Chunk #20 — MATERIALS AND METHODS — Statistical model — Calibration to type I error

Source

QuantiSNP: an Objective Bayes Hidden-Markov Model to detect and accurately map copy number variation using SNP genotyping data.

Embedded

yes

Text

In order to provide calibration of our model to a fixed Type I (frequentist) error rate, we generated 100 pseudo-normal datasets for both the Human-1 and HumanHap300 SNP coverage. This was achieved by randomly sampling log R ratio and B allele frequency values from an individual assayed using the Human-1 and HumanHap300 platforms. These normal datasets allow us to quantify the false positive rates of the algorithms and calibrate the method to a user-specified rate.