Dear Bioc community,

I'm wondering if I am missing something. I have been using Bioconductor in an attempt to visualize the location of tblastx hits (FASTA seqs) on a novel genome. I am developing the impression that: Bioc is not ideal for my needs, and I would like some community feedback. Am I wrong? Have I missed the important packages? Do I misunderstand some fundamental aspect of Bioc?

Here is why I am developing my impression:

1) No native method to import blast results.

For example: I want to make an IRanges object using the tblastx output. I import the tblastx data using read.table. But tblastx uses all six frames and some of the start positions have higher values than the end positions. I write a function to swap the position of these numbers and then IRanges is happy. Have I completely missed the package that does all things blast?

2) import.gff does not allow "negative widths."

I'd like to make ideograms to represent the tblastx hits, some of which are near genomic features that are described in the GFF3 file associated with our genome. The function import.gff3 does not use the "strand" information and kicks out the error 

Msex.gff <- import.gff3("M_sexta/OGS2_20140407.gff3")

Error in .Call2("solve_user_SEW0", start, end, width, PACKAGE = "IRanges") : 
  solving row 357069: negative widths are not allowed

I am new to Bioc and could easily be missing something. So please, if I am wrong let me know. 

Thanks in advance

Chris Hamm (KU)

Hi Chris,

Don't know about 1), but for 2) it looks to me that the problem is with your GFF3 file, not with import.gff3().

According to the official GFF3 specs here

  http://www.sequenceontology.org/gff3.shtml

"start is always less than or equal to end". Therefore I'm not surprised that import.gff3() choke on a file that doesn't satisfy this.

Maybe you could try to use the online validator (the above specs provide the link to it) to validate your GFF3 file, and, if the file turns out to be invalid like I suspect, contact the persons who generated it.

Cheers,

H.