Transition and transversion as a quality of SNP call set

Question

[hpapoli]

Hello,

One of the quality checks for a SNP call set is transition/transversion ratio.

By applying a series of filters, the ratio increases which is said to be an indication of a set with less false positives.

Why with filtering, TS/TV ratio increases?

Thanks!

[ATpoint]

Please provide sources to your statements.

Answer 1

[Santosh Anand]

The whole genome Ts/Tv ratio is around 2.1. If your sample's Ts/Tv deviates significantly from this, most probably there are false positives, and removing them will make the data more close to reality and close to actual Ts/Tv. However, bear in mind that this is an ad-hoc approach and valid (tested) only for human whole genome. Targetted regions and other genomes can have different Ts/Tv ratios. See some interesting discussions here

Ti/Tv Ratio Confirms Snp Discovery. Is This A General Rule?