> My guess is that the 23651 symbol exact match is > somehow causing > problems. Your guess sounds plausible. I wouldn't be the least bit surprised to discover that all of my failing sequences were highly conserved regions. > If you have other sequences you need help with > to keep your project > going, feel free to contact me and I'll do what > I can to help. My focus is on building a tool. The specific sequences are only of interest as test cases. So I don't think that will be necessary. But your insight has already been valuable. Thank you While I'm still interested in hearing whether other mpiblast users can successfully run that sequence, perhaps the mpiblast developers could theorize whether large exact matches might be problematic? ===== Mike Cariaso