Shaping Brain Structure: Genetic and Phylogenetic Axes of Macro Scale Organization of Cortical Thickness

Structural and functional characteristics of the cortex systematically vary along global axes as a function of cytoarchitecture, gene expression, and connectivity. The topology of the cerebral cortex has been proposed to be a prerequisite for the emergence of human cognition and explain both the impact and progression of pathology. However, the neurogenetic origin of these organizational axes in humans remains incompletely understood. To address this gap in the literature our current study assessed macro scale cortical organization through an unsupervised machine learning analysis of cortical thickness covariance patterns and used converging methods to evaluate its genetic basis. In a large-scale sample of twins (n=899) we found structural covariance of thickness to be organized along both an anterior-to-posterior and inferior-to-superior axis. We found that both axes showed a high degree of correspondence in pairs of identical twins, suggesting a strong heritable component in humans. Furthermore, comparing these dimensions in macaques and humans highlighted similar organizational principles in both species demonstrating that these axes of cortical organization are phylogenetically conserved within primate species. Finally, we found that in both humans and macaques the inferior-superior dimension of cortical organization was aligned with the predictions of the dual-origin theory, highlighting the possibility that the macroscale organization of primate brain structure is subject to multiple distinct neurodevelopmental trajectories. Together, our study establishes the genetic basis of natural axes in the cerebral cortex along which structure is organized and so provides important insights into the organization of human cognition that will inform both our understanding of how structure guides function and for the progression of pathology in diseases.

Understanding of large-scale organization of brain structure may offer a novel and compelling and networks affected at different stages of the disorder 27,28 , and its sequence determined by 4 0 8 underlying anatomical axes. Parkinson's is assumed to show early disruptions in the lower 4 0 9 brain stem, followed later disruption in other midbrain structures, meso-cortex and allocortex. Parkinson's disease, as well as other neurodegenerative conditions. To conclude, our novel results establish that two major organizational axes in macro scale 4 1 9 organization of thickness in human and non-human primates that are likely to be at least  It is of note that our findings were made possible thanks to open data initiatives. These Sibships with individuals having severe neurodevelopmental disorders (e.g., autism), DZ-twins) with a mean age of 28.8 years (SD =3.7, range =22-37). Structural imaging processing 4 6 0 MRI protocols of the HCP are previously described in 78,79 . In short, MRI data used in the 4 6 1 study were acquired on the HCP's custom 3T Siemens Skyra equipped with a 32-channel 4 6 2 head coil. Two T1w images with identical parameters were acquired using a 3D-MPRAGE sequence (0.7 mm isotropic voxels, matrix = 320 × 320, 256 sagittal slices; TR = 2,400 ms, 4 6 4 TE = 2.14 ms, TI = 1,000 ms, flip angle = 8°; iPAT = 2). Two T2w images were acquired 4 6 5 using a 3D T2-SPACE sequence with identical geometry (TR = 3,200 ms, TE = 565 ms, 4 6 6 variable flip angle; iPAT = 2). T1w and T2w scans were acquired on the same day. The 4 6 7 pipeline used to obtain the Freesurfer-segmentation is described in detail in a previous article 4 6 8 78 and is recommended for the HCP-data. The pre-processing steps included co-registration of In addition to assess robustness and replicability of the results across different surface options. Surface-extraction and cortical thickness estimation using CIVET were performed then measured as the distance between the estimated "white" and "grey" cortical surfaces, in 4 8 0 the native space framework of the original MR images, using the same approach that is used focus on the granularity of 400 parcels, as averaging will improve signal-to-noise. In order to on the phenotypic correlation (ρ p ), between brain structure and personality with the following 5 4 2 formula: comparing the log likelihood for two restricted models (with either ρ g or ρ e constrained to be 5 4 5 equal to 0) against the log likelihood for the model in which these parameters were estimated.

4 6
A significant genetic correlation (corrected for multiple comparisons using Bonferroni phenotypic correlation, we computed the contribution of the genetic path to the phenotypic Geodesic distance was computed between each vertex in fsaverge5 space using the Eucledian 5 5 6 coordinates of the vertices, creating a 20484 x 20484 distance matrix. Only ipsilateral 5 5 7 distance was considered. Following distances between parcels were computed by taking the 5 5 8 average distance between both parcels. We evaluated the macro scale organization of 5 5 9 thickness while controlling for distance by multiplying the covariance strength by the distance between the respective parcels. Comparisons between gradients and modalities.

