grm

After coverage, truvari grm is run. GRM, or graph-reference mappability, is a procedure to take kmers centered on edges in a variant graph and remap them to the reference. GRM then collects metrics such as the number of hits of the kmer through the reference and their mapping quality.

This step produces a dataframe that is used by the qual_classifier step.

(bg7)$ truvari anno grm -h
usage: grm [-h] -i INPUT -r REFERENCE [-o OUTPUT] [-k KMERSIZE] [-t THREADS]
            [--debug]

Maps graph edge kmers with BWA to assess Graph Reference Mappability

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT, --input INPUT
                        Input VCF
  -r REFERENCE, --reference REFERENCE
                        BWA indexed reference
  -o OUTPUT, --output OUTPUT
                        Output dataframe (results.jl)
  -k KMERSIZE, --kmersize KMERSIZE
                        Size of kmer to map (50)
  -t THREADS, --threads THREADS
                        Number of threads (48)
  --debug               Verbose logging