Remove a row if two specific columns have similar/share contents

Question

[waqasnayab]

Hi,

I have a line in VCF file:

chr1    11785112        .       T       G,T     .       str10   AC=6,0;ADP=72;AN=8;SF=0f,1f;STATUS=.    GT:ADR:ABQ:RDR:FREQ:RDF:ADF:PVAL:AD:SDP:RBQ:DP:RD:GQ    0/0:0:0:0:0%:0:0:1E0:0:46:0:46:0:0      1/1:5:47:0:100%:0:82:6.9142E-52:87:87:0:87:0:255     1/1:4:50:0:100%:0:51:1.0149E-32:55:84:0:84:0:255        1/1:3:49:0:100%:0:31:3.5146E-20:34:71:0:70:0:194

and might be many more like this. I want to remove such lines where REF base is repeated in the ALT bases. I tried:

awk '!($4~$5)' FA-MO-D1B-D1S.mpileup.output.snps.indel_srt_smplrnme_d1b_d1s.vcf > family.vcf

but no luck.

Any help is appreciated.

Waqas.

[cpad0112]

There are dedicated tools such as bcftools, gatk or vcftools as suggested in Remove positions that are non-variant in a subset of samples from a vcf file

[waqasnayab]

Thanks Pierre, it worked like a charm!

Regards,

Waqas.

Answer 1

[Pierre Lindenbaum]

awk -F '\t' '/^#/ {print;next;} {N=split($5,a,/[,]/);P=1;for(i=1;i<=N;i++) {if(a[i]==$4) {P=0;break;}} if(P) print}' in.vcf