A question about merge parameter -p

[zmyzhaomingyang]

hello, i am using delly to analyse my data. I am using delly merge according to Germline SV calling, and the commond i used is delly merge -p -c -m 51 -n 1000 -o path bcf1 bcf2 ... bcf180 the question is：

1. after genotype on every sample, i use the command bcftools query -f '%CHROM\t%POS\t%END\t%ID\n'
   to get the bed file of each sample eacept BND, and caculate the length using POS and END found that the minimum is 50 which is not a value equal or bigger than 51, but the max value of length is 999. I want to know how does delly caculate the length of sv when delly merge?
2. Is it suitable for me to use Germline SV calling methods? My data is wgs data without control and the dna is from muscle.
3. I want to filter sv in every one sample through length after genotype according to Germline SV calling, but i just hava a reference genome and without control sample which makes me can not use somatic sv calling , also can not use delly filter in germline sv calling which needs at least 20 samples? how should i do ? can i use the method to caculate the length in question 1 and do filter?

very very very very thank you for answering my questions!!!!!!!

[tobiasrausch]

If there are only a few samples available it's usually best to call them jointly:

delly call -g <genome.fa> -x <excl.tsv> -o delly.bcf sample1.bam sample2.bam sample3.bam

The filtering then depends on your study. Usually the PASS and PRECISE variants are good choices to start with.