Ion Torrent sequencing, data processing and analysis

Amplicon concentration was determined using a NanoDrop™ and 25 µl of DNA (26 pM) was prepared for emulsion PCR. Emulsion PCR was performed using the Ion PGM™ Hi-Q™ OT2 Kit (Life Technologies), according to manufacturer’s protocol. First, a unique DNA amplicon was amplified and bound to a single Ion Sphere Particle (ISP) by emulsion PCR. Amplification primers that bind to A and P1 adapters were used for clonal amplification so that each ISP was covered with many copies of the same DNA fragment. Second, because the A primer was biotinylated, template positive ISPs could be isolated using Ion Torrent enrichment beads and non-templated ISPs were removed. Third, dsDNA anchored to the ISPs was denatured. This allowed the ISPs with ssDNA to go into solution while the biotinylated strand remain bound to enrichment beads. The solution containing ssDNA enriched ISPs was used for next generation sequencing.

Next generation sequencing was performed using the Ion PGM™ System (Thermo Scientific). Ion 318 chips and Ion PGM Hi-Q Sequencing Kits were used according to manufacturer’s protocol. Base calling, chip analysis, and barcode separation were performed using the Ion Torrent Server software version 5.0. Chip analysis included percentage ISP loaded, percentage of enriched ISPs, percentage polyclonal reads (ISPs with multi-type DNA templates), and percentage of low quality reads. Total raw sequences were retrieved as FASTQ format from Ion Torrent server and counted and plotted afterwards.

Queries on each Ion Torrent read was done with a computational tool for pattern recognition called regular expression (regex)³⁹. Each read was scanned for an established known set of sequence (represented as framework in Table 1) followed by the number of nucleotides equal to the length of the library sequences (in our case, 21) and then the cleaved-XhoI site (‘CTC’). Once all reads had been scanned, counts of each unique sequence were saved to a file using script (python-2.7.9) provided as additional file named as “Script file” according to the method described in Supplementary Fig. ^S4A.