There are two main areas where MIRA could need help: programming and
resources. If you feel you can help, please contact bach@chevreux.org


Programming & Documentation:
============================
- testing compilation and running of MIRA on presently untested platforms
  (see README for a list of platforms).
- writing beginner tutorials for assembling data (including preprocessing
  steps with other programs like lucy, pregap4 etc) would be very welcome
- some helper scripts to automatically download, quickly reformat and then
  assemble data from the NCBI trace archive or the ENSEMBL trace server would
  be nice (perl or  Tcl preferred)
- a linear-space banded Smith-Waterman alignment routine that works without
  runtime-penalty would come in handy. I already have an idea how to realise
  this but no time. Tricky, but feasible.
- multithreading: the Smith-Waterman alignments could be calculated in
  parallel (not trivial, but should not be really hard either). Furthermore,
  the task of building contigs could be distributed on 2 threads or even 2
  threads per independent contig (harder to realise). I did not find the time
  to build this in yet.
- and of course: feedback in any form, be it ideas, bug reports, success
  stories (or failures) etc.


Resources:
==========
- I only have acces to a machine with 4 GB RAM (well, my box at home), which
  makes testing of 454 data assembly only possible for low-coverage small
  genomes. It would be great if someone could provide me an SSH access to a
  reasonably fast machine with 16 or more GB RAM (and some 25 to 50 GB disk).
- even better: sponsoring a reasonably fast machine with 16 GB RAM would also
  be very welcome as this is definitively something I will not be able to
  afford anytime soon :-) This would allow me to test MIRA for special
  questions where I have data that is not public yet.
