Genomic-DNA sequencing and data analysis.

The library for genomic-DNA sequencing was prepared according to the TruSeq DNA sample preparation protocol (Illumina). Briefly, 1 μg of genomic DNA was sonicated into fragments with a median length of 400 bp; after end repair, indexed adapters were ligated at the DNA fragment ends, and the libraries were quantified by quantitative real-time PCR (qPCR) using Kapa Library Quant kits (Kapa Biosystems). After a short amplification step, the library was sequenced on an Illumina GAIIX sequence analyzer to generate 85-bp paired-end reads. The raw reads were individually mapped to the E. coli MC4100 genome (RefSeq accession number {"type":"entrez-nucleotide","attrs":{"text":"HG738867","term_id":"557270520","term_text":"HG738867"}}HG738867) using the accurate alignment BWA mem algorithm (81) allowing 1% error. Removal of duplicated reads was performed with SAMtools (82); only high-quality reads having mapping quality score (MAQ) values of >30 were used for the analysis of variant detection. A VCF file containing all the variants for each sample relative to E. coli MC4100 was obtained by using SAMtools and Bcftools (82) and filtered for low-quality variants. SNVs having coverage lower than five high-quality reads were discarded. Predicted indel mutations having coverage lower than six high-quality reads were discarded. Then, the VCF files were analyzed using SNPeff version 4.0 (83), and high-quality SNVs and indels were subsequently annotated to determine their effects and impacts on coding sequences.