Comparison of CMImpute and baseline imputation performance measured via sample-wise Pearson correlation with held-out observed data across all probes. The y -axis is CMImpute’s performance on each imputed combination. The x -axis is the species baseline’s (top left), logistic regression’s (top right), global baseline (bottom left), or tissue baseline’s (bottom right) performance on each imputed combination. Each dot is a single imputed species-tissue combination mean sample. The black diagonal line represents equal performance between CMImpute and the baseline. If a point is above the diagonal, CMImpute outperforms the baseline on the corresponding imputed combination mean sample and vice versa. Values in the upper left corners are the fractions of samples where CMImpute outperforms the corresponding baseline
