Deciphering complex antibiotic resistance patterns in Helicobacter pylori through whole genome sequencing and machine learning

• DOI: 10.3389/fcimb.2023.1306368
• Publication Date: 2024-01-04T04:19:06Z
• Authors: Jianwei Yu; Yan Jia; Qichao Yu; Lan Lin; Chao Li; Bowang Chen; Pingyu Zhong; Xueqing Lin; Huilan Li; Yinping Sun; Xuejing Zhong; Yuqi He; Xiaoyun Huang; Shuangming Lin; Yuanming Pan
• Keywords: Helicobactor pylori; 0303 health sciences; Helicobacter pylori; Whole Genome Sequencing; Amoxicillin; Drug Resistance, Microbial; Microbial Sensitivity Tests; Microbiology; genomic sequencing data; QR1-502; Helicobacter Infections; Anti-Bacterial Agents; Machine Learning; 03 medical and health sciences; Cellular and Infection Microbiology; antimicrobial resistance (AMR); Clarithromycin; Drug Resistance, Bacterial; machine learning methods; Humans; molecular mechanism

IntroductionHelicobacter pylori (H.pylori, Hp) affects billions of people worldwide. However, the emerging resistance of Hp to antibiotics challenges the effectiveness of current treatments. Investigating the genotype-phenotype connection for Hp using next-generation sequencing could enhance our understanding of this resistance.MethodsIn this study, we analyzed 52 Hp strains collected from various hospitals. The susceptibility of these strains to five antibiotics was assessed using the agar dilution assay. Whole-genome sequencing was then performed to screen the antimicrobial resistance (AMR) genotypes of these Hp strains. To model the relationship between drug resistance and genotype, we employed univariate statistical tests, unsupervised machine learning, and supervised machine learning techniques, including the development of support vector machine models.ResultsOur models for predicting Amoxicillin resistance demonstrated 66% sensitivity and 100% specificity, while those for Clarithromycin resistance showed 100% sensitivity and 100% specificity. These results outperformed the known resistance sites for Amoxicillin (A1834G) and Clarithromycin (A2147), which had sensitivities of 22.2% and 87%, and specificities of 100% and 96%, respectively.DiscussionOur study demonstrates that predictive modeling using supervised learning algorithms with feature selection can yield diagnostic models with higher predictive power compared to models relying on single single-nucleotide polymorphism (SNP) sites. This approach significantly contributes to enhancing the precision and effectiveness of antibiotic treatment strategies for Hp infections. The application of whole-genome sequencing for Hp presents a promising pathway for advancing personalized medicine in this context.