HATCHet v1.0.1

HATCHet 1.0.1

The 1.0.1 release of HATCHet has a number of algorithmic improvements over the 0.x versions:

A variable length binning approach to ensure that each bin contains some minimum number of SNP-covering reads.
Reference-based phasing and quasi- maximum likelihood EM approach to capture allelic imbalance.
Locality-based clustering using a Hidden Markov Model.

These improvements are all turned on by default. To emulate pre-1.0 behavior, you will want to:

Use HATCHet's count-reads-fw command instead of count-reads (fw stands for fixed width).
Use HATCHet's combine-counts-fw command instead of combine-counts.
Use HATCHet's cluster-bins-gmm command instead of cluster-bins.