Missing genotype for merged genotype calling.

[mnshgl0110]

The script for diploid calling does not call genotypes. Probably this line needs fixing:

cuteSV/src/cuteSV/diploid_calling.py

Line 38 in 2cfaaf6

if 'cutesvh1' in i:

[Meltpinkg]

Hello, @mnshgl0110

This script is based on the pre-processing to achieve the diploid assembly based SV calling. Maybe you have missed the 4th step here to add haplotype-unique tag, which leads to the missing genotypes. When I try following the steps in wiki, this script can work well. By the way, you can also compare the workflow in the wiki to have another check.
If there are other problems, please provide the error messages and the running commands, and I will make a further check.

Best,
Shuqi

[mnshgl0110]

Hi Shuqi,

Thanks for sharing the documentation. I think, I followed some earlier version of documentation where this step was missing. I will try with this newer documentation.

May I ask an additional related question.. Let's say there is a 10MB cut-and-paste event between Chr1 and Chr2, such that in the diploid query genome this 10MB sequence is present on Chr1 in one haplotype and on Chr2 on the other haplotype. In the reference, this 10MB sequence is on Chr1. In this setup, it is also possible that there are smaller indels compared to the reference within this 10MB regions. These indels could either be present on one of the two copies or on both.
How would cuteSV handle this case? Would it report 10MB indels on Chr1/Chr2 and not report the smaller variations or would it try to report them as well? What would be the guiding principle for determining the genotypes?

[Meltpinkg]

Hello, @mnshgl0110

Based on your description, this appears to be a complex SV involving both a translocation and insertion/deletion events. When a 10MB sequence is cut-and-pasted from chr1 to chr2, cuteSV will recognize it as a translocation. Since it presents in only one haplotype, it will be a heterozygous translocation.
For the embedded indels within this translocated region, if they are existed in the CIGAR field in alignment, cuteSV will extract them from alignments and report them.
For determining the genotypes, cuteSV will check the number of haplotypes that contains the corresponding SV alleles to assign the genotype.

Hope it will help!

Best,
Shuqi