Cerebral Cortical Folding, Parcellation, and Connectivity
Figure S1. Individual variability in structural MRI volume slices
Left and center columns: individual-subject axial T1-weighted MRI slices after rigidly aligning each native scan to a standard orientation in 3D space for macaque (top row), chimpanzee (middle row) and human (bottom row). Right column: population-average atlas volumes (N=19 for macaque Yerkes19_v1.2; N=29 for chimpanzee Yerkes29_v1.2; N=210 for human HCP Q1-Q6_RelatedValidation210), all generated by iterative nonlinear registration using FSL’s FNIRT algorithm. Scale bars (1 cm) on the left apply to all images in that row. In the macaque, the thickness and the pattern of cortical gray matter folding is comparable in the population average volume to that in individual scans, reflecting the consistency of folding across individuals. Further, the left and right hemisphere images are relatively symmetric by visual inspection. In the chimpanzee, the population average volume is slightly blurry in several cortical regions (green arrows) and is blurrier still in parts of the human population average (red arrows). The increased blurring arises because nonlinear volume-registration algorithms fail to align tissue boundaries between gray and white matter and cerebrospinal fluid in regions of high folding variability, as they do not take advantage of the topology of the cortical sheet. In terms of bilateral symmetry, the left and right hemispheres of individual chimpanzees and humans show many differences, but the population average volumes are relatively symmetric in both species.
Surface Mesh:32k fs LR, Species:Chimpanzee, Registration:MSMAll, Species:Human, Modality:Myelin Map, Atlas:Yerkes 19, Species:Macaque, Modality:T1-weighted, Other Data:gene