Hello,
I have been working with the "tuxedo" package for a little while now, and have started getting the nice results that I want, but I still have a few nagging issues.
One is that when I use the -b option, for multi-read correction in cuffdiff, I get a segmentation fault at differing stages of the run, and thus no result. If I do not use this option, I basically get no results, and the gene_exp.diff is basically full of NOTEST and LOWDATA entries.
This is only an issue for the newest (and unfortunately much better) annotation and assembly, as when I run the package with the older files, it works generally fine (also with the -b option).
Is there anything in the genome fasta or bowtie mapping files that I should look for in order to fix this problem? If anyone wants to take a look, I would be more than happy to send the files...
Another issue I have is that at some point some clearly distinct genes are merged into one cufflinks transcript, so that in the gene_exp.diff file, an entry in the "gene" column would look like: Nasvi2EG000144,Nasvi2EG000145,Nasvi2EG000147,Nasvi2EG000149,Nasvi2EG000151 It is clear to me that I am losing quite some differential expression information due to this issue, and was wondering if there was a way to avoid it....
I have been working with the "tuxedo" package for a little while now, and have started getting the nice results that I want, but I still have a few nagging issues.
One is that when I use the -b option, for multi-read correction in cuffdiff, I get a segmentation fault at differing stages of the run, and thus no result. If I do not use this option, I basically get no results, and the gene_exp.diff is basically full of NOTEST and LOWDATA entries.
This is only an issue for the newest (and unfortunately much better) annotation and assembly, as when I run the package with the older files, it works generally fine (also with the -b option).
Is there anything in the genome fasta or bowtie mapping files that I should look for in order to fix this problem? If anyone wants to take a look, I would be more than happy to send the files...
Another issue I have is that at some point some clearly distinct genes are merged into one cufflinks transcript, so that in the gene_exp.diff file, an entry in the "gene" column would look like: Nasvi2EG000144,Nasvi2EG000145,Nasvi2EG000147,Nasvi2EG000149,Nasvi2EG000151 It is clear to me that I am losing quite some differential expression information due to this issue, and was wondering if there was a way to avoid it....
Comment