Bayesian Piecewise Constant Regression for DNA copy number estimation (mBPCR)
The mBPCR method is a Bayesian regression method for data which are noisy observations of a piecwise constant function.
The original Bayesian Piecewise Constant Regression method (BPCR) was presented by Hutter (2007a) and (2007b). The mBPCR algorithm is
a modification of BPCR, which significantly improves the estimation of the number of segments and the boundaries
of the piecewise constant function (Rancoita et al. (2009)).
Since the copy number of a genomic region can be considered as a piecewise constant function, mBPCR can be used to estimate the copy number profiles.
Publications & Presentations
- Hutter, M. (2007a). Bayesian Regression of Piecewise Constant Functions. In Bayesian Statistics: Proceedings of the Eighth
Valencia International Meeting. Universitat de Valencia and International Society for Bayesian Analysis, Spain.
- Hutter, M. (2007b). Exact Bayesian regression of piecewise
constant functions. Bayesian Analysis 2(4), 635-664.
- Rancoita P.M.V., ďAn Improved Bayesian Method for DNA Copy Number Estimation", Mathematical and statistical aspects of molecular biology (MASAMB08), Glasgow, March 27-28 2008.
- Rancoita, P.M.V., Hutter, M., Bertoni, F., Kwee, I. (2009). Bayesian DNA copy number analysis. BMC Bioinformatics
- To download the zipped folder with the source of the program and some examples, please click here
- To download the instructions and some suggestions for using the program, please click here
- The R package can be downloaded from the Bioconductor website [link]