non-SNP variants in a vcf merge file
Hello,
I have used bwa for alignment and platypus for variant calling and filtered only SNPs using bcftools. I then merged the vcf files with only SNPs for some samples. But the merged file shows some odd positions with variants other than SNPs. Although individual vcf files does not show these calls. I used this code:
bwa mem ref.fa S_1.fastqS_2.fastq | samtools sort -o sorted.bam
samtools index sorted.bam
python Platypus.py callVariants --bamFiles=sorted.bam --refFile=ref.fa --output=1.vcf
bcftools view --types snps 1.vcf -o 2.vcf
bgzip -c 2.vcf > 2.vcf.gz
tabix -p vcf 2.vcf.gz
bcftools merge --file-list list.txt -O v -o merge.vcf
I have pasted one weird call from the merged file:
Chr POS ID REF ALT 1 2 3
ch1 1470 . CGCTCC TGCTCC,TACTCC,TGCTCA TGCTCC NA TGCTCC
• 1,972 views
•
link
updated
by
Kevin Blighe
•
written
by
evelyn
•
written
by
evelyn
1 answer
Some variant callers just call variants in this way. You can probably mitigate these by 'normalising' your data to a reference:
#1st pipe, splits multi-allelic calls into separate variant calls
#2nd pipe, left-aligns indels and issues warnings when the REF base in your VCF does not match the base in the supplied FASTA reference genome
bcftools norm -m-any merge.vcf | \
bcftools norm -Ob --check-ref w -f human_g1k_v37.fasta > merge_norm.vcf ;
Here, human_g1k_v37.fasta, would be whichever reference genome FASTA against which your variants were originally called.
Kevin
• 0 views
•
link
written
by
Kevin Blighe
Please let me know if it works....
• 0 views
•
link
written
by
Kevin Blighe
Log in to answer this question.
More posts like this
-
SNPs not filtered from a GATK vcf file
written by evelyn
-
variant calling with platypus
written by evelyn
-
how to merge gvcf using bcftools
written by evelyn
-
[W::bcf_hdr_check_sanity] GL should be declared as Number=G
written by evelyn
-
bcftools merge duplicate names
written by evelyn
-
bcftools-merge for vcf files
written by evelyn
-
vcf from sorted bam
written by evelyn
-
Incorrect number of AD fields (2) at 1:878906 cannot merge
written by jaybee •
-
My output of variant calling is different by samtools and platypus
written by Anop Singh Ranawat •
-
Three different VCF callers ignoring the same columns with variants
written by dmathog •
Hello evelyn ,
what's the content of
list.txt? Also your output example doesn't look like a valid vcf file. I'm missing theQUAL,FILTER,INFOandFORMATcolumns.fin swimmer
Hello @finswimmer,
list.txtcontains list of vcf.gz samples to be merged. In order to show the problem only, I mistakenly broke the format. And I have more samples, I am just showing three here including the problem.