Question: filtering structural variants from normal-tumor pairs (whole-genome sequencing)
0
gravatar for Bogdan
2.4 years ago by
Bogdan850
Palo Alto, CA, USA
Bogdan850 wrote:

Dear all,

we have been sequencing pediatric tumors by using Illumina technology (normal-tumor pairs, paired end reads of 150nt each), and I have been using several SV callers (including DELLY, MANTA, LUMPY) in order to call the Structural Variants.

Considering the initial files of SV calls, what is the set of criteria you would recommend for filtering ? particularly, the number of PR (paired-end reads) or SR (split - reads) ? thank you,

-- bogdan

filtering lumpy manta delly sv • 1.3k views
ADD COMMENTlink modified 2.4 years ago by d-cameron2.1k • written 2.4 years ago by Bogdan850
6
gravatar for d-cameron
2.4 years ago by
d-cameron2.1k
Australia
d-cameron2.1k wrote:

This depends on the relative cost of false positives vs false negatives. What FDR and sensitivity are you aiming for?

I've done some fairly extensive benchmarking of SV callers for which you can find germ-line and simulation results at http://shiny.wehi.edu.au/cameron.d/sv_benchmark to given you some idea of the trade-off between FDR and sensitivity for various callers. The DREAM Somatic Mutation Calling Challenge results also give you a basic idea of these trade-offs.

In my somatic usage of GRIDSS, I've found that requiring assembly support has been a very effective filter - something that you could also do for manta. As for direct read support, the relative strength of RP and SR support depends on the library fragment size distribution and will be different for each library. For 2x150bp sequencing, you'll have weak RP support even for real variants in libraries with mean fragment size <300bp. Ideally, you would filter on variant quality directly, instead of RP and SR independently but not many SV callers actually report a variant quality score, and none that I know of report an even remotely calibrated quality score.

Disclaimer: I am the author of GRIDSS.

ADD COMMENTlink written 2.4 years ago by d-cameron2.1k

Dear Daniel, took a quick look in your paper and it looks fascinating : GRIDSS is among the first algorithms, that combines assembly, SR and PR analysis. Wondering how extensive have you tested it in on cancer samples, for somatic SV detection. We can also take the conversation via email, if you'd like. Thanks.

ADD REPLYlink written 2.4 years ago by Bogdan850
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 542 users visited in the last hour