HyPhy message board - Evolutionary Networks

	Welcome, Guest. Please Login

Home

Help

HyPhy message board › Methodology Questions › How to › Evolutionary Networks

(Moderators: Sergei, Simon)

‹ Previous Topic | Next Topic ›

Pages: 1

Send Topic

Evolutionary Networks (Read 2664 times)

skryazhi

YaBB Newbies

Offline

Feed your monkey!

Posts: 2

Evolutionary Networks
Oct 12^th, 2010 at 9:10pm

Hi Sergei,

I am interested in using HyPhy to detect the evolutionary networks in viral proteins (like in your PLoS Comp Biol paper in 2007). So, how do I go about doing that? I am just starting to learn how to use HyPhy. I have read your documents on HyPhy but didn't see anything on co-evolution or BGMs there. I do see the BGM.bf script, but it contains many functions, and it's not clear to me how to use them. In particular, my really basic questions are:

1. Which functions do I need to use to follow your analysis in that paper?
2. Which inputs do these functions take?
3. What are the outputs of these functions?

Thanks a lot!

Best,
Sergey

IP Logged

Sergei

YaBB Administrator

Offline

Datamonkeys are forever...

Posts: 1658
UCSD
Gender: male

Re: Evolutionary Networks
Reply #1 - Oct 13^th, 2010 at 9:38am

Hi Sergey,

You are correct in noticing that, regrettably, there is little documentation on using BGM in HyPhy.
For some general background, data preparation, etc, take a look at Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

Login

When our cluster is back online (should be today or tomorrow), you can run a BGM analysis online, otherwise, you may need to run using the standard analysis for selection detection. Take a look at sections 2.6.x of Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

Login to learn how to run the preliminary step (model fitting and ancestral sequence reconstruction) in HyPhy. At some point while you are running QuickSelectionDetection.bf, you will be presented with an option "BGM co-evolution" - select that and follow the prompts from there onwards to perform a BGM analysis.

HTH,
Sergei

Associate Professor
Division of Infectious Diseases
Division of Biomedical Informatics
School of Medicine
University of California San Diego

WWW

IP Logged

skryazhi

YaBB Newbies

Offline

Feed your monkey!

Posts: 2

Re: Evolutionary Networks
Reply #2 - Oct 15^th, 2010 at 7:51am

Sergei,

thanks for your prompt response.

Unfortunately, I can't run my analysis on SPIDERMONKEY because the data set that I would like to analyze is too big (more than 400 sequences). So, I am trying to run it locally.

Following your instructions, I have selected the 'QuickSelectionAnalysis' from the 'Standard Analyses' menu, and loaded the p51.nex file for testing. However, when I select 'BGM co-evolution' option, I get the following error:

An error occured: Expected: BGM ident = (<nodes>) Call stack 1 : ExecuteAFile from "BGM.bf" using basepath Macintosh HD:Applications:HYPHY 2.0:TemplateBatchFiles:. Current operation/job has terminated

When I hit 'Skip further' and select various options for the MCMC run (e.g., the number of parents for each node, not include ancestral resampling, selecting the number of burn-in steps, etc.), I get another error:

An error occured: gen_bgm is not a valid likelihood function/data set filter/tree topology in call SetParameter Call stack 2 : Set parameter BGM_DATA_MATRIX of gen_bgm to bgm_data_matrix 1 : ml_bgm_results = runBGM(_bgm_data) Current operation/job has been terminated.

So, what am I doing wrong?

IP Logged

Sergei

YaBB Administrator

Offline

Datamonkeys are forever...

Posts: 1658
UCSD
Gender: male

Re: Evolutionary Networks
Reply #3 - Oct 15^th, 2010 at 11:59am

Hi Sergey,

You are doing everything correctly. I realized, based on your bug report, that the Mac binary was compiled with incorrect define flags (some development code for BGM was included but the older analysis files were not updated to work with it properly). Please download the version of HyPhy I just compiled and placed on our web site and try again.

Thanks for the bug report and sorry for the inconvenience.

Sergei

Associate Professor
Division of Infectious Diseases
Division of Biomedical Informatics
School of Medicine
University of California San Diego

WWW

IP Logged

Pages: 1

Send Topic

‹ Previous Topic | Next Topic ›

« Home

‹ Board

Top of this page