On Dienstag 02 Juni 2009 Jan Paces wrote: > [...] > I think Mira keeps in memory few bytes about each read, which makes it > impossible to use it with huge amount of SOLiD or SOLEXA reads. > [...] Yep, this is currently a real problem. Unfortunately, not only a few bytes. On 64 bit machines, I have ~280 bytes of overhead per read just for alll the empty pointers, strings, vectors, clipping points and lists. Then around 10 bytes per read base (nucleotide, quality, adjustement positions etc.). An then MIRA uses an assembly strategy the keeps copies of reads untouched until the contig they were put in has been accepted as good. This was perfectly reasonable for Sanger project up to medium eukaryote size (say, 80 megabases with 1m Sanger reads) and is still manageable now for 454 Titanium, but the small reads break my neck there. I sometimes wish I could squeeze in more work in the evening hours and weekends to get around this ;-) Regards, Bastien -- You have received this mail because you are subscribed to the mira_talk mailing list. For information on how to subscribe or unsubscribe, please visit http://www.chevreux.org/mira_mailinglists.html