Starting the analysis of 454 data

[Naira]

Hi all,

I am newbie and I will have to start the analysis of 454 Roche data very soon.
I was wondering if you have any recommendation about the analysis.

The fist step will be to map the reads to the genome (here Arabidopsis thaliana)
For that, I am still not sure of which tool to use.. Eland or MaQ simply BLAST??

I will be happy to have any suggestion from the people who are already into this.

Thanks a lot,
Naira

[cabauser]

For GS-FLX reads you need an alignment program that can handle the 250bp / 400bp reads generated by the GS-FLX. Blast is probably only good enough for a first estimte of where a read might align (possibly several positons) and to give an estimate of how many reads do not align. Even if you do a Smith-Waterman alignment of the regions around every Blast hit (+/-300bp, for example) it will only take a weekend to map the reads from a flx run (depending on the computer). That is the easy part. The hard part is designing and implementing a database that can help with the analysis A database and genome browser to store the reads and positions as well as quality metrics (things like number of mismatches) is probably a good idea. If need be, you can exclude things that are trivial. For example, if you are looking for SNPs, you don't need to keep reads that are 100% identical. Of course the structure will depend on the questions you are asking.
Gook Luck