Closed hollybik closed 12 years ago
Liase with Eric to subsample the DiRisi dataset
I'm going to close this issue because I think its becoming a bit too vague, and I have done some comparisons about our PhyloSift results vs. the DiRisi datasets (we're recovering some of the same taxa, so its looking good.) Perhaps we need to define a more specific test case for this if needed for the manuscript.
Viruses are going to be a mess, but we should start digging into some test datasets (e.g. Joe DeRisi's sequence data, or genomes from easier virus groups like Mimiviruses) to see what the PhyloSift output looks like. Then we can iteratively assess what is happening and how accurate are placements are looking.