Skip to content

Latest commit

 

History

History
60 lines (47 loc) · 3.53 KB

caller_catalog.md

File metadata and controls

60 lines (47 loc) · 3.53 KB
Raw

Caller catalog format & choosing a catalog

Caller catalog format

For the --loci argument, strkit call takes a list of loci in a modified BED / TSV format, similar to methods like Straglr/Tandem-genotypes/GangSTR.

The file must be structured with a row per locus, where each row looks like:

chr#    10000    10101    [...]    AC

The important requirements here are:

  • The fields are tab-separated
  • The rows are sorted by contig, and then by starting position
  • Locus coordinates are 0-based and half-open (start is inclusive, end is exclusive)
  • The locus motif must come last in the row, but any number of fields can separate the end position and the motif.

As a result, STRkit can take myrid different TSV-type catalog formats as input, including those produced from the TRF UCSC browser track, or for GangSTR, or for Straglr.

Here are a few notes on catalogs:

  • TODO: talk about expanding coordinates, TRF multiple for same TR, ...
  • Some disease expansions can contain multiple different motifs, which may be not present in the reference genome at all (for example: CANVAS, BAFME2). As such, we provide a mechanism to specify motifs using any IUPAC code. Thus, the CANVAS and BAFME2 motifs can be represented as AARRG and AAAWK, respectively. We also add in a non-IUPAC code, X, which behaves like N in that it represents any base, but instead of giving a reward of +2 it neither penalizes nor rewards alignment, and penalizes a gap. We use this internally to represent low-confidence base calls.
  • Related to the above, this can be important for diseases such as SCA37, where the motif composition (rather than the actual copy number) is associated with disease (Seixas et al. 2017). Here, STRkit's motif-sized k-mer counting function can be used during calling with the --count-kmers flag. See the advanced usage page for more.

Choosing an existing catalog

Other researchers have done extensive work in identifying and cataloguing loci for genotyping:

  • The Tandem Repeats Finder track for the UCSC browser, available as a downloadable BED file, with the caveat that this file includes overlapping entries, and TRs may not always be represented in their most 'essential' form (e.g., using the motif TATATATA instead of just TA). Thus, some work may be required to create a desirable locus catalog.
  • The researchers behind the GangSTR short-read STR genotyping method have prepared several extensive STR catalogs for different human reference genomes, containing motifs up to 20bp in length. However, these files use 1-based closed-interval coordinates, and should be adjusted (subtracting 1 from all start coordinates) to transform them into the 0-based half-open interval coordinates when using them with STRkit.
  • We have prepared a catalog of disease-causing or disease-associated loci for the hg38 reference genome, based on the review research done by. It will be available soon (TO COME).