Fig. 3.

MIRTH achieves high accuracy in cross-dataset imputation and preserves biological characteristics in the data. a The same metabolites distinguish tumor and normal samples in RC12 and RC3. b Tumor-distinguishing metabolite patterns are recovered by MIRTH imputation of RC3 and RC12. c Schematic of experiment to assess the imputation of features that were entirely unmeasured in a dataset. d Typical by-metabolite $\rho$ values when those metabolite features are entirely masked from a target dataset. Metabolites are ordered by decreasing imputation performance. e By-metabolite performance summarized across target datasets, showing that many of the same metabolites are well-predicted in many target datasets. f Relationship between actual and predicted within-dataset metabolite ranks for two reproducibly well-predicted metabolites. Each point represents one sample in which the metabolite was measured