Prediction model to estimate CMV serostatus from DNA methylation levels measured by the Illumna EPIC array. The model is estimated by elastic net regression.
We are trying to use your CMVest tool to predict CMV infection in our samples (we also have the measure based on antibodies against the virus), and we have some questions:
1) the weights in this repository are for M or beta values of methylation?
2) which is the cutoff you would recomend to identify infected samples?
3) have you trained the same algorithm but only using 450K probes (this is the array we have)?
JacobBergstedt
commented
1 year ago
Dear @JacobBergstedt ,
We are trying to use your CMVest tool to predict CMV infection in our samples (we also have the measure based on antibodies against the virus), and we have some questions: 1) the weights in this repository are for M or beta values of methylation? 2) which is the cutoff you would recomend to identify infected samples? 3) have you trained the same algorithm but only using 450K probes (this is the array we have)?
Regards,
Dolors