Skip to content

wiedenhoeft/HaMMLET

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HaMMLET

HaMMLET – Fast Bayesian HMM segmentation for big data

This software implements Forward-Backward Gibbs sampling for Bayesian segmentation in Hidden Markov Models (HMM). It uses dynamic wavelet compression to drastically improve convergence and memory consumption, making inference possible on large-scale data.

For instance, HaMMLET can be used on a regular laptop for segmentation of genomic data, such as array-CGH or depth-of coverage from whole-genome sequencing (WGS), to find candidates for copy-number variants (CNV). For details, please refer to the doc/ directory.

For implementation details and the theory behind this approach, please refer to my thesis (DOI: 10.7282/t3-4e1k-ph18).