Greetings,
Currently, I am doing a project that involves creating a database to store cuffdiff output. I have recently been running some data through the workflow Tophat=>cufflinks=>cuffcompare=>cuffdiff to get a better understanding of the data. For the cuffdiff output, I noticed that all of my fpkm_tracking files have a "-" in column 9, the coverage column. To see how best to design my database.
To see under what circumstances the coverage column might be filled in with numbers instead of "-", I download the cufflinks source code for both version 1.3.0 and 2.0.2. I looked in the cuffdiff.cpp file and in both cases was surprised to find this:
fprintf(fout, "%s\t%c\t%s\t%s\t%s\t%s\t%s\t%s\t%s",
description.c_str(),
track.classcode ? track.classcode : '-',
track.ref_match.c_str(),
all_gene_ids.c_str(),
all_gene_names.c_str(),
all_tss_ids.c_str(),
track.locus_tag.c_str(),
length_buff,
"-");
The coverage column appears to be hardcoded to always have the value "-", although perhaps there's another part of the source code where the coverage column gets filled in with something else that I don't know about. After all, the cuffdiff documentation makes no mention about the coverage column only being able to have a "-" in it.
To summarize, has anyone found there to be a value besides "-" in the coverage column of their fpkm_tracking files from the output of cuffdiff?
I appreciate any help you can offer.
Currently, I am doing a project that involves creating a database to store cuffdiff output. I have recently been running some data through the workflow Tophat=>cufflinks=>cuffcompare=>cuffdiff to get a better understanding of the data. For the cuffdiff output, I noticed that all of my fpkm_tracking files have a "-" in column 9, the coverage column. To see how best to design my database.
To see under what circumstances the coverage column might be filled in with numbers instead of "-", I download the cufflinks source code for both version 1.3.0 and 2.0.2. I looked in the cuffdiff.cpp file and in both cases was surprised to find this:
fprintf(fout, "%s\t%c\t%s\t%s\t%s\t%s\t%s\t%s\t%s",
description.c_str(),
track.classcode ? track.classcode : '-',
track.ref_match.c_str(),
all_gene_ids.c_str(),
all_gene_names.c_str(),
all_tss_ids.c_str(),
track.locus_tag.c_str(),
length_buff,
"-");
The coverage column appears to be hardcoded to always have the value "-", although perhaps there's another part of the source code where the coverage column gets filled in with something else that I don't know about. After all, the cuffdiff documentation makes no mention about the coverage column only being able to have a "-" in it.
To summarize, has anyone found there to be a value besides "-" in the coverage column of their fpkm_tracking files from the output of cuffdiff?
I appreciate any help you can offer.
