An expectation–maximization algorithm for estimating proportions of deletions among bacterial populations with application to study antibiotic resistance gene transfer in Enterococcus faecalis  

作　　者：Yu Zhang Cong Zhang Wenwen Huo Xinlei Wang Michael Zhang Kelli Palmer Min Chen 

机构地区：[1]School of Mathematical Sciences,Ocean University of China,Qingdao,266000,China [2]Department of Mathematical Sciences,University of Texas at Dallas,Richardson,TX,75080,USA [3]Department of Biological Sciences,University of Texas at Dallas,Richardson,TX,75080,USA [4]Department of Statistical Science,Southern Methodist University,Dallas,TX,75205,USA [5]MOE Key Laboratory of Bioinformatics,Tsinghua University,Beijing,100084,China [6]Department of Population and Data Sciences,UT Southwestern Medical Center,Dallas,TX,75390,USA

出　　处：《Marine Life Science & Technology》2023年第1期28-43,共16页海洋生命科学与技术（英文）

基　　金：This work was supported by the National Institutes of Health[Grant R15GM131390 to X.W.,Grant R01CA245294 to M.Z.,and Grant R01AI116610 to K.P.]。

摘　　要：The emergence of antibiotic resistance in bacteria limits the availability of antibiotic choices for treatment and infection control,thereby representing a major threat to human health.The de novo mutation of bacterial genomes is an essential mechanism by which bacteria acquire antibiotic resistance.Previously,deletion mutations within bacterial immune systems,ranging from dozens to thousands of base pairs(bps)in length,have been associated with the spread of antibiotic resistance.Most current methods for evaluating genomic structural variations(SVs)have concentrated on detecting them,rather than estimating the proportions of populations that carry distinct SVs.A better understanding of the distribution of mutations and subpopulations dynamics in bacterial populations is needed to appreciate antibiotic resistance evolution and movement of resistance genes through populations.Here,we propose a statistical model to estimate the proportions of genomic deletions in a mixed population based on Expectation–Maximization(EM)algorithms and next-generation sequencing(NGS)data.The method integrates both insert size and split-read mapping information to iteratively update estimated distributions.The proposed method was evaluated with three simulations that demonstrated the production of accurate estimations.The proposed method was then applied to investigate the horizontal transfers of antibiotic resistance genes in concert with changes in the CRISPR-Cas system of E.faecalis.

关 键 词：Bacterial genomes CRISPR-Cas system Antibiotic resistance EM algorithm Proportion estimation 

分 类 号：Q178[生物学—水生生物学]