Genetic and environmental influences on structural variability of the brain in pediatric twin: Deformation based morphometry

Neuroscience Letters 493 (2011) 8–13

Contents lists available at ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

Genetic and environmental influences on structural variability of the brain in pediatric twin: Deformation based morphometry Uicheul Yoon a,c,d , Daniel Perusse b , Jong-Min Lee c,∗ , Alan C. Evans a a

McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal, Quebec H3A 2B4, Canada The Research Centre at the Sainte Justine Hospital, Montreal, Canada Department of Biomedical Engineering, Hanyang University, Sung-dong P.O. Box 55, Seoul 133-605, South Korea d Department of Biomedical Engineering, Catholic University of Daegu, 330 Geumnak 1-ri, Hayang-eup, Gyeongsan 712-702, South Korea b c

a r t i c l e

i n f o

Article history: Received 15 September 2010 Received in revised form 28 January 2011 Accepted 28 January 2011 Keywords: Magnetic resonance imaging Pediatric twin Jacobian determinant Genetic effect Environmental effect Structural equation modeling

a b s t r a c t Twin studies are one of the most powerful study designs for estimating the relative contribution of genetic and environmental influences on phenotypic variation inhuman brain morphology. In this study, we applied deformation based morphometry, a technique that provides a voxel-wise index of local tissue growth or atrophy relative to a template brain, combined with univariate ACE model, to investigate the genetic and environmental effects on the human brain structural variations in a cohort of homogeneously aged healthy pediatric twins. In addition, anatomical regions of interest (ROIs) were defined in order to explore global and regional genetic effects. ROI results showed that the influence of genetic factors on cerebrum (h2 = 0.70), total gray matter (0.67), and total white matter (0.73) volumes were significant. In particular, structural variability of left-side lobar volumes showed a significant heritability. Several subcortical structures such as putamen (h2ROI = 0.79/0.77(L/R), h2MAX = 0.82/0.79) and globus pallidus (0.81/0.76, 0.88/0.82) were also significantly heritable in both voxel-wise and ROI-based results. In the voxel-wise results, lateral parts of right cerebellum (c2 = 0.68) and the posterior portion of the corpus callosum (0.63) were rather environmentally determined, but it failed to reach statistical significance. Pediatric twin studies are important because they can discriminate several influences on developmental brain trajectories and identify relationships between gene and behavior. Several brain structures showed significant genetic effects and might therefore serve as biological markers for inherited traits, or as targets for genetic linkage and association studies. © 2011 Elsevier Ireland Ltd. All rights reserved.

Twin studies yield valuable information for distinguishing the relative contribution of genetic and environmental influences on phenotypic variation of human brain morphology [19]. Classical twin analysis is based on the difference in intra-pair similarity between monozygotic (MZ) twin pairs who are genetically identical, and dizygotic (DZ) twin pairs who share approximately half of their segregating genes. If MZ twin pairs resemble each other more closely than DZ ones do for a certain trait, it can be inferred that there is a significant genetic influence on that trait [7]. Traditional volumetric measures for regions of interest (ROIs) have been used to quantify the similarity between MZ and DZ twins in the early stages of neuroanatomical genetic studies. This approach has revealed highly heritable brain features, such as total cerebral volume [8,19] and individual gray matter (GM) and white matter (WM) volumes [8]. Corpus callosum morphometry has been

∗ Corresponding author. Tel.: +82 2 2220 0685; fax: +82 2 2296 5943. E-mail address: [email protected] (J.-M. Lee). URL: http://cna.hanyang.ac.kr (J.-M. Lee). 0304-3940/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2011.01.070

shown to be controlled by genetic factors and this has been verified at different stages in life [19]. However, ventricular volumes and superficial gyral patterns seem to be almost entirely mediated by environmental factors [2]. Although it has been reported that individual lobar brain volumes have lower heritability than the whole brain volume, this might be a consequence of poor reliability in regional partitioning rather than any substantial difference in heritability for these regions [8]. Our previous heritability study of cortical thickness, with the same population as examined in this study, found that leftlateralized genetic and right-lateralized common environmental influences accounted for individual differences in human brain structures, but it did not investigate genetic influences on subcortical structures [27]. A recent heritability study of voxel-based morphometry in pediatric twin pairs showed high heritability of intracranial, total brain, cerebellum, and GM and WM volumes, a finding that is consistent with previous adult studies [2,8,19,20]. By contrast, deformation based morphometry (DBM) computes information on regional volumetric differences between groups by warping brain images to a canonical brain template. DBM may be

