| Objective: The aim of this paper was to study the drug resistant characters among thosebacteria. Those bacteria all collected from Shenyang and had some characters of resistanceβ-lactamase antibiotics. It was meaningful to analyze drug resistant characters by using somebioinformatic methods. One of the fashionable methods to do molecular evolutionary wascombining Bayesian inference and Markov Chain Monte Carlo (MCMC) procedure. Thismethod was used for estimating the time to the most recent common ancestor (TMRCA) andconstructing phylogenetic tree. Some bioinformatic methods were used to solve a toughproblem in microbiology. Those methods had several advantages in finding the phylogeneticrelationships among bacteria.Methods:113strains Gram-negative bacteria were collected from Shenyang between2012and2013. Those bacteria were identified and did drug sensitive tests. It was efficient touse biclustering method to analyze the results of drug sensitive tests. After that, we madesome programs by Perl language and BioPerl modules. Downloading nucleotide sequencesefficiently and semiautomatically and integrating download nucleotide sequence (.gbk)documents to new FASTA document could be achieved by BioPerl modules. Using Bayesianinference and Markov Chain Monte Carlo (MCMC) procedure to do the molecularevolutionary research were fashionable recently.Results: Those bacteria consist of13kinds of species, such as Klebsiella pneumoniae,Escherichia coli, Pseudomonas aeruginosa, Acinetobacter baumannii and so on. The drugresistant rates to18kinds of antibacterial agents were relative lower than rates from otherpapers. In this experiment, we found that drug resistant rates of Klebsiella pneumoniae werelow. But, the drug resistant rates of Escherichia coli, Pseudomonas aeruginosa and Acinetobacter baumannii were relative higher than the normal. The biclustering resultrepresent that all the bacteria were divide into3different groups. The first group denoted as I,the bacteria drug resistant rates from this group were the highest. The second group denotedas II, the number of bacteria from this group was more than other groups. But, the bacteriadrug resistant rates from this group were the lowest. The last group denoted as III, the bacteriadrug resistant rates were between the group I drug resistant rates and group II drug resistantrates. The II group could also be divided into three small clusters (A, B, C) depending ondifferent antibacterial agents. Polymerase Chain Reaction (PCR) was a useful method todetect drug resistant gene. We found that the positive rate of blaIMP was14.8%and positiverate of blaVIM was0.9%. It was usually to use BLAST finding the simliar percentage in thedatabase of NCBI. The result showed that the type of IMP was IMP-1and the type of VIMwas VIM-2.Bayesian inference and Markov Chain Monte Carlo (MCMC) procedure are fashionablemethods to do the molecular evolutionary research. It was useful to analyze total MβLssequences and MβLs VIM gene type respectively. The the maximum clade credibility (MCC)tree result of VIM gene type show up those sequences were form to three clusters. The totalsubstitution rate of this gene type was2.36×E-3events/sites/year. The most recent commonancestor (TMRCA) time was from10to13years. Dynamic analysis by the Bayesian skylineplot method showed that the MβLs gene evolutionary mutation rate had a peak at the year of2003. This trend was decline after2003. But the evolutionary mutation rate was still high. Theevolutionary pressure of total MβLs gene was heavy. The evolutionary mutation rates of totalMβLs gene type was more or less at the same level. The trend of Bayesian skyline plot had aslight rise from2000to2003. The rates of mutation fluctuated, hitting a peak at2003. Therewas a steady decline from2003to2013. The evolutionary mutation rates were still in highlevel. This result showed that the genetic diversity of MβLs and selective pressure ofantibiotics had a similar tendency. |
PDF Full Text Request