HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
HYPHY Package >> HyPhy feedback >> How to mention the name of the nucleotide evolutionary model
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1352125819

Message started by RTuteja on Nov 5th, 2012 at 6:30am

Title: How to mention the name of the nucleotide evolutionary model
Post by RTuteja on Nov 5th, 2012 at 6:30am
Hi All,

Modelgenerator program has predicted that the best model for my sequences is TVM+I+G. I don't see this model in the menu while using singlebreakpointReco.

Can you please suggest me the way to specify this model. I don't know what is the six letter designation for the models.

Please help!

Thanks,
Reetu

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by Sergei on Nov 5th, 2012 at 1:36pm
Hi there,

TVM == REV with fixed (0.25) base frequencies.
Use (012345); see Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login Login; (page 16)

Sergei

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by RTuteja on Nov 12th, 2012 at 3:07am
Hi Seregi,

Thanks for your help so far on this.

I could not find why I have to specify TVM == REV, instead of only TVM (are these submodels?). I have around 200 gene families for analysis. I am reading up in the area of nucl evolutionary models but any suggestions/hints from you will be a helpful to explore myself further.

Many thanks for your help!

Regards,
Reetu

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by Sergei on Nov 13th, 2012 at 2:11pm
Hi Reetu,

Generally speaking, in most HyPhy analyses it is not terribly important to specify the exact "best" nucleotide model; so long as you don't force a model that's too simplistic (e.g. F81). REV is more general than TVM, and base frequencies in REV are not estimated as a part of the likelihood function (they are estimated by nucleotide counts), so it won't hurt much effect on the inference.

For some good background reading on the subject of model selection Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login Login

Sergei

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by RTuteja on Nov 16th, 2012 at 10:01am
Hi Sergei,

I am getting the following error while running Singlebreakpint.bf module on cluster. Can you please help where I am doing mistake.

Thanks,
Reetu
+++++++++++++++++++++++++++++
2). Looking for a breakpoint...
Error:

Master node received an error:Cannot reuse the filter 'filteredData0' because it is already being used by likelihood function 'lf', and the two likelihood functions impose different leaf-to-sequence mapping. Create a copy the filter and pass it to the second likelihood function to resolve this issue.

Function call stack
1 : Construct the following likelihood function:filteredData0 ; givenTree1 ; filteredData1 ; givenTree0 ;
        Standard input redirect:
                Empty Associative List-------
2 : RunASample(individual)
        Standard input redirect:
                Empty Associative List-------

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by RTuteja on Nov 19th, 2012 at 4:51am
P.S.- If I skip the KH-test then everything is working fine. Is this a bug or I am making some mistake?

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by Sergei on Nov 19th, 2012 at 1:00pm
Hi Reetu,

I think this is a bug in the script. Let me investigate.

Sergei

Title: Re: How to mention the name of the nucleotide evolutionary model
Post by yuda on Oct 4th, 2013 at 2:33am
Dear Sergei,

I was wondering whether you could fix the bug in the SBP script.

Thank you,
Haruo

HyPhy message board » Powered by YaBB 2.5.2!
YaBB Forum Software © 2000-2017. All Rights Reserved.