Nowhere in the documentation does the GOTHiC function explicitly state that it only takes bowtie files as input. Please make note of this. I tried to feed in bwa-aligned bams and the tool failed. Thanks for your help and efforts.
Ok I am now getting the same error using bowtie2-aligned bams:
Error in .Call2("solve_user_SEW0", start, end, width, PACKAGE = "IRanges") :
solving row 67984077: range cannot be determined from the supplied arguments (too many NAs)
Calls: GOTHiC ... GRanges -> new_GRanges -> IRanges -> solveUserSEW0 -> .Call2
I see a similar issue in another thread - Errors using wavClusteR package - "range cannot be determined from the supplied arguments (too many NAs)" - the solution to get passed this issue was to run bowtie - does GOTHiC only support bowtie? not bowtie2 or bwa?
Any help would be much appreciated.
Appears it only supports bowtie (I conclude after running an additional test)? Clearly this package is incredibly useful, but how can it be that it is so widely used but there is no documentation or discussion on the support thread regarding obvious things like this?
In fairness to the authors the vignette at https://bioconductor.org/packages/3.8/bioc/vignettes/GOTHiC/inst/doc/package_vignettes.pdf does say:
2.3.1 Pairing aligned paired NGS reads
The pairReads function takes bowtie output files, ...
although I'm not aware of any fundamental differences between bam files produced by bwa vs bowtie - it is a standard after all.
Can you perhaps make available two small example bam files that produce this error so others can investigate?
Login before adding your answer.
Use of this site constitutes acceptance of our User Agreement and Privacy