In-silico fragment analysis

Harry Mangalam mangalam at home.com
Fri Sep 17 17:02:10 EST 1999


tacg will do up to several thousand patterns at a time, including all
combinations (if you have the memory).

I've already written a pseudo AFLP analysis routine for it as well:
from the (spare) docs:

-x {Label(=),Label..(,C)} selects SPECIFIC REs (<=15) from the REBASE file;
               If '=' is appended to 1 RE Label, it will 'tag' that RE for
the
               Hookey analysis.  See man page for details.  If ',C' is
               appended to a list of >1 Labels, requests a multiple digest, 
               enabling output of both Sites and Fragments of the combined 
               results.  If '=' is used, only tagged fragments will be
printed 
               (in both order-of-sequence and sorted-by-size).

The bit you might be interested in is the 'Hookey' analysis (named for John
Hookey, who initially requested it).  He was using it for mapping short
bits of a bacterial genome in an AFLP-like way (I believe).

What this allows you to do is 'tag' one RE (say a 6- or 8-cutter) and then
use another 4- cutter to free it, allowing the tagged fragment to be
detected.

so ... -x 'HindIII=,MboI,C'  will show only those fragments that has a
HindIII site on one end (or both ends) and an MboI site on the other end.

I also modified my gel simulation routine so that you can now 'blow up' a
section of the gel to get a better view of the bit you're interested in. 
It's pretty crude (only section by decades (10, 100, 1000, etc bps ) but
it's better than nothing.

Give me shout by email or phone if you're interested.

There's a web interface to the old version at:
http://24.1.175.29/tacg/

(but I believe it's restricted to 500,000 characters of input (as I don't
want people throwing MBs at my poor little PC), but it handles up as much
DNA as you have memory for (it'll do several MB on a Linux PC with 32MB
RAM.  This one does not provide access to the Hookey routine, but it does
do a lot of other useful stuff.

and the new version will be out for beta in a matter of a week or so.  If
you want to try a working, but pretty grungy version of the new one, let me
know.

Best Wishes
Harry


Scott Handley wrote:
> 
> Does anyone have any suggestions for software packages that can find all
> possible restriction sites in a genome (~2 Mb) for two different
> restriction enzymes?  I want to be able to build a list of fragment
> sizes that would be generated by a double digestion and then be able to
> pick those fragments.  This is analagous to a software AFLP protocol.  I
> know Lasergene and MacVector have been used to do this, but if there is
> something else I would be interested.

-- 
Cheers,
Harry

Harry J Mangalam -- (949) 856 2847 -- mangalam at home.com




More information about the Bio-soft mailing list