Issue in SNP calling

[faizee-ali]

I am trying to use ANGSD's for SNP callign from whole genome low coverage reads, through a pipeline that uses Indel Realignment through GATK-3.6, to finally construct genotype likelihoods with angsd 0.940-dirty (htslib: 1.16).

When I run the angsd command:

angsd-GL 2 -doGlf 2 -doMajorMinor 1 -minInd 5 \
    -doMaf 1 -doGeno 8 -doPost 1 -skipTriallelic 1 \
    -minMapQ 20 -minQ 20 -SNP_pval 2e-6 -minMaf 0.02 \
    -setMinDepth 3 -setMaxDepth 200 -uniqueOnly 1 \
    -remove_bads 1 -only_proper_pairs 1 -C 50 -baq 1 \
    -doBcf 1 -doCounts 1 --ignore-RG 0 \
    -out "/mnt/d/Faizee/wgs/bams/angsd" \
    -nThreads 20 \
    -bam "/mnt/d/Faizee/wgs/bamlist.txt" \
    -ref "/mnt/d/Faizee/wgs/ref/Parus_major.fna"

I encounter the following error. For context I have 6 bam files names as *RI.bam, generated from the IndelRealigner function in GATK-3.6.

Error:

-> angsd version: 0.940-dirty (htslib: 1.16) build(Sep 13 2024 01:04:18)
-> angsd -GL 2 -doGlf 2 -doMajorMinor 1 -minInd 5 -doMaf 1 -doGeno 8 -doPost 1 -skipTriallelic 1 -minMapQ 20 -minQ 20 -SNP_pval 2e-6 -minMaf 0.02 -setMinDepth 3 -setMaxDepth 200 -uniqueOnly 1 -remove_bads 1 -only_proper_pairs 1 -C 50 -baq 1 -doBcf 1 -doCounts 1 --ignore-RG 0 -out /media/birdlab/naman_hdd2/Naman/culicicapa_project/angsd/angsd -nThreads 20 -bam /media/birdlab/naman_hdd2/Naman/culicicapa_project/bamlist.txt -ref /media/birdlab/naman_hdd2/Naman/culicicapa_project/ref/Parus_major.fna
        -> Inputtype is BAM/CRAM
[multiReader] 5 samples in 6 input files
        -> You have choosen to supply a very high number of threads. The number of threads will be capped at 8
        -> Reading fasta: /media/birdlab/naman_hdd2/Naman/culicicapa_project/ref/Parus_major.fna
        -> SNP-filter using a pvalue: 2.000000e-06 correspond to 22.595043 likelihood units

**[bcfoutput]     Please add the following parameters
                 '-gl 1 -dopost 1 -domajorminor 1 -domaf 1 -dobcf 1 --ignore-RG 0 -dogeno 1 -docounts 1'

        -> Inputtype is BAM/CRAM
        -> Parsing 6 number of samples
munmap_chunk(): invalid pointer
Aborted (core dumped)**

However, when I run the same command for a bamlist.txt containing a single sample I do not encounter this error and angsd runs through.

[TonyKess]

I have seen this error occur when you have duplicate entries in your bam list. Try doing cat | sort | uniq > bamlistuniq.tsv or just going through it manually and see if that fixes the problem.