Entering edit mode
giakountis
•
0
@giakountis-14946
Last seen 6.2 years ago
Dear community members,
I am new and relatively inexperienced with R and/or this forum so please be patient with my input. I am trying to download gene and exon annotation for human (hg19) and convert it to GenomicRanges object. I am performing the analysis in Windows10,64bit / Rstudio using a Bioconductor package called MetaseqR. It was working fine till I updated to R3.4.3. In brief I have installed the following :
R version 3.4.3, 64bit, Bioconductor version 3.6 [BiocInstaller 1.28.0], GenomicRanges 1.30.1 [installed from source, Rtools installed in the system-also tried binary 1.30.0 with no effect)
It seems to download the annotation but immediately I get the following error message :
Downloading gene annotation for hg19... Downloading exon annotation for hg19... Converting annotation to GenomicRanges object...Error in .find_start_end_cols(df_colnames0, start.field0, end.field0) : cannnot determine start/end columns
Tried downgrading to R3.4.2, R3.4.1 etc to no effect. Tried to run through R gui again the same...Please advice, I am desperate!
Thanks James for the reply-it is just more automated to do this when you run the metaseqR package. I can (and have done) it manually in the past by downloading the annotation file and feeding it in. I am more curious to know why it stopped working and whether this is a GenomicRanges issue that has to do with its installation. I noticed that the 1.30.1 version that is available as source installation,requires compilation and which is newer compared to the binary (1.30.0), gave several Rd warning messages such as the following :
Could this be part of the problem?How can i deal with such messages in general?
Many thanks for you time
No. The Rd files are the help pages, and wouldn't have anything to do with code functionality.
I doubt this has anything to do with
GenomicRanges. The error you see comes frommakeGRangesFromDataFrame, which expects there to be a 'start' and 'end' column in thedata.frame. If the parser inmetaseqrdoesn't generate adata.framewith a start and end column, thenmakeGRangesFromDataFramewill fail.I tried to look through the code for
metaseqr, and it's some legit spaghetti code, so I ran out of patience right quick. The maintainer is supposed to be watching for questions pertaining to their package, but you tagged this question (incorrectly) as agenomicrangesquestion instead ofmetaseqR. I don't know if you can add that tag at this late date, so you might consider starting another thread with the correct tag.Thanks James, I know the maintainer in person, I just wanted to figure out if it is a metaseqR issue or if something else causes the problem.
Many thanks for your time and help!
Problem resolved you were right it was a metaseqR bug. Now corrected and runs smoothly, thanks for your time once again!