[Biodevelopers] Blixem help...

[Brian Pino]

Group --

I am currently performing a study to observe the evolutionary fusion, deletion and translocation of telomeric DNA throughout the Human chromosome.  In order to make these observations I have blasted several short (100kbase) telomeric sequences against the human genome build 16.  I have written a program in Perl to graphically display an overview of the results of this data.  Next, to get a more in depth look, I have written my program to call Blixem to view specific alignments of interest.

Here's where the problems start.  Blixem causes a FATAL (memory allocation) ERROR everytime.  For those of you unfamiliar with Blixem it requires the original sequence file which is in the order of 100 kbytes and the Blixem file (200 kbytes).  With the computer's Im using, this is nothing in terms of file size, however it will NOT run.  The only logical explanation I can accrue is that since the blast is of the entire genome the, blixem is allocating memory for large gaps, and huge chromosomal representations??

Basicly, what my question boils down to is:

Has anyone encountered an error like this before??
Is there a work around?
Is there a fix?
Is there a similar program that can handle larger analysis?

Any question, comments or suggestions::: bpino at wistar.upenn.edu

--
Riethman Lab
The Wistar Institute
3601 Spruce St.
Philadelphia, Pa 19104
G64 (215) 898-3873
--