Fig. 1
From: Machine learning identifies a compact gene set for monitoring the circadian clock in human blood

Using ZeitZeiger to predict circadian time in control samples from three datasets (tenfold cross-validation). a Boxplots of absolute error for various values of sumabsv (regularization parameter) and nSPC (number of SPCs). b Boxplots of absolute error for each dataset at sumabsv = 2 and nSPC = 2. c Mean number of genes in the predictors from cross-validation for various values of sumabsv and nSPC