inst/shiny-app/help.md

In airr-community/rep-cred: Repertoire Credibility

Good analysis requires good data, but how can how can a high-quality AIRR dataset be distinguished from a low-quality one? RepCred fills this need by identifying and reporting several key metrics that can be indicators of potential problems. Modeled after fastqc, RepCred provides an easily digestible summary that can be used by novices and experts alike.

The only requirement to run RepCred is a file describing IG and/or TCR rearrangements in the AIRR TSV format. You can find a small example repertoire here.

By default, all repertoires are randomly down-sampled to 5000 rearrangements; we find that this still provides an accurate assessment of repertoire credibility while maintaining a reasonable run time.

Finally, if a custom (non-IMGT?) database was used to annotate the rearrangements, this should be provided to RepCred as well, so that SHM and related statistics can be calculated accurately.

To get help or report a bug, please file an issue on RepCred’s github page at https://github.com/airr-community/rep-cred/issues.

RepCred is designed to be modular and accept new functions/metrics with ease. We welcome pull requests (please see https://github.com/airr-community/airr-standards/blob/master/CONTRIBUTING.rst). Other feature requests will be evaluated and prioritized.

airr-community/rep-cred documentation built on July 13, 2024, 1 p.m.