6 3
To make comparisons across gradient and distance maps, we used spin-tests to control for paleo-cortex distance and macro scale organizational gradients were assessed using statistical 5 6 6 energy test, a non-parametric statistic for two sample comparisons 51 (https://github. University of California, Davis).

7 5
Oxford data: The full data set consisted of 20 rhesus macaque monkeys (macaca mulatta) scanned on a 3T scanner with 4-channel coil. The data were collected while the animals were 5 7 7 under anesthesia. Briefly, the macaque was sedated with intramuscular injection of ketamine 1-2%. The details of the scan and anesthesia protocol can be found at 5 9 0 (http://fcon_1000.projects.nitrc.org/indi/PRIME/ucdavis.html).

9 6
MRI data processing: The structural processing includes 1) spatial denoising by a non-local 5 9 7 mean filtering operation 87 , 2) brain extraction using ANTs registration with a reference brain 5 9 8 mask followed by manually editing to fix the incorrect volume (ITK-SNAP, Quality control: We excluded macaque monkeys that showed a hemispheric difference of 6 0 3 more 0.2 cm (UC Davis (0); Oxford (7), Newcastle (5)) for our final analysis, as gradient 6 0 4 models were estimated based on covariance of ipsi-and contra-lateral covariance. global thickness. Following we performed gradient analysis analogue to described in humans. Archi-paleo cortex distance: Distance from the archi -and paleo cortex was computed in Cortical microstructure and microstructural covariance networks.

1 8
We estimated MPC using myelin-sensitive MRI (MPCMRI), in line with the previously 6 1 9 reported protocol 5 , in our main sample (HCP S900). The myelin-sensitive contrast was T1w/T2w from the HCP minimal processing pipeline, which uses the T2w to correct for 6 2 1 inhomogeneities in the T1w image. We generated 12 equivolumetric surfaces between the  (1) 6 2 7 in which α represents a fraction of the total volume of the segment accounted for by the 6 2 8 surface, while A out and A in represents the surface area of the outer and inner cortical surfaces, . In turn, MPC MRI (i,j) for a 6 3 2 given pair of parcels i and j is defined by (5): in which s is a participant and n is the number of participants. We used the MPC MRI to (re-) 6 3 5 compute the gradient of microstructure. The functional connectivity gradient was downloaded from (https://www.neuroconnlab.org) 6 3 9 computed as part of 13 , based on 820 individuals from the HCP S900 release. As the gradient 6 4 0 was reported at the fs_32k standard space surface, values were resampled for the Schaefer  correspondence between personality and cortical brain structure in the enhanced Nathan Kline Institute-Rockland Sample (NKI). The sample was made available by the Nathan-Kline (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3472598/). In short, eNKI was designed to community. Participants below 6 years were excluded to balance data losses with scientific 6 5 4 yield, as well as participants above the age of 85, as chronic illness was observed to analyses, we selected individuals with complete personality and imaging data. Our sample for  3D magnetization-prepared rapid gradient-echo imaging (3D MP-RAGE) structural scans 88 6 6 5 were acquired using a 3.0 T Siemens Trio scanner with TR=2500 ms, TE=3.5 ms, surface area maps were standardized to fsaverage5 for further analysis. Segmentations were  Modulation of structural covariance of thickness by age 6 7 5 In the eNKI sample, we also computed the modulation of structural covariance by probing the 6 7 6 interaction of covariance by age in the following model: Following the parcel to parcel t-maps were used to compute large-scale gradients age-related 6 7 9 changes in covariance. Cortical thickness of the individuals of the HCP S1200 release were computed as part of an 6 8 3 independent study (Kharabian, under review) and resampled to Schaefer 400 parcels. We (http://www.bic.mni.mcgill.ca/ServicesSoftware/CIVET). gradients are available at In a last step we evaluated the association between the two main axis of regional covariance participants (HCP S900 sample). We observed a strong negative relationship between G1 scov  with the two most outer strata (layer 1: r=0.60, layer 2: r=0.40), but not with layers closer to profile covariance and topological organization of structural covariance of cortical thickness. To do so, we computed the mean microstructural profile covariance (MPC) maps across 8 1 3 individuals and preformed gradient decomposition. We observed, as previously reported 6 , a Shaping Brain Structure sensory-motor to frontal cortices. We found that the first MPC gradient showed a close 8 1 7 correlation with the inferior-superior gradient of genetic covariance of thickness (r=0.62, p<0.00001), but not with the posterior-anterior gradient of genetic covariance of thickness 8 1 9 (r=-0.02). Conversely, the secondary gradient of MPC was associated with the posterior- anterior gradient of genetic covariance of thickness (r=0.30, p<0.00001), but not with the 8 2 1 inferior-superior gradient of genetic covariance (r=-0.09, p>0.1).
Relationship between large-scale organization of genetic correlation of regional thickness and functional connectivity topology.

