Thank you for posting this.
I was about to create a Biostars post asking about free alternatives to GENCOVE, but is there an open version that is free to use?
When I looked around, I thought that the only command line option was for uploading files to the interface:
https://docs.gencove.com/cli/
It looks like there is a free trial, but I am not sure how often I might want to check the GENCOVE imputation results. A slower run-time is acceptable to me - I just want to be able to validate what companies say is acceptable low-coverage WGS data as an independent 3rd party (and verify what genetic data can be considered identifying information).
UPDATE (3/6/2020): It took ~2 weeks (I originally submitted a request for a trail on 2/22/2020, I had a phone call with Gencove on 3/2/2020, and I was able to run the analysis on my first successful sample today), but I was able to submit FASTQ files for free, as promised.
I'l provide another update with some additional analysis of the .vcf file that I downloaded, but I also have the following notes:
- You upload FASTQ files with the Command Line Interface (CLI).
- There is a message about checking if a file is uploaded, but I tried to add new files from a folder with previous files and they weren't recognized (so, all of my first set of files were duplicated).
- There are some caveats with the file formatting (depends upon what is found before "_R1" or "_R2" in your .fastq.gz file, if I understand things correctly), but perhaps this can be changed in the future (along with the file checking bug in the CLI)
- For example, I am not sure if you can delete files after they are uploaded, but you can re-name files by creating a new "Client ID" (and you can save the change my immediately assigning them to a project). This was not immediately obvious to me.
- Also, I don't think the "_1" and "_2" format for paired in read was recognized, but you can force files to be forward and reverse reads by the order they appear under the "Client ID".
- I have concerns about the ancestry and PRS provided in the reports, but I was really only interested in the genotypes in the imputed .vcf file.
Original poster has not logged in for 2.5 yrs. You may want to contact the company directly.
Thank you!
I sent them an e-mail (to support@gencove.com) relatively recently (I believe a little less than a month ago) with a slightly different question (even though it contained a portion asking about free use of GENCOVE), and I haven't heard back.
I am thinking about looking into the free trial this weekend. If my trial expires (and/or STITCH looks like it isn't acceptable), then perhaps I will try again with an e-mail. I think this will probably be OK, but I will contribute to the forum if I find problems with the free trial (so, others will also know).
Thank you again for seeing this and contributing to the discussion!
For example, has anybody tried STITCH?
https://github.com/rwdavies/STITCH
After I take a look, I can post some notes of my experiences. However, if anybody has any relevant information, then I am interested in learning about that.
FYI, I have completed some testing with STITCH, which I think looks reasonable good (as an open-source option, given that Gencove will require a license after the free trial, and they check to get some sense of how many samples you will process during that free trial):
You can also see more details from the following links:
https://github.com/cwarden45/DTC_Scripts/tree/master/Nebula/Gencove
https://github.com/cwarden45/DTC_Scripts/tree/master/Color/lcWGS_Genotype_Estimations
https://github.com/rwdavies/STITCH/issues/29