rwdavies / STITCH

STITCH - Sequencing To Imputation Through Constructing Haplotypes
http://www.nature.com/ng/journal/v48/n8/abs/ng.3594.html
GNU General Public License v3.0
77 stars 17 forks source link

NA in correlation #37

Closed madleina closed 4 years ago

madleina commented 4 years ago

Hey!

We verify the performance of our STITCH test runs by looking at the interim.r2-files produced based on a genfile. However, for some individuals it outputs NA (example attached). I'm not sure what the exact reason is, but I think it is caused by calculating the mean in R without na.rm=T. Would it be possible to fix this?

Many thanks! randomIND_largerdistance_K100.txt