Fig. 4. Use of neuromaps to contextualize two exemplar brain maps.

To demonstrate the utility of neuromaps we use the toolbox to transform, profile and quantitatively assess structural and functional enrichment for two example brain maps (that is, source maps). a, A volumetric map of cortical thinning in patients diagnosed with chronic schizophrenia from the NUSDAST repository (n = 133 patients versus n = 113 controls⁴⁶) was estimated by applying deformation-based morphometry to T1-weighted MRI scans to calculate the extent of gray matter expansion or contraction in patients relative to controls⁴⁷. Warm colors represent regions with greater thinning. The map was transformed to the native space of each brain map to which it was correlated (that is, the target maps). b, A surface-based brain map of structural evolutionary expansion represents the ratio of the surface area in humans to that of macaques, as computed using interspecies surface-based registration¹⁹. Warm colors indicate regions with greater evolutionary expansion. This map (the source map) was transformed to the native space of each brain map to which it was correlated. Spatial Pearson’s correlations were assessed against a two-sided spatial autocorrelation-preserving null model (‘spin test’)³⁸. Points represent the empirical Pearson’s correlation between source and target maps (with significance defined as P_spin < 0.05). In the boxplots the ends of the boxes represent the first (25%) and third (75%) quartiles, the center line (median) represents the second quartile of the null distribution (n = 1,000 rotations), the whiskers represent the non-outlier end-points of the distribution, and open circles represent outliers. All correlations were corrected for multiple comparisons⁴⁸.