A Snakemake workflow for mapping transposon integration sites from tagmentation mapping data (https://www.biorxiv.org/content/10.1101/037762v5.full), either from pooled NGS libraries or from individual Sanger reads of clonal lines.

Two complementary ways of finding an integration site are supported, and can be run together or on their own:

• NGS: paired-end tagmentation libraries are mapped, filtered down to read pairs anchored at an ITR primer, and turned into peaks and then single-base insertion sites. Suited to pooled populations, and reports both sides of an insertion when both were captured.
• Sanger: individual long reads from a clonal line - typically .ab1 traces from one well per clone - are mapped directly, and the exact cassette/genome junction is read off each read. Suited to identifying the integration site in a specific clone.

When both are configured for the same material, Sanger sites can be cross-validated against the NGS data (validate_sanger_with_ngs): a Sanger read that only worked from one ITR primer can still be trusted if the NGS library shows an insertion at the same position from both sides.

Alongside the site calls, the workflow writes a set of summary tables to stats_folder - NGS mobilization rate and insertion-site sidedness, Sanger read/QC counts and per-clone forward/reverse coverage, and the Sanger-vs-NGS validation outcome - and assembles them into one report.md. A founder/control clone that still carries the cassette at its original, pre-mobilization locus can be flagged as such (original_insertion_site), rather than counted as a new integration.

See config/README.md for how to configure the workflow, and config/example_config.yaml for a fully commented example. A minimal test dataset lives under .test/.

If you use this workflow in a paper, don't forget to give credits to the authors by citing the URL of this (original) repository and its DOI (see above).