2 5
Next, we evaluated the association between the posterior-anterior and inferior-superior heteromodal association areas (frontal and temporal cortex). Genetic correlation was observed 8 3 7 to vary as a function of the combination of gradients and was strongest in regions at similar to regions at different gradient levels. Functional topography along macro scale organizational patterns of thickness We conducted a meta-analysis using the Neurosynth 97 database and estimated the center of anteriorly. Various terms such as 'emotion' and 'reward' related to both posterior and anterior inferior regions. the average covariance of regions within the respective functional networks.   Correlation between layer-based T1w/T2w and the primary gradient of thickness covariance. the first two gradients in the eNKI dataset, using the Schaefer 400 parcellation. B). Gradient decomposition of t-maps of age-related modulation of structural covariance. based on CIVIT 2.1.0. standard pipeline.  G  r  a  d  i  e  n  t  s  i  n  c  y  t  o  a  r  c  h  i  t  e  c  t  u  r  a  l  l  a  n  d  s  c  a  p  e  s  o  f  t  h  e  i  s  o  c  o  r  t  e  x  :  1  1  1  8   D  i  p  r  o  t  o  d  o  n  t  m  a  r  s  u  p  i  a  l  s  i  n  c  o  m  p  a  r  i  s  o  n  t  o  e  u  t  h  e  r  i  a  n  m  a  m  m  a  l  s  .   J  C  o  m  p  N  e  u  r  o  l  5  2  5   ,  1  1  1  9   1  8  1  1  -1  8  2  6  ,  d  o  i  :  1  0  .  1  0  0  2  /  c  n  e  .  2  4  1  6  0  (  2  0  1  7  ) .  i  l  g  e  t  a  g  ,  C  .  C  .  T  o  w  a  r  d  s  a  "  c  a  n  o  n  i  c  a  l  "  a  g  r  a  n  u  l  a  r  c  o  r  t  i  c  a  l  m  i  c  r  o  c  i  r  c  u  i  t  .  1  1  6  5   F  r  o  n  t  N  e  u  r  o  a  n  a  t  8   ,  1  6  5  ,  d  o  i  :  1  0  .  3  3  8  9  /  f  n  a  n  a  .  2  0  1  4  .  0  0  1  6  5  (  2  0  1  4  )  .  1  1  6  6   7  6  W  a  g  s  t  y  l  ,  K  .  &  E  v  a  n  s  ,  A  .  B  i  g  B  r  a  i  n  3  D  a  t  l  a  s  o  f  c  o  r  t  i  c  a  l  l  a  y  e  r  s  :  c  o  r  t  i  c  a  l  a  n  d  l  a  m  i  n  a  r  1  1  6  7   t  h  i  c  k  n  e  s  s  g  r  a  d  i  e  n  t  s  d  i  v  e  r  g  e  i  n  s  e  n  s  o  r  y  a  n  d  m  o  t  o  r  c  o  r  t  i  c  e  s  .  (  B  i  o  a  r  X  i  v  )  .  1  1  6  8   7  7  M  a  r  c  u  s  ,  D  .  S  .   e  t  a  l  .   H  u  m  a  n  C  o  n  n  e  c  t  o  m  e  P  r  o  j  e  c  t  i  n  f  o  r  m  a  t  i  c  s  :  q  u  a  l  i  t  y  c  o  n  t  r  o  l  ,  d  a  t  a  b  a  s  e  1  1  6