The Journal of Neuroscience, May 17, 2006, ():
Symmetry of Cortical Folding Abnormalities in Williams Syndrome Revealed by Surface-Based Analyses
J. Neurosci. Van Essen et al.
26: 5470
Supplemental data
Files in this Data Supplement:
- supplemental material -
Supplemental material
- supplemental material
-
Supplementary-Figure 1. A, B. Structural MRI and fiducial surface contours for the right hemispheres of WS case 1 (sn 107.R) and CONTROL case 1 (115n.R). Note that the illustrated contours are from the resampled fiducial surface (see Methods) and thus are slightly smoothed and run slightly closer to the gray/white boundary than the original fiducial surface. C, D. Landmark contours visible from the medial wall of WS case 1, viewed on the fiducial surface (C) and the inflated surface (D). In the fiducial surface the non-cortical medial wall region has been smoothed relative to its configuration in the initial segmentation. Landmark contours are visible along the medial wall (purple, pink) and the calcarine sulcus (orange)of the fiducial (C) and inflated (D) surfaces. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6436987
- supplemental material
-
Supplementary-Figure 2. T-statistic maps of WS vs. CONTROL sulcal depth for the right hemisphere, comparing registration to two different targets. A. T-map obtained by registration to the average landmarks for the PALS-B12 right hemisphere target displayed on lateral view of the very inflated atlas surface. B. T-map obtained by registration to the combined (left + right) hybrid PALS-B12 landmarks. C. Difference map between the results for panels A and B. In all three panels, the colored regions indicate values of |t| > 3, whereas the grayscale regions indicate values of |t| < 3. D – F. Same as panels A – C, except displayed on the medial view of the very inflated atlas surface. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6414215.
- supplemental material
-
Supplementary-Figure 3. Maps of the average areal distortion in comparing individual fiducial surfaces to the PALS-B12 average fiducial surface. For each of the 12 left and right individual hemispheres contributing to the PALS-B12 atlas (Van Essen, 2005a), the distortion between the individual fiducial and the average fiducial surface was computed. The 12 distortion maps were then averaged, and are displayed on various PALS-B12 atlas surface configurations. A-E. Left hemisphere lateral (A) and medial (B) average fiducial surfaces, lateral (C) and medial (D) very inflated surfaces, and flat map (E). F-J. Corresponding maps for the right hemisphere. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6436972
- supplemental material
-
Supplementary-Figure 4. Thresholded t-statistics maps involving the age-matched WS group and the age-outlier WS group. A-C. T-statistics maps for the left hemisphere, comparing WS-age-matched vs CON13 (A), WS16 vs CON13 (B), and WS age-outlier vs WS age-matched (C). D-F. Corresponding comparisons for the right hemisphere. G-I. T-correlation maps for the left vs. right hemispheres for each group comparison, thresholded at |t|>4. Arrow in panel C points to a statistically non-significant region of interest in the parieto-occipital sulcus. Arrows in panels D – F point to a region of interest in the superior temporal gyrus. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6436975.
- supplemental material
-
Supplementary-Figure 5. Depth-vs-age scatterplots for selected regions of interest in the right hemisphere. A. Average depth within the olfactory sulcus (region 17 in Fig. 3) for each individual right hemisphere. B - D Depth vs age scatterplots for the intraparietal sulcus (panel B), posterior insula (panel C), and superior temporal gyrus (panel D). There is no age-related trend that would account for the group difference between WS and CONTROL subjects in panels A – C, but in panel D there is more overlap for the age-matched subgroup than for the younger or older subgroups.
- supplemental material
-
Supplementary-Figure 6. Thresholded t-statistics maps involving the WS-16 group, the B24 group (24 Washington University normal subjects), and the CON-13 group (13 UCSD normal subjects). A-C. T-statistics maps for the left hemisphere, comparing WS16 vs B24 (A), WS16 vs CON13 (B), and CON13 vs B24 (C). D-F. Corresponding comparisons for the right hemisphere. G-I. T-correlation maps for the left vs. right hemispheres for each group comparison, thresholded at |t|>4. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6436978. A potential confound of this analysis is that the distortions arising from susceptibility artifacts are not identical for different scanners. However, given the scanning parameters and bandwidth values used for the MP-RAGE sequences, we estimate that the distortion differences for overall head dimensions were negligible (< 1mm) for the two scanners resulted.
- supplemental material
-
Supplementary-Figure 7. Multi-dimensional scaling (MDS) plots based on cross-correlation analyses of individual sulcal depth maps for individual hemispheres and hemispheric subregions. A. MDS plot for the 29 WS and CONTROL left hemispheres. B. MDS plot for the corresponding 29 right hemispheres. C. MDS plot for a broad swath of the right hemisphere that encompasses just the regions containing statistically significant folding abnormalities. D. MDS plot for the region outside this swath. Blue = data points for the CONTROL hemispheres; red = data points for the WS hemispheres. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6414221.
- supplemental material
-
Supplementary-Figure 8 Volume-based differences between WS and CONTROL groups. A. Coronal slice through the average structural MRI volume for 16 WS subjects after registration to 711-2B space. B. Corresponding slice through the average CONTROL MRI volume. C. Intensity difference volume (average WS minus average CONTROL). D. A thresholded volume-intensity t-statistic (|t| > 2) for the WS vs. CONTROL volumes (red/yellow = positive t; green/blue = negative t). (This is analogous to the node-based t-statistic maps illustrated in Fig. 2 generated from sulcal depth maps.) The average CONTROL MRI is the underlay. Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6414212.
- supplemental material
-
Supplementary-Figure 9. A, B. Maps of t′-statistics for the left (A) and right (B) hemispheres, generated by multifiducial mapping of the volume-intensity t-statistics map to the PALS atlas left and right hemispheres (red/yellow = t′ > 1; green = t′ < -1). Results are displayed on the atlas flat map and very inflated map. Spheres represent folding abnormalities projected from their stereotaxic coordinates (cf. Table 2) onto the PALS-B12 average fiducial surface. C. A t′-correlation map, generated by taking the product of t′ maps for the left and right hemispheres. The nine largest patches are labeled by Roman numerals in an approximately posterior to anterior sequence. D,E. Stereotaxic foci from VBM studies of Reiss et al., (2004), and Meyer-Lindenberg et al. (2004), overlaid on the map of t′ from panels A and B. Blue squares indicate WS > CONTROL and maroon squares indicate CONTROL > WS gray matter density from Reiss et al., (2004); gray squares indicate the Meyer-Lindenberg et al. (2004) study.
Data are available at http://sumsdb.wustl.edu/sums/directory.do?dirid=6436984.
