{"title":"Machine-learning models utilizing <i>CYP3A4*1G</i> show improved prediction of hypoglycemic medication in Type 2 diabetes.","authors":"Yi Yang, Xing-Yun Hou, Weiqing Ge, Xinye Wang, Yitian Xu, Wansheng Chen, Yaping Tian, Huafang Gao, Qian Chen","doi":"10.2217/pme-2022-0059","DOIUrl":null,"url":null,"abstract":"<p><p>The effectiveness and side effects of Type 2 diabetes (T2D) medication are related to individual genetic background. SNPs <i>CYP3A4</i> and <i>CYP2C19</i> were introduced to machine-learning models to improve the performance of T2D medication prediction. Two multilabel classification models, ML-KNN and WRank-SVM, trained with clinical data and <i>CYP3A4</i>/<i>CYP2C19</i> SNPs were evaluated. Prediction performance was evaluated with Hamming loss, one-error, coverage, ranking loss and average precision. The average precision of ML-KNN and WRank-SVM using clinical data was 92.74% and 92.9%, respectively. Combined with <i>CYP2C19*2*3</i>, the average precision dropped to 88.84% and 89.93%, respectively. While combined with <i>CYP3A4*1G</i>, the average precision was enhanced to 97.96% and 97.82%, respectively. Results suggest that <i>CYP3A4*1G</i> can improve the performance of ML-KNN and WRank-SVM models in predicting T2D medication performance.</p>","PeriodicalId":19753,"journal":{"name":"Personalized medicine","volume":"20 1","pages":"27-37"},"PeriodicalIF":1.7000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Personalized medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.2217/pme-2022-0059","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"PHARMACOLOGY & PHARMACY","Score":null,"Total":0}
引用次数: 1
Abstract
The effectiveness and side effects of Type 2 diabetes (T2D) medication are related to individual genetic background. SNPs CYP3A4 and CYP2C19 were introduced to machine-learning models to improve the performance of T2D medication prediction. Two multilabel classification models, ML-KNN and WRank-SVM, trained with clinical data and CYP3A4/CYP2C19 SNPs were evaluated. Prediction performance was evaluated with Hamming loss, one-error, coverage, ranking loss and average precision. The average precision of ML-KNN and WRank-SVM using clinical data was 92.74% and 92.9%, respectively. Combined with CYP2C19*2*3, the average precision dropped to 88.84% and 89.93%, respectively. While combined with CYP3A4*1G, the average precision was enhanced to 97.96% and 97.82%, respectively. Results suggest that CYP3A4*1G can improve the performance of ML-KNN and WRank-SVM models in predicting T2D medication performance.
期刊介绍:
Personalized Medicine (ISSN 1741-0541) translates recent genomic, genetic and proteomic advances into the clinical context. The journal provides an integrated forum for all players involved - academic and clinical researchers, pharmaceutical companies, regulatory authorities, healthcare management organizations, patient organizations and others in the healthcare community. Personalized Medicine assists these parties to shape thefuture of medicine by providing a platform for expert commentary and analysis.
The journal addresses scientific, commercial and policy issues in the field of precision medicine and includes news and views, current awareness regarding new biomarkers, concise commentary and analysis, reports from the conference circuit and full review articles.