James McNellie
James McNellie

Reputation: 1

ASReml-R GBLUP cannot find predict labels

* tmf <- asreml(fixed = oil ~ 1, random = ~ vm(line_code, grm_train), maxiter = 100, data = temp_train)

pred <- predict.asreml(tmf, classify = 'line_code') [ASReml Version 4.2 25/07/2024 16:01:13 Model fitted using the gamma parameterization. LogLik Sigma2 DF wall 1 -1247.260 4.257250 959 16:01:13 2 -1247.260 4.257249 959 16:01:13 3 -1247.260 4.257245 959 16:01:13 Error in asreml(fixed = oil ~ 1, random = ~vm(line_code, grm_train), data = temp_train, : Cannot find predict labels for line_code *

Online guides use this approach with no problems. If I have 'line_code' as a random effect but not within vm() everything works, but once I use a genomic relationship matrix, I get the above error.

Thank you.

Tried to use the predict function within asreml to predict the value of random effects using a genomic relationship matrix.

Upvotes: 0

Views: 32

Answers (1)

Max_H
Max_H

Reputation: 1

predict(), for whatever reason, gives an error if the name of the variable for vm() has an underscore in it.

Example: line_code gives an error, but linecode should work. At least that is how I resolved my issue.

Upvotes: 0

Related Questions