-
Notifications
You must be signed in to change notification settings - Fork 27
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
error running relateness check #3
Comments
could you send a sessionInfo() please? Could you share your dataset with me (confidentially)? |
Thanks for the reply! where can I upload the data? |
Can you share a Dropbox/google drive link?
… On 20 Jun 2019, at 22:46, blueskypie ***@***.***> wrote:
Thanks for the reply! where can I upload the data?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.
|
I sent the link of the data to your work email. Thanks! |
I've received the data. Could you please send your sessionInfo()? |
Matrix products: default locale: attached base packages: other attached packages: loaded via a namespace (and not attached): |
Thank you for pointing me to this issue: in your example data set, there are no related individuals, all IBD estimates are 0. When trying to select non-related individuals plinkQV failed as it was lacking a check for this scenario. Can you try again by installing the github version of the package via: |
I used |
I've updated the version, can you check again. Thanks! |
Thanks, working now! However, I'm a bit surprised that "all IBD estimates are 0". If I use |
rel-cutoff estimates the relationship based on a genetic relationship matrix that build using all genetic variants that you provide and default threshold of 0.025 as suggested in the original GTCA paper. check_relatedness uses IBD estimation which leads (in my opinion) to more interpretable values for the level of relatedness, with eg 1 being identical, 0.5 first degree relatives etc. Either way, the example data you provided has a very poor genotyping rate of 0.308368, a high degree of missingness per individual and therefore many low minor allele frequencies. As minor allele frequency and missingness are considered respectively in estimating relatedness and selecting the individual of a pair of related individuals to keep, the two individuals marked as related with rel-cutoff were not detected with the highIBD filter. I have updated the documentation making this process clearer and added user input of MAF as an option (8ac935b). I am closing this issue now as the initial bug has been resolved. |
Hi,
Thanks for make the package! I'm getting error whiling running on some data. Here is the command
part of the output:
The text was updated successfully, but these errors were encountered: