The present invention concerns methods for predicting MED of a subject's skin, comprising either determining the methylation levels of at least 2 CpG sites selected from certain genes in a skin sample obtained from the subject, and predicting the subject's MED based on the determined methylation levels using a machine learning model trained on data comprising known MEDs and corresponding known methylation levels of the at least 2 CpG sites, and/or determining the RNA expression levels of at least 2 genes selected from certain genes in a skin sample obtained from the subject, and predicting the subject's MED based on the determined RNA expression levels using a machine learning model trained on data comprising known MEDs and corresponding known RNA expression levels of the at least 2 genes. The invention further relates to computer programs for carrying out the methods of the invention.
展开▼
