CNGPLD: case–control copy-number analysis using Gaussian process latent difference

Abstract

Motivation Cross-sectional analyses of primary cancer genomes have identified regions of recurrent somatic copy-number alteration, many of which result from positive selection during cancer formation and contain driver genes. However, no effective approach exists for identifying genomic loci under significantly different degrees of selection in cancers of different subtypes, anatomic sites, or disease stages. Results CNGPLD is a new tool for performing case-control somatic copy-number analysis that facilitates the discovery of differentially amplified or deleted copy-number aberrations in a case group of cancer compared to a control group of cancer. This tool uses a Gaussian process statistical framework in order to account for the covariance structure of copy-number data along genomic coordinates and to control the false discovery rate at the region level. Availability CNGPLD is freely available at https://bitbucket.org/djhshih/cngpld as an R package. Supplementary information Supplementary data are available at Bioinformatics online.

Identification and Analysis of Sex-Biased Copy Number Alterations

“…The previous studies have detected the sex-biased CNAs in multiple cancers by applying conventional statistical method such as Fisher’s exact test to test each loci independently [ 9 , 10 ]. However, the CNAs of adjacent loci are highly linked, which can cause standard adjustment methods to fail in controlling the FDR [ 14 ]. To dissect the sex-biased CNAs in each cancer type, we used a published framework called CNGPLD that is specifically designed for comparative copy-number analysis such as discovering differentially amplified or deleted regions in metastatic cancer compared to primary cancer on the basis of a Gaussian process prior and is conductive to disclose more genuine differential CNAs [ 14 ].…”

“…However, the CNAs of adjacent loci are highly linked, which can cause standard adjustment methods to fail in controlling the FDR [ 14 ]. To dissect the sex-biased CNAs in each cancer type, we used a published framework called CNGPLD that is specifically designed for comparative copy-number analysis such as discovering differentially amplified or deleted regions in metastatic cancer compared to primary cancer on the basis of a Gaussian process prior and is conductive to disclose more genuine differential CNAs [ 14 ]. Although the distribution of CNAs along genome are similar between the sexes, for example, Fig.…”

“…We used the case-control copy-number analysis using Gaussian process latent difference (CNGPLD) method to identify the genome regions with differential copy number aberrations between male and female patients [ 14 ]. CNGPLD is a tool that uses a statistical framework (Gaussian process latent difference model) to capture covariance structure of copy number data between 2 groups and achieves controlling the FDR at the region level [ 14 ]. Here, we applied this approach to compare 2 copy number data distribution of male and female cohorts for identifying genomic regions with different CNAs between sexes.…”

“…Nevertheless, only limited functional variation events are found, and it still needs in-depth investigation of linking this molecular difference to the cancer outcomes between the sexes. In this work, through a strict pipeline on comparing the copy number profiles of male and female patients presented by [ 14 ], we identified the CNA regions showing sex difference in 16 cancer types using the TCGA database. Further analysis of these sex-biased CNAs unveils crucial biological processes potentially linked to distinct cancer outcomes between the sexes and sex-related prognostic biomarkers in specific cancer types.…”