MFA data analysis

Paired-end raw datasets from an Illumina HiSeq 2000 sequencing platform (obtained from Edinburgh Genomics) were mapped against the genomic sequence of the reference strain ‘BW27784’ using BWA sequence aligner (version 0.7.11) and subsequently analysed using SAMtools (version 1.2). ‘BW27784’ is a modified version of E. coli K12 MG1655 (NC000913.3) including all published differences between the strains [62,63]. Replication profiles of exponentially growing cultures were calculated by normalizing to the number of uniquely mapped sequence reads (to correct for differences in depth of sequencing) and then to the normalised reads of a non-replicating stationary-phase wild-type culture (a Rec⁺ strain without palindrome) to correct for differences in sequence-based recovery across the genome. An in-lab R-script (available on request) has been used to calculate the enrichment (normalised read depth) in 1 kb non-overlapping windows across the genome and a non-parametric smoothing method (LOESS, Local regression) has been applied to the data points of the replication profiles of each strain.