Continued execution of userGWAS() even with bad SNPs

[Thanh Le]

Hi all,

My model without SNPs for the seven traits demonstrates a good fit, and userGWAS() was run successfully for most SNPs, with the exception of a subset of SNPs. Upon examining these problematic SNPs, I did not identify any anomalies in the beta estimates, standard errors, or MAF.

Additionally, the Z-heritability for all traits exceeds 4, and there are no rows where beta or standard error (SE) are exactly zero.

Could you kindly review my model? Furthermore, would it be possible in the future to consider an option in the userGWAS() function that allows the analysis to proceed despite singularity issues, while reporting only the affected SNPs?

Best regards,

Thanh Le

[Thanh Le]

I forgot to mention the error I encountered. It is ""Error message: task 1 failed - \"Lapack routine dgesv: system is exactly singular: U[11,11] = 0"