Understanding biser output

[YingChen94]

Hello, thank you for this wonderful tool! I notice that sometimes the genomic position is much larger than the actual alignment. For example, region 1 is 9326bp and region 2 is 12585bp, but the alignment is only 1789bp. How should I interpret the genomic position, and how does biser decide on the boundary? Thank you in advance for your help!

chr16_pat 2101992 2111318 chr16_pat 3962784 3975366 pat:pat 2.3 + - 1789 1789

[YingChen94]

Also it seems that occasionally it will report the same genomic region as a SD pair...

chr1A_pat 46524596 46530781 chr1A_pat 46524596 46530781 pat:pat 7.8 + - 2290 2347 1M565S565N50M75S75N1M14I118M3D58M1D166M19D122M10D59M10D512M2611N29M2611S542M10I59M10I122M19I164M1I55M3I123M14D1M89S89N50M555S555N1M X=2.9;ID=4.9 2.9 4.9

[inumanag]

Hi @YingChen94,

Thank you for using BISER!

1. You need to use CIGAR string to parse the alignment. This might be due to a large indel, or it is just a weird artefact.

2. This is a palindrome (as the strands are + and -). BISER is not yet able to filter them out; you will need to have a post-processing script for them if you don't want them.

[YingChen94]

Hi @inumanag , thank you for your answer! Do you have any suggestions on general rules of filtering BISER SDs? Sorry I am pretty new to the topic, and I want to investigate whether inversions may be associated with SDs (in a non-model organism). Do you have any suggestions?

[inumanag]

Hi @YingChen94,

Initially, I'd recommend filtering out overlapping SDs, identical SDs and palindromes (known BISER artifacts). Also increase the sensitivity: use --max-error 15 --max-edit-error 10 (or even lower) for your application.

Note that overlapping SDs might be actually biologically useful, esp. if it is a part of core block or if the mate is the reverse (-) strand.

And yes, there are inversions that are SD-induced; easiest way to check for those is to look for SDs whose second mate is reversed (- strand).

[YingChen94]

Hi @inumanag , thank you very much for your inputs! They are super helpful. Just to confirm I understand it correctly, is --max-error filtering the 8th column score (i.e. score = mismatch rate X: + gap rate ID:), and --max-edit-error filtering the mismatch rate (X:)?