug AdministrationDefense Biological Product Assurance Office (DBPAO), Joint Program Executive Office for Chemical, Biological, Radiological and Nuclear Defense (JPEO-CBRND), Joint Project Lead, Enabling Biotechnologies JPL-EBDivision of Microbiology, Office of In Vitro Diagnostics, Office of Product Evaluation and Quality, Center for Devices and Radiological Health, U.S. Food and Drug AdministrationDivision of Microbiology, Office of In Vitro Diagnostics, Office of Product Evaluation and Quality, Center for Devices and Radiological Health, U.S. Food and Drug AdministrationAbstract Molecular assays are critical tools for the diagnosis of infectious diseases. These assays have been extremely valuable during the COVID pandemic, used to guide both patient management and infection control strategies. Sustained transmission and unhindered proliferation of the virus during the pandemic resulted in many variants with unique mutations. Some of these mutations could lead to signature erosion, where tests developed using the genetic sequence of an earlier version of the pathogen may produce false negative results when used to detect novel variants. In this study, we assessed the performance changes of 15 molecular assay designs when challenged with a variety of mutations that fall within the targeted region. Using data generated from this study, we trained and assessed the performance of seven different machine learning models to predict whether a specific set of mutations will result in significant change in the performance for a specific test design. The best performing model demonstrated acceptable performance with sensitivity of 82% and specificity of 87% when assessed using tenfold cross validation. Our findings highlighted the potential of using machine learning models to predict the impact of emerging mutations on the performance of specific molecular test designs.https://doi.org/10.1038/s41598-025-98444-8Signature erosionqPCR performanceIn silico predictionFalse negative resultSupervised learning
spellingShingle Brittany Knight
Taylor Otwell
Michael P. Coryell
Jennifer Stone
Phillip Davis
Bryan Necciai
Paul E. Carlson
Shanmuga Sozhamannan
Alyxandria M. Schubert
Yi H. Yan
Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
Scientific Reports
Signature erosion
qPCR performance
In silico prediction
False negative result
Supervised learning
title Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
title_full Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
title_fullStr Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
title_full_unstemmed Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
title_short Using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
title_sort using machine learning models to predict the impact of template mismatches on polymerase chain reaction assay performance
topic Signature erosion
qPCR performance
In silico prediction
False negative result
Supervised learning
url https://doi.org/10.1038/s41598-025-98444-8
work_keys_str_mv AT brittanyknight usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT taylorotwell usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT michaelpcoryell usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT jenniferstone usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT phillipdavis usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT bryannecciai usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT paulecarlson usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT shanmugasozhamannan usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT alyxandriamschubert usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance
AT yihyan usingmachinelearningmodelstopredicttheimpactoftemplatemismatchesonpolymerasechainreactionassayperformance