checkphase_summary

#!/bin/bash

#
#  Copyright (C) 2018-2023 by Lars Wienbrandt,
#  Institute of Clinical Molecular Biology, Kiel University
#  
#  This file is part of Checkphase.
#  
#  Checkphase is free software: you can redistribute it and/or modify
#  it under the terms of the GNU General Public License as published by
#  the Free Software Foundation, either version 3 of the License, or
#  (at your option) any later version.
#  
#  Checkphase is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.
#  
#  You should have received a copy of the GNU General Public License
#  along with Checkphase. If not, see <https://www.gnu.org/licenses/>.
#

# summarizes information from checkphase sample-wise outputs
if [[ $# == 0 ]]; then
  echo "Usage: checkphase_summary <list of checkphase sample-wise output files>"
  exit 1
fi

awk 'BEGIN {
 Mcheck=0
 Totswerr=0
 Totgterr=0
 Totgterrsoft=0.0
 Nquery=0
}
# add checked variants from this file
/Checked variants:/{Mcheck += $3; next}
# ignore header
/Idx/{next}
# general
{
 # number of samples based on indices
 if ($1 >= Nquery) Nquery = $1
 # sum up errors
 Totswerr += $2
 Totgterr += $3
 Totgterrsoft += $4
 # sum up per ID
 swerr[$1] += $2
 gterr[$1] += $3
 gterrsoft[$1] += $4
}

END {
 Nquery++ # indices are 0-based

 # Summary output of:
 # - Number of files
 # - Number of samples
 # - Number of (checked) variants
 # - Total sw errors
 # - Av. sw errors
 # - Av. sw error rate
 # - Total gt errors
 # - Av. gt errors
 # - Av. gt error rate
 print "# Num files:\t" ARGC-1 # first argument is the awk program itself
 print "# Num samples:\t" Nquery
 print "# Num (checked) variants:\t" Mcheck
 print "# Total sw errors:\t" Totswerr
 print "# Av. sw errors:\t" (Totswerr/Nquery)
 print "# Av. sw error rate:\t" (Totswerr/Nquery/Mcheck)
 print "# Total gt errors (hard):\t" Totgterr
 print "# Av. gt errors (hard):\t" (Totgterr/Nquery)
 print "# Av. gt error rate (hard):\t" (Totgterr/Nquery/Mcheck)
 print "# Total gt errors (soft):\t" Totgterrsoft
 print "# Av. gt errors (soft):\t" (Totgterrsoft/Nquery)
 print "# Av. gt error rate (soft):\t" (Totgterrsoft/Nquery/Mcheck)

 # Per samples output of:
 # - sw errors
 # - sw error rate
 # - gt errors (hard)
 # - gt error rate (hard)
 # - gt errors (soft)
 # - gt error rate (soft)
 print "Idx\tsw err\tsw err rate\tgt err (hard)\tgt err rate (hard)\tgt err (soft)\tgt err rate (soft)"
 for (idx=0; idx<Nquery; idx++) {
  print idx "\t" swerr[idx] "\t" swerr[idx]/Mcheck "\t" gterr[idx] "\t" gterr[idx]/Mcheck "\t" gterrsoft[idx] "\t" gterrsoft[idx]/Mcheck
 }
}' $@