FilterAlignmentArtifacts error

[kasiatom]

Bug Report

Tool

FilterAlignmentArtifacts

Affected version(s)

4.2.0.0 and 4.1.9.0, run from local jar or docker

Description

FilterAlignmentArtifacts crushes repetitively in the same position of the input mutect2 vcf m2.vcf.gz (chrX:63457865). But, it finishes task successfully when only the last variant from the output file is present in the input vcf file.
I cut the input vcf around the troublesome variant to reproduce the error on a smaller input and:

1. Error did not occur when the input was very small
2. FilterAlignmentArtifacts finished run at different variant (chrX:73769127) when analyzing the smaller input (test.vcf.gz)

The log issue looks very similar to that described here #7162), but the Problematic frame information is different.
As suggested in this issue #5690, the problem disappears when using gatk 4.1.3.0 on the same inputs.

log:

17:37:20.674 INFO  ProgressMeter -       chr20:43968267             10.6                 44000           4132.2
17:37:38.646 INFO  ProgressMeter -       chr22:22736335             10.9                 45000           4110.5
17:37:52.672 INFO  ProgressMeter -         chrX:7000139             11.2                 46000           4113.9
17:38:05.421 INFO  ProgressMeter -        chrX:26360893             11.4                 47000           4125.0
17:38:17.207 INFO  ProgressMeter -        chrX:44917184             11.6                 48000           4141.4
17:38:29.312 INFO  ProgressMeter -        chrX:77681733             11.8                 49000           4155.3
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fc0ccec5cdb, pid=15987, tid=15988
#
# JRE version: OpenJDK Runtime Environment (11.0.11+9) (build 11.0.11+9-Ubuntu-0ubuntu2.18.04)
# Java VM: OpenJDK 64-Bit Server VM (11.0.11+9-Ubuntu-0ubuntu2.18.04, mixed mode, sharing, tiered, compressed oops, g1 gc, linux-amd64)
# Problematic frame:
# C  [libc.so.6+0x97cdb]  cfree+0x31b
#
# Core dump will be written. Default location: Core dumps may be processed with "/usr/share/apport/apport %p %s %c %d %P %E" (or dumping to /home/kt/core.15987)
#
# An error report file with more information is saved as:
# /home/kt/hs_err_pid15987.log

Steps to reproduce

My commands:

gatk --java-options "-Xmx11g"  \
     FilterAlignmentArtifacts \
      -R GRCh38.no_alt_analysis_set.fa   \
      -V in.vcf.gz \
      -I bamout.bam \
      --bwa-mem-index-image Homo_sapiens_assembly38.fa.img  \
      --num-regular-contigs 194 \
     --max-reasonable-fragment-length 2000 \
     --drop-ratio 0.1 \
     --indel-start-tolerance 8 \
     -O out.vcf.gz

I copied the input vcfs (small: test.cf.gz and initial: m2.vcf.gz), bamout and "hs_err_pid.logs" to gs://iseq/kt/strange-bug/
I hope you can access them.
Best,
Kasia