Figure 3.

Evaluation of imputation methods

(A) Cosine similarity scores (CSSs) for OsmFISH, STARmap, and MERFISH spatial transcriptomic datasets.

(B) Line plots of imputation performances in different uncertainty quantile ranges, where the x axis denotes the quantiles of predicted uncertainty and the y axis denotes the median cosine similarity score of genes below the corresponding uncertainty quantile.

(C) Averaged clustering scores from multiple metrics (“method”).

(D) Averaged computational runtime of different methods.