simonwhelan / modelomatic

Modelomatic - A software for model selection across state-spaces

#Modelomatic

Written by Simon Whelan, with contributions from James Allen, Ben Blackburne and David Talavera

ModelOMatic is a C++ program designed for rapid phylogenetic model selection on protein coding genes. Please see the manual for details of the program and its settings.

ModelOMatic is distributed free of charge for personal and academic use only under a GNU GPL v3. Modelomatic is a piece of academically developed software and should be used at your own risk!

Pay attention to the output and use your own judgements when performing analysis. Modelomatic uses lots of heuristics that seem to perform well when tested across the PANDIT database, but they may not perform well on your data.

##Download information

• Precompiled binaries are available here
• The user manual is available here
• Example files are available here

##Version history Modelomatic will be periodically updated and changes that may affect the outcome of your analyses will be announced here. Minor fixes that address bugs that are not expected to change results are not included here.

• v1.03 -> v1.04 (5Aug16): There was a critical error where likelihoods from large families were not computed correctly (usually for codon models). The symptoms of this bug are lots of weird short branches and unreasonably high likelihoods.

##Models tested in Modelomatic

• RY
• DNA
  • JC
  • FEL
  • K2P
  • HKY
  • REV
• AA
  • EQU
  • BLOSUM62
  • DAY
  • JTT
  • WAG
  • LG
  • VT
  • cpREV
  • mtArt
  • mtMam
  • mtREV
  • rtREV
  • HIVb
  • HIVw
• Codon
  • F0
  • F1X4
  • F3X4
  • F64

##How to specify models with a configuration file

Modelomatic will look for configuration files $HOME/.modelomatic.ini and $PWD/modelomatic.ini. Configurations in the second file will be laid over the first (i.e. models turned off in $HOME/.modelomatic.ini can be switched back on in $PWD/modelomatic.ini). The format is standard INI, and models are specified with 0/1, as below:

[Amino Acid]
;Ignore Viral models
rtREV=0
HIVb=0
HIVw=0

[Codon]
F0=0

Currently only Amino Acid and Codon models can be specified. The default case is to run everything. The INI file is case insensitive.

modelomatic's Issues

Broken Prob() error

What steps will reproduce the problem?
#running a large data set with either "fast" or "normal"
>modelomatic_OSX  Phyllos_OR137_Funct_tAlign.phylip bionj 
Phyllos_OR137_Funct_tAlign_mOm_results_fast.txt 0 fast

#or "normal"
>modelomatic_OSX  Phyllos_OR137_Funct_tAlign.phylip bionj 
Phyllos_OR137_Funct_tAlign_mOm_results_fast.txt 0 normal
What is the expected output? What do you see instead?
#Expected (using smaller data set):

--------------------------------------------------------------------------------
-------------
  ModelOMatic (v1.01 (release)).
    A program for choosing substitutions models for phylogenetic inference.
    Written by Simon Whelan.
    Contributions from James Allen, Ben Blackburne and David Talavera.
--------------------------------------------------------------------------------
-------------
Data: 
</Volumes/Yango/Hayden_batORs/OR_alignments/Phyllos_OR4_Funct_tAlign.phylip>: 
127 sequences of length 753 (DataMatrix: 127 x 666)
Checking for sparse (>85% gaps) sequences
    Starting with 127 ... after removal there are 127 sequences
Creating start tree ...  estimated using bionj (1.32627s)
Optimisation settings: normal 
Scanning for modelomatic.ini file ... done
Working with genetic code: Universal
>>> Doing model analysis <<< 
RY Done  (26.7797s)
NT Done  (124.74s)
AA Done  (903.174s).........
Codon Done  (2785.52s)

Outputting results to 
</Volumes/Yango/Hayden_batORs/model_o_matic/Phyllos_OR4_Funct_tAlign_mOm_result_
normal.txt>
Successful exit

#What I got:
--------------------------------------------------------------------------------
-------------
  ModelOMatic (v1.01 (release)).
    A program for choosing substitutions models for phylogenetic inference.
    Written by Simon Whelan.
    Contributions from James Allen, Ben Blackburne and David Talavera.
--------------------------------------------------------------------------------
-------------
Data: 
</Volumes/Yango/Hayden_batORs/OR_alignments/Phyllos_OR137_Funct_tAlign.phylip>: 
452 sequences of length 804 (DataMatrix: 452 x 766)
Checking for sparse (>85% gaps) sequences
    Starting with 452 ... after removal there are 452 sequences
Creating start tree ...  estimated using bionj (5.02254s)
Optimisation settings: fast 
Scanning for modelomatic.ini file ... done
Working with genetic code: Universal
>>> Doing model analysis <<< 
RY Done  (7.94116s)
NT Done  (44.6001s)

Broken Prob(): -40.8111

What version of the product are you using? On what operating system?
I am using the macOSX binary. I am running Yosemite 10.10.3

Please provide any additional information below.
I attached the input both nexus file. The OR4 is working, while the OR137 is 
not working.
What does the Broken Prob(): error mean and how can I address it?

Thanks for your help!

Original issue reported on code.google.com by [email protected] on 3 May 2015 at 5:41

Attachments:

• Phyllos_OR137_Funct_tAlign.phylip
• Phyllos_OR4_Funct_tAlign.phylip

Bounding failed in DoBraOpt(...)...

After running the following command,
modelomatic_OSX OR6_transcriptomes_transAlign_root_noStops.phy bionj OR6_transcriptomes_transAlign_root_noStops_results.txt 0 normal
I get the following output:

ModelOMatic (v1.01 (release)).
A program for choosing substitutions models for phylogenetic inference.
Written by Simon Whelan.
Contributions from James Allen, Ben Blackburne and David Talavera.
Data: <OR6_transcriptomes_transAlign_root_noStops.phy>: 458 sequences of length 1503 (DataMatrix: 458 x 1144)
Checking for sparse (>85% gaps) sequences
Starting with 458 ... after removal there are 458 sequences
Creating start tree ... estimated using bionj (8.06263s)
Optimisation settings: normal
Scanning for modelomatic.ini file ... done
Working with genetic code: Universal
Doing model analysis <<<
RY Done (296.442s)
Bounding failed in DoBraOpt(...)...
after Right: (0.99995: -90846.9765,1: -90846.9761,0.9999999735: -90846.9761)
Returning with no improvement and please report to Simon Whelan

I have run this program many times with other very similar alignments and this is the first time I have come across this error. Any thoughts? I am happy to share the alignment if you would like. Thanks.