|
From: Eccles, D. <dav...@mp...> - 2011-07-14 06:49:38
|
From: Sébastien Boisvert [mailto:seb...@ul...] > VirtualNextGenSequencer dumps pairs of reads so you should provide both. > Can you rerun with -p tests/phix/phix_500k_1.fasta tests/phix/phix_500k_2.fasta > Otherwise I think your graph won't be connected enough because of the random number generator used to simulate reads. Okay, done. Coverage: http://pastebin.com/p6v7m1jA And output: http://pastebin.com/rEjqveCz Still failing to assemble. I must say that these results surprised me, because the minimum coverage from Coverage Distribution has bumped up to 96 now. > Also, what is the read length ? Reads were generated using this command: ../readSimulator/VirtualNextGenSequencer tests/phix/phix_genome.fasta 0 200 10 500000 50 phix_500k_1.fasta phix_500k_2.fasta In other words, no sequence errors, 200bp outer distance (SD = 10bp), 50bp read length. I chose those outer distance / read length values (but SD was a guess) because they are the same as the parameters used on a human genome sequence that was done at MPI a month or so ago (which included a phiX lane). I guess I could choose the reads from that as input, but then I'd be testing a different thing, because those reads would likely have errors. > I added a system test for phiX and it works just fine on my branch master. > see system-tests/tests/phix You're simulating errors / SNPs (substitution rate = 0.005), which is a different test from the one I'm doing. -- David |
