Skip to content
A program for divvying or partially filtering multiple sequence alignments
C++ C Makefile
Branch: master
Clone or download
simonwhelan Merge pull request #6 from simonwhelan/gcc-bug-fix
Updated Makefile and version number
Latest commit 70ceca5 Oct 10, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Example Examples added Jun 25, 2019
Cluster.cpp Added partialfiltering as a subset of divvying Nov 23, 2017
Cluster.h Minor bug fixes and updates to some functionality Jun 9, 2018
Divvier.cpp
Divvier.h Updated Makefile and version number Oct 10, 2019
DivvierManual.pdf
LICENSE Added LICENSE Oct 12, 2017
Makefile Updated Makefile and version number Oct 10, 2019
README.md Release version updates Nov 5, 2018
Random.cpp Full divvying update Oct 24, 2017
Random.h Full divvying update Oct 24, 2017
Sequence.cpp Phylip reader error Nov 7, 2017
Sequence.h Adding partial filtering Nov 6, 2017
Tree.cpp Minor bug fixes shared with PREQUAL May 7, 2018
Tree.h Adding partial filtering Nov 6, 2017
bionj.cxx Alpha release Nov 1, 2017
hmm.c Minor bug fixes shared with PREQUAL May 7, 2018
hmm.h Full divvying update Oct 24, 2017
matrices.c Adding partial filtering Nov 6, 2017
matrices.h Initial commit Oct 12, 2017
utils.c
utils.h Adding partial filtering Nov 6, 2017

README.md

Divvier: a program for removing MSA uncertainty by Simon Whelan

Divvier is now submitted for publication. The following options are available in divvier

Clustering options:

	-divvy       : do standard divvying (DEFAULT)
	-partial     : do partial filtering by testing removal of individual characters
	-thresh X    : set the threshold for divvying to X (DEFAULT = 0.801)

Approximation options:

	-approx X    : minimum number of characters tested in a split during divvying (DEFAULT X = 10)
	-checksplits : go through sequence and ensure there's a pair for every split. Can be slow
	-HMMapprox   : Do the pairHMM bounding approximation (DEFAULT)
	-HMMexact    : Do the full pairHMM and ignore bounding

Output options:

	-mincol X    : Minimum number of characters in a column to output when divvying/filtering (DEFAULT X = 2)
	-divvygap    : Output a gap instead of the static * character so divvied MSAs can be used in phylogeny

Recommended usage

There are two main modes for running divvier on an alignment file: full divvying and partial filtering. Full divvying is the default option and will aim to retain as much information in the MSA as possible. For immediate phylogenetic use we suggest the following that will require at least 4 character per column and produce output to myfile.divvy.fas that will work as input into standard phylogeny programs:

./divvier -mincol 4 -divvygap myfile.fas

Partial filtering can be done by adding a single option to this command line:

./divvier –partial -mincol 4 -divvygap myfile.fas
You can’t perform that action at this time.