Discovery of copy number variations (CNVs) from exome read depth using XHMM (eXome-Hidden Markov Model)
Call copy number variation (CNV) from next-generation sequencing data, where exome capture was used (or targeted sequencing, more generally).
Input files:
- list of exome targets ("EXOME.interval_list", one column format of exome regions of one specific chromosome, e.g., 10:92895-94177)
- exome sequencing reads ("DATA.RD.txt", matrix of all coverage means for all EXOME.interval_list regions of all samples, obtained from GATK pipeline)
- Markov model parameters ("params.txt", all 9 model parameters)
Please see the following link for original source code: http://atgu.mgh.harvard.edu/xhmm/download.shtml
The XHMM R code is now available as an R library. There are still two options for using the scripts:
Install the xhmmScripts package via one of the following methods (you only need to do this once):
-
Compile it from the main XHMM source code downloaded from the Git repository above: cd statgen-xhmm-* make R Then, run in R: install.packages(list.files(path=".", pattern="xhmmScripts_.+\.tar\.gz"), repos=NULL, type="source")
-
Download and install the package from CRAN using the install.packages() command in R: install.packages("xhmmScripts")
-
Download it from the xhmmScripts page at CRAN. Then, run in R: install.packages(list.files(path=".", pattern="xhmmScripts_.+\.tar\.gz"), repos=NULL, type="source")
To use the R code, preface your R scripts with: library(xhmmScripts)