Fig. 3
From: Image guided construction of a common coordinate framework for spatial transcriptome data
Validation of the image registration module of STaCker trained by synthetic data. A synthetic image not used for the training (a) or a real tissue histology image (b) was digitally distorted to a low, medium, or high level, or manually warped to generate a series of moving images. The original unwarped images served as the references. The degree of the distortions is quantified as the “moving” NCC score in the bar plots. The image module of STaCker, the Affine alignment, and the non-linear methods offered by ANTs and WSIreg, were applied to align each of the moving images to the reference. The aligned images from each method are displayed, together with their post-alignment NCC scores shown in the bar plots. The displayed NCC score of ANTs and STaCker is the mean over 10 repeated runs, with the associated error bars showing the standard errors. The statistical significance of the increase in STaCker’s NCC values relative to other methods is assessed by two-sided student t-test and marked by asterisks and explained in the annotation.