U. Yoon et al. / Neuroscience Letters 493 (2011) 8–13

more sensitive than VBM since it requires no a priori knowledge of GM or WM distributions throughout the brain, and the warping algorithm for DBM allows for better detection of subtle GM differences in comparison with VBM, as explained previously [1]. The aim of the present study was to examine the genetic and environmental effects on human brain structural variation in a cohort of homogeneously aged (8-year-old) healthy pediatric twins. We employed DBM to analyze neuroanatomical variability and compared cross-twin correlations of structural variation across the two groups using structural equation modeling to quantify both genetic and environmental effects on a voxel-wise basis. In addition, various anatomical ROIs were defined in order to explore global and regional genetic effects. The pediatric subjects in the present study are enrolled in the Quebec Newborn Twin Study (QNTS), an ongoing longitudinal study of twins from the Province of Quebec, Canada [27]. It has been reported that heritability estimates are not significantly affected by sex [15], but we chose same-sex twin only to avoid any confounding effect. Fifty seven MZ (male/female = 22/35) and 35 DZ (15/20) twin pairs were scanned at 8 years old using the same MRI scanner (Supplement Table 1). All the subjects scanned were within the normal gestational age and birth weight for a twin population, and these measures were not significantly different between groups [27]. Written informed consent was obtained from parents after full explanation of the aims and procedures for this study. The study protocol was approved by the scientific and ethics committees of Sainte-Justine and Notre Dame Hospitals in Montreal, Canada. MRI data were obtained on a 1.5 Tesla system (Magnetom Vision, Siemens Electric, Erlangen, Germany). A three-dimensional T1 -weighted, sagittal, fast low-angle shot (FLASH) of the whole head, designed to optimally discriminate between brain tissues (TE = 10 ms, TR = 22 ms, flip angle = 30◦ , 160 contiguous slices; matrix size = 224 × 256; 1 mm × 1 mm × 1 mm voxels) was acquired. The following pipeline image processing steps were applied for further analysis, as described in detail elsewhere [6,22,24,28]. At first, the native MRIdata of all subjects were registered into the pediatric template using a linear transformation and corrected for intensity non-uniformity artifacts [22,27]. A hierarchical multi-scale non-linear fitting algorithm [6] was then applied to: (i) normalize the individual MR images in stereotaxic space, (ii) provide a priori information i.e. tissue probability maps for subsequent tissue classification using a neural network classifier [28] and (iii) obtain the 3D deformation vector field that mapped each individual brain volume onto the template. The determinant of the Jacobian matrix (||J||) of the deformation field expresses the volume of the unit-cube at each voxel after deformation [13]. A value of ||J|| > 1 at any voxel implies a local expansion of the image being studied in comparison to the template, while ||J|| < 1 signifies a local shrinkage. An artificial neural network classifier was applied to identify GM, WM and cerebrospinal fluid. Partial volume errors were estimated and corrected using a trimmed minimum covariance determinant method [24,28]. As well as voxel-wise analysis, we also investigated the genetic and environmental contributions to various anatomical ROIs: whole hemisphere, GM and WM, lobar cortical GM and WM, subcortical GM including putamen, thalamus, caudate and globus pallidus, lateral ventricle, corpus callosum, and cerebellum (Supplement Fig. 1). The GM map was divided into cortical and subcortical regions using a subcortical GM stereotaxic mask, and the corpus callosum was defined from the WM map using a midsagittal brain mask. The other ROIs including lobar cortical GM and WM, lateral ventricle, cerebellum and several subcortical structures were automatically identified using ANIMAL (Automated Non-linear Image Matching and Anatomical Labeling), a well-established non-linear warping algorithm that uses a multi-

