Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review]
Background The TaqMan Array Card (TAC) is an arrayed, high-throughput qPCR platform that can simultaneously detect multiple targets in a single reaction. However, the manual post-run analysis of TAC data is time consuming and subject to interpretation. We sought to automate the post-run analysis of...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
F1000 Research Ltd
2025-01-01
|
| Series: | Gates Open Research |
| Subjects: | |
| Online Access: | https://gatesopenresearch.org/articles/9-1/v1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592859845361664 |
|---|---|
| author | David Garrett Brown Darwin J. Operario Lan Wang Shanrui Wu Daniel T. Leung Eric R. Houpt James A. Platts-Mills Jie Liu Ben J. Brintz |
| author_facet | David Garrett Brown Darwin J. Operario Lan Wang Shanrui Wu Daniel T. Leung Eric R. Houpt James A. Platts-Mills Jie Liu Ben J. Brintz |
| author_sort | David Garrett Brown |
| collection | DOAJ |
| description | Background The TaqMan Array Card (TAC) is an arrayed, high-throughput qPCR platform that can simultaneously detect multiple targets in a single reaction. However, the manual post-run analysis of TAC data is time consuming and subject to interpretation. We sought to automate the post-run analysis of TAC data using machine learning models. Methods We used 165,214 qPCR amplification curves from two studies to train and test two eXtreme Gradient Boosting (XGBoost) models. Previous manual analyses of the amplification curves by experts in qPCR analysis were used as the gold standard. First, a classification model predicted whether amplification occurred or not, and if so, a second model predicted the cycle threshold (Ct) value. We used 5-fold cross-validation to tune the models and assessed performance using accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and mean absolute error (MAE). For external validation, we used 1,472 reactions previously analyzed by 17 laboratory scientists as part of an external quality assessment for a multisite study. Results In internal validation, the classification model achieved an accuracy of 0.996, sensitivity of 0.997, specificity of 0.993, PPV of 0.998, and NPV of 0.991. The Ct prediction model achieved a MAE of 0.590. In external validation, the automated analysis achieved an accuracy of 0.997 and a MAE of 0.611, and the automated analysis was more accurate than manual analyses by 14 of the 17 laboratory scientists. Conclusions We automated the post-run analysis of highly-arrayed qPCR data using machine learning models with high accuracy in comparison to a manual gold standard. This approach has the potential to save time and improve reproducibility in laboratories using the TAC platform and other high-throughput qPCR approaches. |
| format | Article |
| id | doaj-art-5a64703afb7148bd9c758e13c19db059 |
| institution | Kabale University |
| issn | 2572-4754 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | F1000 Research Ltd |
| record_format | Article |
| series | Gates Open Research |
| spelling | doaj-art-5a64703afb7148bd9c758e13c19db0592025-01-21T01:00:00ZengF1000 Research LtdGates Open Research2572-47542025-01-01917704Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review]David Garrett Brown0Darwin J. Operario1Lan Wang2Shanrui Wu3Daniel T. Leung4Eric R. Houpt5James A. Platts-Mills6https://orcid.org/0000-0002-4956-0418Jie Liu7Ben J. Brintz8https://orcid.org/0000-0003-4695-0290University of Utah Department of Internal Medicine, Salt Lake City, Utah, USAUniversity of Virginia, Charlottesville, Virginia, USAQingdao University School of Public Healh, Qingdao, Shandong, ChinaQingdao University School of Public Healh, Qingdao, Shandong, ChinaUniversity of Utah Department of Internal Medicine, Salt Lake City, Utah, USAUniversity of Virginia, Charlottesville, Virginia, USAUniversity of Virginia, Charlottesville, Virginia, USAUniversity of Virginia, Charlottesville, Virginia, USAUniversity of Utah Department of Internal Medicine, Salt Lake City, Utah, USABackground The TaqMan Array Card (TAC) is an arrayed, high-throughput qPCR platform that can simultaneously detect multiple targets in a single reaction. However, the manual post-run analysis of TAC data is time consuming and subject to interpretation. We sought to automate the post-run analysis of TAC data using machine learning models. Methods We used 165,214 qPCR amplification curves from two studies to train and test two eXtreme Gradient Boosting (XGBoost) models. Previous manual analyses of the amplification curves by experts in qPCR analysis were used as the gold standard. First, a classification model predicted whether amplification occurred or not, and if so, a second model predicted the cycle threshold (Ct) value. We used 5-fold cross-validation to tune the models and assessed performance using accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and mean absolute error (MAE). For external validation, we used 1,472 reactions previously analyzed by 17 laboratory scientists as part of an external quality assessment for a multisite study. Results In internal validation, the classification model achieved an accuracy of 0.996, sensitivity of 0.997, specificity of 0.993, PPV of 0.998, and NPV of 0.991. The Ct prediction model achieved a MAE of 0.590. In external validation, the automated analysis achieved an accuracy of 0.997 and a MAE of 0.611, and the automated analysis was more accurate than manual analyses by 14 of the 17 laboratory scientists. Conclusions We automated the post-run analysis of highly-arrayed qPCR data using machine learning models with high accuracy in comparison to a manual gold standard. This approach has the potential to save time and improve reproducibility in laboratories using the TAC platform and other high-throughput qPCR approaches.https://gatesopenresearch.org/articles/9-1/v1qPCR PCR amplification cycle threshold machine learningeng |
| spellingShingle | David Garrett Brown Darwin J. Operario Lan Wang Shanrui Wu Daniel T. Leung Eric R. Houpt James A. Platts-Mills Jie Liu Ben J. Brintz Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] Gates Open Research qPCR PCR amplification cycle threshold machine learning eng |
| title | Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] |
| title_full | Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] |
| title_fullStr | Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] |
| title_full_unstemmed | Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] |
| title_short | Automated post-run analysis of arrayed quantitative PCR amplification curves using machine learning [version 1; peer review: awaiting peer review] |
| title_sort | automated post run analysis of arrayed quantitative pcr amplification curves using machine learning version 1 peer review awaiting peer review |
| topic | qPCR PCR amplification cycle threshold machine learning eng |
| url | https://gatesopenresearch.org/articles/9-1/v1 |
| work_keys_str_mv | AT davidgarrettbrown automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT darwinjoperario automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT lanwang automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT shanruiwu automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT danieltleung automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT ericrhoupt automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT jamesaplattsmills automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT jieliu automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview AT benjbrintz automatedpostrunanalysisofarrayedquantitativepcramplificationcurvesusingmachinelearningversion1peerreviewawaitingpeerreview |