search

schneebergerlab / syri

Synteny and Rearrangement Identifier
https://schneebergerlab.github.io/syri/
MIT License
323 stars 35 forks source link

Merge SyRI VCF results #127

Closed ahishsujay closed 2 years ago

ahishsujay commented 2 years ago

Hi! I've recently used SyRI and have ~40 VCF files I would like to merge. Would you recommend using SURVIVOR (https://github.com/fritzsedlazeck/SURVIVOR) in order to merge these SVs? I plan on separating SNPs and INDELs, merging them using bcftools merge and using SURVIVOR to merge the remaining SVs.

Please let me know if you think this is a good idea, or any suggestions.

Thanks!

mnshgl0110 commented 2 years ago

Hi Ahish,

As far as I know, merging of non-SNP and non-indel variants is still not possible. In fact, we are developing methods that can merge complex non-linear rearrangements identified using genome comparisons.

Regarding SURVIVOR, I have not used the merge option of SURVIVOR extensively so I cannot comment on its applicability. However, my impression is that it works best for merging large indels and is designed for read based SV identification. I would suggest that you give it a try. SURVIVOR is quite easy to install and use, and is quite fast, so better to test and see if the output suits your project.

Best Manish

ahishsujay commented 2 years ago

Hi Manish,

Thanks for responding so quick! I see, looking forward to using the methods you talked about when it's developed.

I have used SURVIVOR before for merging SVs from multiple SV callers, but was a little wary of using it on SyRI's results as I am not sure if it makes sense to "merge" the annotations which are not a SNP, INS or DEL if that makes sense. Also from what I read, SURVIVOR isn't suitable for handling variants < 20 bp.

Nonetheless, thank you for your inputs. I appreciate it!

Best, Ahish