Delly misses translocation chr14-18 in our positive control

[Andrea stacchetti]

Good Everyone and Tobias, 

I am writing here seeking for help; we are using delly for calling SV, in particular to call the translocation.
However, we are experiencing quite a issue.
We have a cell line that is positive for the translocation of the chr 14-18, in our data once we try to call the translocation delly is missing it.

Nevertheless, we are sure there is that SV since we can see it by the file bam.

More in details, delly calls in "Low Quality - Imprecise" a BND in the chr 14 from the chr 18; but once delly constructs the consensus and then we tried to BLAT it that sequence has 100% match in the chr 16. Furthermore, it doesn't seen the same translocation on the other way around, we mean chr18-14.

We are using this command:

delly call --genome $ref_GRCh38 -t BND -q 0 -r 10 -c 5 -m 10 -n 50 -o DOHH2_CAPPSEQ_S362.bcf DOHH2-CAPPSEQ_S36_merged.bam $Germinal

Thanks in advance for your time,

Best,

Andrea

[Tobias Rausch]

Hi,

Delly outputs for each SV breakpoint only one record. If you do want to split the translocations into 2 records for each chromosome, you can use this script: https://github.com/dellytools/delly/tree/main/scripts

python scripts/delly2bnd.py -v delly.bcf -r hg38.fa -o delly.bnd.bcf

I am not sure if that is the problem you are facing or if the translocations paired-ends are somehow missed by delly?

Best, Tobias

--
You received this message because you are subscribed to the Google Groups "delly-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to delly-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/delly-users/a4b51dd5-9739-46f2-aea9-caaf4f61c3b3n%40googlegroups.com.

[andrea stacchetti]

Hi Tobias,

Thank you very much for your answer! We will use the python record for splitting the translocation in 2 records.

However, the problem we are facing is that delly is not calling the translocation chr14-chr18 in our positive control. More in detail, when delly claims to have called the translocation in exam, the consensus sequence it builds is found by BLAT in Chr16 with 100% of match. Nor chr14 nor chr18 are involved in the consensus sequence of the translocation.

Thanks a lot, Andrea

 

[Tobias Rausch]

Hi Andrea,

Sounds like the translocation is flanked by repeats or a segdup or I may misunderstand your problem. Do you see high-quality SV supporting paired-ends and/or clipped reads in IGV?

Best, Tobias