Fig. 3: Interpretation of sub-predictors across ChronoGauge’s ensemble. | Nature Communications

Fig. 3: Interpretation of sub-predictors across ChronoGauge’s ensemble.

From: Machine learning models highlight environmental and genetic factors associated with the Arabidopsis circadian clock

Fig. 3: Interpretation of sub-predictors across ChronoGauge’s ensemble.

a Comparison of phase estimates made by each sub-predictor within ChronoGauge for the test accession published by Rugnone et al.24. Colour-scale represents the mean-error of phase estimates made for each timepoint. b Enrichment of biological processes terms across different sub-predictor groups in ZT10, including those with accurate (<60 min absolute-error), advanced (+60 min error) and delayed (−60 min error) CT estimates. Term enrichment was determined using Fisher’s exact test in TopGO38 (Q < 0.05) using all circadian-regulated genes as a background. Significance determined by Benjamini–Hochberg adjusted P-values. Figure source data are provided as a Source Data file.

Back to article page