leoisl / GCP

Genotype confidence percentile header-only c++ implementation

A header-only c++ implementation of genotype confidence percentile (GCP) production.

Usage

Copy "GCP.h" in your project.

The requirements are:

1. Defining a struct or class holding data: let's call it ModelData.
2. Defining a genotyper class. This must have:
   • A constructor taking a ModelData object.
   • A method called get_genotype_confidence, with signature double get_genotype_confidence(), returning the confidence of the genotype call.
3. Overriding a method to simulate data produced as a ModelData.

Only point 3 is actually required by GCP. Points 1 and 2 would usually be internally implemented as part of a tool's genotyping.

Follow this example for a concrete walkthrough!

Users

• pandora
• gramtools

Credit

Originally based on minos Genotype confidence percentile (https://github.com/iqbal-lab-org/minos/blob/ed975d1099c6403eff79b04b0d2064eebfa95e73/minos/genotype_confidence_simulator.py) by Martin Hunt