Daniel Panyard

Results 10 comments of Daniel Panyard

Hi, I'm not sure you're referring to by "this sample", but if you mean the string "u997", then that string is not an identifier for a sample in this data...

I think I've noticed a pattern with these errors. The "unexpected column" will have a number of one more than the number of samples with valid phenotypes in the analysis....

I discovered today that if I run rvtests but without including the xHemi.kinship file, the error doesn't occur (but of course then it skips the X chromosome for the analysis)....

A couple more things I've tried: I tried using a tab-delimited phenotype file instead of space-delimited, but that did not prevent the error. I tried generating and using a kinship...

The error seems to occur during KinshipHolder.cpp, line 358 Looks like the "u997" column must be coming from the eigen file (this->eigenFileName). I just (re)discovered that there's a hidden file...

Now, if I'm understanding what's happening correctly, I think this issue might be benign if, when rvtests detects any sort of unexpected problem with the loading of .xHemi.eigen, it just...

So, some potential changes to help mitigate this issue: - If reasonable, don't save and use the .xHemi.eigen file at all--just calculate the kinship eigen-decomposition for the X chromosome each...

For myself (and others who might run into this), it seems like a potential workaround could be to include a line of code in a shell script that would manually...

> > > This is great investigative work @dpanyard ! Thanks! Glad someone else read my detective-novel-of-a-post here. Out of curiosity, have you encountered this warning message as well? I...