2.9. Next-generation sequencing of the capture libraries
Illumina library construction. Post-capture PCR products (16 μl) were enzymatically shared by adding 2 μl 10X Fragmentase Reaction Buffer v2 and 2 μl NEBNext dsDNA Fragmentase. The sheared DNA (22 μl) was subjected to end repair, 5′ phosphorylation, dA-tailing and Illumina adaptor ligation using the NEBNext Ultra DNA Library Prep Kit for Illumina (NEB) as described by the vendor. PCR enrichment of adaptor ligated DNA was performed using NEBNext Multiplex Oligos (NEB) with index primers. The thermal profile was 30 s at 98 °C, followed by eight cycles of 10 s at 98 °C, 75 s at 63 °C and, 5 min at 72 °C. The PCR products were finally purified using the Agencourt AMPure XP system as described in the NEB protocol. The quality of the Illumina library was verified by checking the size distribution on an Agilent Bioanalyzer using a high-sensitivity DNA chip. The concentration of the Illumina library was measured by qPCR using the NEBNext Library Quant Kit for Illumina (NEB). DNA sequencing was performed using Illumina NextSeq instrument and standard reagents (Illumina).
Next-generation sequencing computational analyses. To check whether the sequencing and sample preparation went well, we did a quality check on the raw data using the FastQC tool (version 0.11.5). Low-quality read trimming, along with adapter clipping, was performed using Trimmomatic version 0.36. [35] The resulting fastq files from the trimmomatic output were then mapped against a reference genome sequence using Bowtie 2 version 2.4.1 (ref. 15). The reference genome used was E. coli K12 Using SAMtools. version 0.1.19-44428cd (ref. 16), we filtered the reads to include only those satisfying mapping quality scores  of at least 30 and then sorted the resulting bam file. Since the probes were made for genes that satisfy the requirements of the current protocol, we considered these genes as targets. The rest of the genes, along with the intergenic regions, were considered as non-targets.
We then used bed tools version 2.24.0 (ref. 17) to estimate the depth of the regions. The depth of coverage was normalized to the length of the target to allow for accurate comparison between targets of different size. Matches between the reads and the LASSO probes were determined by using BLAT v35x1. [35]