I've been using tophat with some success until recently. I have one data set that seems to me to be the exact same as my other files, in the same format etc (I even changed the name lines to match the older files that I know work). When I run these files I do not get any errors I just get very very little aligning. Bowtie will align less then 1% of the reads so there is basically no islands for tophat and hence no junctions. With the really big files I might get a few thousand junctions but i don't think I should trust it given the situation. I would be happy to give anyone with any ideas more details, I am just not sure what would be useful.
tophat -o/.../outfolder /home/.../genome /home/.../readfile
@SRR012345.1 FC36:2:1:597:470 length=32
GGGATACATTGCCAAGCTCCGCTGCCAGATTA
+SRR012345.1 FC36:2:1:597:470 length=32
IIIIIIIIIIIII?3II3I$C>I><I&&4&C%
32bp reads not paired end
I am on ubuntu 9.10
tophat -o/.../outfolder /home/.../genome /home/.../readfile
@SRR012345.1 FC36:2:1:597:470 length=32
GGGATACATTGCCAAGCTCCGCTGCCAGATTA
+SRR012345.1 FC36:2:1:597:470 length=32
IIIIIIIIIIIII?3II3I$C>I><I&&4&C%
32bp reads not paired end
I am on ubuntu 9.10
Comment