Fig. 2: Spot2vector demonstrates superior domain identification performance. | Communications Biology

Fig. 2: Spot2vector demonstrates superior domain identification performance.

From: Flexible integration of spatial and expression information for precise spot embedding via ZINB-based graph-enhanced autoencoder

Fig. 2

a Performance comparison of seven domain identification methods (Spot2vector, DeepST, SpaGCN, Scanpy, Stlearn, STAGATE, GraphST) tested on six ST datasets, using ARI as the evaluation metric. b ARI scores (x-axis) of seven methods (y-axis) tested on 12 DLPFC sections. Each box plot ranges from the first and third quartiles with the median as the vertical line, while whiskers represent 1.5 times the interquartile range from the lower and upper bounds of the box. Data beyond the end of the whiskers are plotted individually. c Spatial plots of domain annotation, and seven methods on the section 151676 of DLPFC dataset. Each method colors the spot with their predictive domain labels. d Spatial plots of domain annotation, and seven methods on the Mouse Brain dataset. Three regional boundaries are framed by dashed lines: yellow (Fiber_tract), black (Hypothalamus_2), white (Hypothalamus_1).

Back to article page