About

Have you heard that time-reversible codon substitution models are biased when multiple sequence alignments exhibit non-stationarity? Would you like to fit non-stationary codon substitution models to your data? This package provides a suite of command-line tools to fit non-stationary codon substitution models.

It was used to produce the results in our paper on the subject.

There are several undocumented features of this package, such as interfacing with MongoDB databases, processing large datasets in parallel using MPI, and accessing our extensions of PyCogent models programmatically using Python. If you would like information on any of these, please post a ticket with your request.

Example

You have an alignment of Human, Mouse, and Opossum in the file aln.fasta.

You have a tree in the file tree.nwk:

(Human,Mouse,Opossum);

You can fit GNC from the paper using:

codon fit aln.fasta tree.nwk GNC.txt

The result will be in GNC.txt:

Likelihood Function Table
=============================================================================
   edge    parent    length       A>C       A>G       A>T       C>A       C>G
-----------------------------------------------------------------------------
  Human      root    0.0808    5.9707    4.3949    6.5200    8.0418    0.0500
  Mouse      root    0.2119    0.9965    1.4484    0.8000    0.2648    0.0500
Opossum      root    0.3457    0.9909    1.7200    2.3427    4.3204    1.2068
-----------------------------------------------------------------------------

continued:
===============================================================================
   edge       C>T        G>A       G>C       G>T       T>A        T>C     omega
-------------------------------------------------------------------------------
  Human    8.8596    20.0000    0.0500    0.0500    0.0500    10.7019    0.0150
  Mouse    0.4895     0.9670    0.0500    0.0500    0.2828     1.5363    0.0000
Opossum    2.9152     6.9261    0.0500    2.3511    1.0258     2.7257    0.0148
-------------------------------------------------------------------------------

===============
motif    mprobs
---------------
  CTT    0.0180
  ACC    0.0121
  ACA    0.0387
  ACG    0.0000
  ATC    0.0111
  ATA    0.0123
  AGG    0.0128
  CCT    0.0170
  AGC    0.0133
  AGA    0.0171
  ATT    0.0246
  CTG    0.0065
  CTA    0.0109
  ACT    0.0152
  CCG    0.0000
  AGT    0.0468
  CCA    0.0193
  CCC    0.0058
  TAT    0.0188
  GGT    0.0121
  CGA    0.0091
  CGC    0.0000
  CGG    0.0061
  GGG    0.0031
  GGA    0.0115
  GGC    0.0182
  TAC    0.0112
  CGT    0.0059
  GTA    0.0087
  GTC    0.0063
  GTG    0.0151
  GAG    0.0324
  GTT    0.0090
  GAC    0.0109
  ATG    0.0240
  AAG    0.0269
  AAA    0.0452
  AAC    0.0335
  CTC    0.0090
  CAT    0.0098
  AAT    0.0295
  CAC    0.0202
  CAA    0.0094
  CAG    0.0386
  TGT    0.0208
  TCT    0.0128
  GAT    0.0402
  TTT    0.0090
  TGC    0.0032
  TGG    0.0060
  TTC    0.0060
  TCG    0.0000
  TTA    0.0352
  TTG    0.0165
  TCC    0.0086
  GAA    0.0487
  TCA    0.0147
  GCA    0.0412
  GCC    0.0160
  GCG    0.0000
  GCT    0.0149
---------------

Installation

pip install numpy # for PyCogent
pip install codon

Documentation

On Read the Docs.

Support

Issue tracker: https://bitbucket.org/nonstationary/codon/issues

Contribute

Source Code: https://bitbucket.org/nonstationary/codon

License

GPLv3 or any later version.