9

scale approach to deform one MRI volume to match a previously labeled template [5]. The volume of each ROI for any subject was calculated by integrating ||J|| over the corresponding ROI in the pediatric template. Univariate twin analysis using the ACE model tests the hypothesis that the variance of any phenotype is decomposed into the proportion of additive genetic (A), common environmental (C), and unique environmental (E) influences with the total variance being equal to A + C + E [18]. The A term manifests an approximate factor of two difference in correlation between MZ and DZ twins and the C term quantifies the environmental factors that influence both members of a pair equally. Although the E term nominally reflects the degree to which MZ twins raised together are dissimilar, it is in practice a residual term that also includes the effect of interactions between genes and environment, as well as measurement error. Univariate ACE models were constructed using the maximum likelihood-based structural equation modeling (SEM) software Mx [18]. The difference in maximum likelihood between a full model and a nested submodel generated by the removal of a parameter of interest follows a 2 distribution with degrees of freedom equivalent to the difference in the number of parameters between the competing models. The threshold for statistical significance was set at an ˛ of 0.05. Correction for multiple comparisons was performed using false discovery rate (FDR) at a q-value of 0.05. The color-coded influences of the genetic, common and unique environmental factors on a voxel-wise structural variability are mapped on the pediatric template (Fig. 1). Although there was no significant gender difference, regions with significant genetic effects were found bilaterally in the lateral fronto-orbital gyrus, cerebellum and several subcortical structures including thalamus, amygdala, putamen, and globus pallidus. Unilateral foci were found in the left frontal WM, inferior temporal gyrus and uncus, and the right temporal WM and superior frontal gyrus. In addition, the brain stem and the anterior portion of the corpus callosum (genu) were shown to be considerably influenced by genetic factors (Fig. 2 and Supplement Table 2). Common environmental effects were estimated up to 0.71 and more prominent than the genetic contribution in the bilateral insula, posterior cingulate region, lateral parts of right cerebellum (cerebrocerebellum) as well as the posterior portion of the corpus callosum (splenium), but failed to reach statistical significance in any region of the brain (Supplement Fig. 2 and Supplement Table 3). Even though the unique environment variance maps demonstrate high variance throughout the brain including cortical GM, caution must be exercised when interpreting these maps since it is impracticable to distinguish unique environmental effects from sources of measurement errors that are uncorrelated between the twins (Fig. 1). Similar to the voxel-wise analysis, univariate ACE models were applied to estimate the effects of the three ACE factors on global and regional structure volumes. The influence of genetic factors on cerebrum, total GM, and total WM volumes were significant, but only left-side volumes showed a significant heritability (Table 1). Even though it was not statistically significant, the genetic influence on WM volumes (Left [95% confidence interval]/Right: 0.72 [0.44,0.82]/0.62 [0.07,0.74]) was relatively higher than GM volumes (0.62 [0.19,0.75]/0.57 [0.00.70]) and the similar pattern was observed in the lobar GM and WM volumes (Table 2). The frontal lobe showed bilaterally significant genetic effects on GM and WM volumes, but statistically significant heritability was shown in only left temporal (WM: 0.62 [0.16,0.83]) and parietal (GM: 0.59 [0.20,0.73]; WM: 0.61 [0.17,0.74]) lobes. In contrast, a rather high common environmental effect was shown in right temporal lobe (GM: 0.26 [0.00,0.66]; WM: 0.24 [0.00,0.65]). In the case of subcortical GM structures, the bilateral putamen and globus pallidus were significantly influenced by genetic fac-

10

U. Yoon et al. / Neuroscience Letters 493 (2011) 8–13

Fig. 1. Voxel-wise maps of variance components for additive genetic (A), common (C) and unique environmental (E) factors for structural variability of the brain. In each map, the proportion of the overall variance is expressed on a scale of 0.0 (purple) to 0.8 (red) for the convenience of visualization. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

tors, but right thalamus and caudate showed a rather high common environmental effect (Table 2). Additive genetic influences on lateral ventricles were relatively high (0.49–0.64), but failed to reach statistical significance. Finally, the heritability estimates for the

cerebellum (0.42–0.69) and corpus callosum (0.79) were statistically significant even though relatively high influence of common environmental factors on right cerebellum was observed (0.49, p = 0.022).

