DNA-R1B1C7-L Archives

Archiver > DNA-R1B1C7 > 2007-08 > 1186821018


From:
Subject: [DNA-R1B1C7] Network.exe
Date: Sat, 11 Aug 2007 04:30:18 EDT


I don't know if this went through or not. I got a bunch of bounced message
emails.



In a message dated 8/10/2007 8:47:27 A.M. Central Standard Time,
writes:

John,

Fascinating stuff!!!

I'm not qualified to respond on the methodology you use, I hope David Ewing
or David Wilson sound off?!

Can you provide the URL to the latest version of the Network software - I'd
love to attempt to analyze some data myself?



You can find the Network.exe program at this site:

_http://www.fluxus-technology.com/_ (http://www.fluxus-technology.com/)


I was interested in the Time Calculations section of the program because
Trinity College used that to determine their 1730 years date for the NW Irish in
their DNA study.

Network.exe uses .ych files.

To make these, use the McGee utility:

_http://www.mymcgee.com/tools/yutility.html_
(http://www.mymcgee.com/tools/yutility.html)

Make sure you check the box to generate Fluxus charts.

Generate _Fluxus_ (http://www.mymcgee.com/tools/yutility.html#fluxus)
phylogenetic network .ych data


You also have an option in McGee to calculate a modal for your data set. If
you leave this checked it will throw that into the Fluxus charts. One
alternative might be to uncheck the modal option and enter your own modal.

Copy and paste the Fluxus data into a .txt file. Save as text. Then
rename the file with a .ych extension.

There are a few caveats to using this program.

1. You can't have any missing markers, all have to be 12, 25 or 37 marker
sets.

2. Each DNA string requires a unique ID string - and the ID strings can only
be so long or Network.exe truncates them.

If you get weird errors trying to load a .ych file into Network.exe it's
generally because you have either missing markers in a sample or duplicate ID
strings.

The Trinity College study stated that they first used the Reduced Median
option in processing the .ych file. This results in an .rmf file. They then
applied the Median Joining option to the .rmf file, which results in an .out
file. You can then draw the .out file and see the resulting phylogenic
network. Then you save the phylogenic chart as either .fdi or .dia files.
Either one of these can be taken into the Time Calculations of the program.

You can also assign weights to each marker in your data set. The default
rate is 10; but you can change this for fast moving markers to something like
5. This can often result in cleaner phylogenic charts. I'm not sure if it
affects TMRCA calculations or not.

If you want to see a list of markers with a lot of mutations in a given
data set, you can use the "Statistics" option while in the draw network mode.

I tried loading 37 marker data sets into Network.exe but kept getting out
of memory errors. So far it's working fine for 100 sets of 25 marker data.

I understand that everyone has their own way of doing things but I thought
if it was good enough for Trinity College than it's good enough for me.



John






************************************** Get a sneak peek of the all-new AOL at
http://discover.aol.com/memed/aolcom30tour

This thread: