plantimals / 2vcf

convert 23andme or Ancestry.com raw genotype calls into VCF format, with dbSNP annotations
MIT License
51 stars 4 forks source link

make not working on Mac OS X with current go installed #7

Closed njbowen closed 2 years ago

njbowen commented 5 years ago

make error in Mac OS X 10.14.3 command and error pasted below

G2A:2vcf-develop nbowen$ make go build -o bin/2vcf ./cmd/2vcf/ cmd/2vcf/2vcf.go:11:2: cannot find package "github.com/briandowns/spinner" in any of: /usr/local/go/src/github.com/briandowns/spinner (from $GOROOT) /Users/nbowen/go/src/github.com/briandowns/spinner (from $GOPATH) cmd/2vcf/2vcf.go:12:2: cannot find package "github.com/fatih/color" in any of: /usr/local/go/src/github.com/fatih/color (from $GOROOT) /Users/nbowen/go/src/github.com/fatih/color (from $GOPATH) cmd/2vcf/2vcf.go:13:2: cannot find package "github.com/plantimals/2vcf/convert" in any of: /usr/local/go/src/github.com/plantimals/2vcf/convert (from $GOROOT) /Users/nbowen/go/src/github.com/plantimals/2vcf/convert (from $GOPATH) cmd/2vcf/2vcf.go:14:2: cannot find package "github.com/plantimals/2vcf/genomicsuploader" in any of: /usr/local/go/src/github.com/plantimals/2vcf/genomicsuploader (from $GOROOT) /Users/nbowen/go/src/github.com/plantimals/2vcf/genomicsuploader (from $GOPATH) cmd/2vcf/2vcf.go:16:2: cannot find package "gopkg.in/alecthomas/kingpin.v2" in any of: /usr/local/go/src/gopkg.in/alecthomas/kingpin.v2 (from $GOROOT) /Users/nbowen/go/src/gopkg.in/alecthomas/kingpin.v2 (from $GOPATH) make: *** [2vcf] Error 1 G2A:2vcf-develop nbowen$

njbowen commented 5 years ago

used go get command to pull all the packages listed in error, now error is go build -o bin/2vcf ./cmd/2vcf/

github.com/plantimals/2vcf/genomicsuploader

../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:95:26: c.GenomicsService.Variants undefined (type genomics.Service has no field or method Variants) ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:95:44: undefined: genomics.ImportVariantsRequest ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:113:41: undefined: genomics.Dataset ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:115:27: c.GenomicsService.Datasets undefined (type genomics.Service has no field or method Datasets) ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:130:25: c.GenomicsService.Datasets undefined (type genomics.Service has no field or method Datasets) ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:130:43: undefined: genomics.Dataset ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:141:44: undefined: genomics.Dataset ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:141:64: undefined: genomics.VariantSet ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:143:27: c.GenomicsService.Variantsets undefined (type genomics.Service has no field or method Variantsets) ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:143:48: undefined: genomics.SearchVariantSetsRequest ../go/src/github.com/plantimals/2vcf/genomicsuploader/genomicsuploader.go:143:48: too many errors make: *** [2vcf] Error 2

plantimals commented 5 years ago

Hi @njbowen,

I see you closed the issue here, were you able to get the project built? If you're still having trouble, please let me know, I'd be happy to help.

Rob

plantimals commented 5 years ago

If your troubles are coming from the Google Genomics package, I'm working on a newer version that removes the deprecated package.

njbowen commented 5 years ago

Not sure, think it was one of the .go calls. I’ll wait for your update and try again later, thanks.

plantimals commented 5 years ago

@njbowen I've got a branch that builds with the Google Genomics integrations removed. I'll push it tonight and let you know. Thank you for your patience.

plantimals commented 5 years ago

@njbowen ok, you should be able to pull the latest commits and build. I'm also working on including some pre-built binaries so people who just want to use the tool will be able to.

njbowen commented 5 years ago

got it to make, command is in 2vcf-develop/bin/2vcf not sure of basic command, but issued this from the 2vcf-develop folder

nbowen$ bin/./2vcf conv ancestry X.zip -o X.vcf.zip

vcf output at: X.vcf.zip

however, don't think it ran, can't find output. in --help found a few pointers, but still need assistance. thanks,

njbowen commented 5 years ago

even though I installed go using brew. I still had errors with the make in the 2vcf-develop/ . It presented errors that it was looking for go directories or packages and couldn't find them. I installed each as the errors were returned using go get GitHub.com/xyz until finally no errors were returned. is this how it should make? or is there something wrong with my go installation maybe? thanks,

plantimals commented 5 years ago

@njbowen yeah, when you build a new project you need to go get any packages you don't already have. I just added a dependency manager called dep. If you check out the latest commits, brew install dep, then run dep ensure, it should download all the required packages.

in any case, you managed to build the executable, so let's see if we can get it working for you. I think you might be missing the reference data path. if you run from the root of the repo that you checked out, it should work automatically, otherwise you can specify a path to the reference. try:

./bin/2vcf conv ancestry path/to/my-genotypes.zip

You may have noticed that the reference/ directory holds a VCF containing the reference data, and the binary needs to be able to load that. if you don't run it from the root directory, you can use the --vcf-ref parameter to specify a path to that reference file.

plantimals commented 5 years ago

@njbowen if you have any future problems with building the binary, now you can download a pre-built binary https://github.com/plantimals/2vcf/releases/tag/v0.3.3

njbowen commented 5 years ago

@plantimals I was in the root directory, but tomorrow I’ll try giving it the reference path and I’ll install a pre built binary and give that a try. Thanks,

njbowen commented 5 years ago

@plantimals I replaced the 2vcf with the Darwin build. I run the command as in the picture below and get a line that the output is..... but there is no output file in the folder. Thanks for helping with this. 2vcf_Darwin_run_image

njbowen commented 5 years ago

@plantimals I've also run the same and pointed it to the reference file as you suggested, got same blue output line but no output file.

plantimals commented 5 years ago

@njbowen well, I feel pretty dumb, the ancestry path was simply not plugged in. I no longer have an example of ancestry data to test on. I've made the obvious update, but I can't test it locally. If you wouldn't mind pasting a line or two of your file here, replace the genotype calls with nonsense to protect your identity, I would greatly appreciate it. Here's the latest update: https://github.com/plantimals/2vcf/releases/tag/v0.3.5

I really appreciate that you're taking the time to help get this right. I'd be happy to help in a more direct way if you have a time crunch at all here. We could exchange emails, get on a google hangout or something. Let me know if I can be of assistance.

njbowen commented 5 years ago

@plantimals Ancestry_DNA_test.txt

njbowen commented 5 years ago

@plantimals looks like it works now, thanks.

plantimals commented 5 years ago

@njbowen awesome! thank you so much for your help. If you come across any more issues, I would greatly appreciate hearing about them. Especially any features that would make your use easier.

njbowen commented 5 years ago

hi just tried to run on a 23andme v5 raw data file and it only returned ~190K SNPs in the .vcf do I need a different reference vcf to process v5 results from 23andMe? the v3 raw results seem to work fine.

plantimals commented 5 years ago

@njbowen hmmm. would it be possible for you to extract the rs_id's from your results? I'd like to intersect them with column 3 of the reference file. I really doubt there's been such a huge turnover that only ~20% of the markers would have successful observations.