Fig. 2: The comparison of imputed results for osmFISH-based data (known as osmFISH-brain).

a Cell-type distribution across spatial location for osmFISH data. b Metrics information of different imputation methods. For SCC and SSIM, higher scores represent better performance, and the methods are sorted in descending. For RMSE and JS, lower scores represent better performance, and the methods are sorted in ascending (sample size (n) = 10 for each group; center point, median; box limits, upper and lower quartiles; whiskers, up to 1.5 × interquartile range; points, outliers). c Examples of the expression levels of reliably imputed genes by VISTA. We compared the predicted gene expression levels with observed expression levels and highlighted the regions with relatively higher gene expression levels with circles. d Examples of the top marker genes by VISTA for each cell type of osmFISH-brain. The sequence of genes corresponds to the sequence of cells in (a).