A Normalization and Copy Number Variation Detection Method for Whole
Exome Sequencing
Description
A normalization and copy number variation calling procedure for
whole exome DNA sequencing data. CODEX relies on the availability of
multiple samples processed using the same sequencing pipeline for
normalization, and does not require matched controls. The normalization
model in CODEX includes terms that specifically remove biases due to GC
content, exon length and targeting and amplification efficiency, and latent
systemic artifacts. CODEX also includes a Poisson likelihood-based recursive
segmentation procedure that explicitly models the count-based exome
sequencing data.