Skip to content
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

decomposition failed error #5

Closed
prnakka opened this issue Nov 23, 2020 · 5 comments
Closed

decomposition failed error #5

prnakka opened this issue Nov 23, 2020 · 5 comments

Comments

@prnakka
Copy link

prnakka commented Nov 23, 2020

Hi, Thanks for making STAAR so easy to install and use! I am trying to run Staar on a test dataset of 5000 individuals for one chromosome and I get the following error when I try to run the STAAR() function:

error: eig_sym(): decomposition failed
Error in STAAR_O_SMMAT(G, P, residuals.phenotype, weights_B = w_B, weights_S = w_S, :
eig_sym(): decomposition failed

Any thoughts on how to resolve this error would be greatly appreciated, thanks!!

@xihaoli
Copy link
Owner

xihaoli commented Nov 23, 2020

Hi Priya,
Thanks for your question. In STAAR(), the eig_sym() is a c++ function embedded in the Armadillo library that performs eigen-decomposition of the underlying dense symmetric matrix (in this case, the weighted covariance matrix). May I ask what would be the dimension for your eigen-decomposition, i.e. what is the dimension of the G matrix (you may follow lines 76-102 in the STAAR() function to obtain this G matrix from the input genotype matrix)?
Best,
Xihao

@prnakka
Copy link
Author

prnakka commented Dec 4, 2020

Hi, The dimension of the G matrix I was running this function on was 4620 x 285. Thanks, Priya

@xihaoli
Copy link
Owner

xihaoli commented Dec 5, 2020

Hi Priya,
Thanks for your reply. The dimension of your G matrix is manageable for STAAR() function. In this case, could you please check whether your G matrix or annotation weights, i.e. w_B, w_S, w_A, contain any missing values? If not, could you please try updating your Rcpp and RcppArmadillo packages to the latest version and try it again?
Best,
Xihao

@prnakka
Copy link
Author

prnakka commented Dec 7, 2020

Hello, Removing variants with missing values in the annotations fixed the problem for me, thank you so much!! I had one last question: is there a method you would recommend to combine STAAR p-values across multiple studies? Thank you again!

@xihaoli
Copy link
Owner

xihaoli commented Dec 7, 2020

Hi Priya,

Thanks so much for letting me know that the problem has been fixed. For your question on combining STAAR p-values, could you please send us an email to have a further discussion, as it may depend on the specific study design of your project?

Thanks again.

Best,
Xihao

@prnakka prnakka closed this as completed Dec 8, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants