Here are some more questions about mpiBLAST. We have a 240 node cluster where we are running different bio-things on... But when I tried using mpi blast with a database of about ~120 megs and 32 nodes it wount split the database into 6 fragments and some of the .00 files where missing. I tried upgrading the blast tools. But when I did that i created a file called '(null).nal' ... Basicly I'm getting a bit tiered of Blast as a software package... Can't someone convince NCBI to use autotools insted of csh? -- Mvh|Regards, Lars Graduate student Department of Computer Science, University of Copenhagen http://www.diku.dk When's the last time you used duct tape on a duct? -- Larry Wall