Fig. 2: Contribution of receptor distributions to explaining multimodal brain reorganization in PD.

Pathological factors are quantified by 6 neuroimaging-derived metrics: gray matter density (GM), neuronal activity (fractional amplitude of low frequency fluctuations; fALFF), dopamine transporter density (DAT) from SPECT, directed microstructure (fractional anisotropy; FA), undirected microstructure (mean diffusivity; MD), and dendrite density (t1/t2 ratio). The improvement in modeling the accumulation of pathology was evaluated in terms of (i) the additional explanatory power due to receptor information, and (ii) the significance of true receptor maps compared to null distributions. The histograms show the distribution of the coefficient of determination (R²) of N = 71 individual models of longitudinal neuroimaging changes including (a) and excluding (b) receptor predictors. Notably, including receptor terms improves model fit for all biological factors, although to varying extents. c Subject-wise F-tests between models with and without receptor maps (113 and 8 parameters, respectively) show proportions of subjects for whom the F-statistic is above the critical threshold (red dotted line). This critical threshold corresponds to a statistically significant (P < 0.05) improvement due to the receptor terms in the re-MCM model, accounting for the increase in adjustable model parameters. Furthermore, to validate the benefit of the receptor templates over randomized null maps, re-MCM models were fit with 1000 spatially permuted receptor maps for each subject. The p value of the model fit (R²) using true receptor templates compared to the distribution of R² of models using randomized templates was calculated for each subject. d Proportion of subjects for whom the true receptor maps resulted in a statistically significant improvement in model fit (P < 0.05; red dotted line).