So I'm trying to find the closest features to a set of SNPs using something along the lines of:
closest-features --closest --dist <(vcf2bed snps.vcf) hg38.bed > answer.bed
But unfortunately, since I am on a Windows machine running Cygwin, there was an error in the makefile and I get the error:
Error: Cannot find sort-bed binary required for sorting BED output
Digging further, I found that this did not install correctly because after running
make, I get this:
SortDetails.cpp: In function 'FILE* createTmpFile(const char*, char**)': SortDetails.cpp:132:22: error: 'mkstemp' was not declared in this scope fd = mkstemp(tmpl); SortDetails.cpp:138:26: error: 'fdopen' was not declared in this scope fp = fdopen(fd, "wb+");
As far as I can tell, this messes with the install of
sort-bed because when I run
make install I get this:
cp applications/bed/sort-bed/bin/sort-bed- bin/sort-bed make: [Makefile:75: install] Error 1 (ignored) cp: cannot stat 'applications/bed/sort-bed/bin/update-sort-bed-slurm-': No such file or directory
There is, as it says, no "sort-bed" in the bin folder of bedops.
I haven't been able to find anything substantial in solving this problem so my questions are:
- Is there some online portal like Galaxy that can convert vcf to bed? The only step of this which seems to effect me is the conversion step, so if I can get a bed file from my snps.vcf I should be able to proceed.
- Has anyone run into this sort of thing before? Is it impossible to get bedops on Cygwin fully? Any sort of work-around?
- Should I just bypass this whole thing? My goal is to use these SNPs to link them to closest features and then run those features through some sort of pathway analysis software to look for patterns or associations that might be related to my given phenotype... Is this the totally wrong way to go about it?
Thank you so much, this is my first post here and if any of this is against the rules or needs to be changed, please let me know and I'll edit and/or remove it!