Fig. 2. Brain regions with significant genetic effects on structural variability. Each voxel is color-mapped for level of significance following the application of an FDR threshold. (a) Globus pallidus (Z = −2), (b) right temporal white matter (Z = 24), (c) brain stem and the anterior portion of the corpus callosum (X = 0), (d) cerebellum (X = −18), (e) left frontal white matter and lateral fronto-orbital gyrus (Y = 24) and (f) bilateral putamen (Y = −3). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

U. Yoon et al. / Neuroscience Letters 493 (2011) 8–13

11

Table 1 Maximum likelihood parameter estimates of univariate ACE models for structural variability of the brain. Brackets indicate 95% confidence intervals. Variance components (95% confidence intervals)

Cerebrum

Gray matter

White matter

W L R W L R W L R

Corpus callosum

ACE vs. CE

a

c

e

2

p

0.70 (0.34, 0.81) 0.67 (0.31, 0.78) 0.58 (0.06, 0.72) 0.67 (0.24, 0.78) 0.62 (0.19, 0.75) 0.57 (0.00, 0.70) 0.73 (0.40, 0.82) 0.72 (0.44, 0.82) 0.62 (0.07, 0.74) 0.79 (0.37, 0.86)

0.00 (0.00, 0.32) 0.00 (0.00, 0.32) 0.00 (0.00, 0.45) 0.00 (0.00, 0.38) 0.00 (0.00, 0.38) 0.00 (0.00, 0.51) 0.00 (0.00, 0.30) 0.00 (0.00, 0.24) 0.00 (0.00, 0.48) 0.00 (0.00, 0.40)

0.30 (0.20, 0.45) 0.33 (0.22, 0.50) 0.42 (0.28, 0.61) 0.33 (0.22, 0.50) 0.38 (0.25, 0.55) 0.43 (0.30, 0.63) 0.27 (0.18, 0.42) 0.28 (0.18, 0.43) 0.38 (0.26, 0.56) 0.21 (0.14, 0.32)

11.29 9.88 4.62 8.18 6.90 3.42 13.37 14.83 4.83 14.44

<0.001 0.002 0.032 0.004 0.009 0.064 <0.001 <0.001 0.028 <0.001

2

2

Most previous twin studies have focused on the genetic influence on the variation in whole brain volume and total GM and WM volume [2,8,19], but only a few studies have examined genetic effects on more specific brain areas [3,9,20,21]. Cortical GM areas exhibited relatively high unique environmental effects in the voxelwise results of the present study while significant genetic effects were shown in several subcortical structures and the anterior portion of the corpus callosum, and relatively high common environmental ones were found from the insula, cingulate region and the posterior portion of the corpus callosum. These results were consistent with the latest VBM study with 9-year-old twin pairs showing that the main proportion of individual variation in overall brain volume is influenced by genetic factors, but GM density as measured with VBM versus structural variability from DBM might have yield different heritability estimates [20]. In general, VBM removes global shape differences between brains at the spatial normalization step, but retains local differences. However, DBM characterizes these global differences in

2

macroscopic anatomy that complements VBM, allowing one to examine differences at mesoscopic and macroscopic scales [1]. And, it may be more sensitive than VBM since it requires no a priori knowledge of GM or WM distributions throughout the brain. The nonlinear registration algorithm with a multi-scale approach allows for better detection of subtle GM differences in comparison with VBM. However, registration accuracy influenced by the degree of spatial regularization involved in the warping approach is the most important issue for DBM. A highly regularized method, in which the deformation fields are extremely smooth, may increase the genetic proportion of variance locally if genetic effect was high for global measures [3]. Actually, the alignment error due to the considerable inter-individual variability of sulcal patterns may cause a high unique environmental effect in the surrounding cortical area. Registration methods developed for more precise cortical pattern matching can be used to compare data from subjects whose sulcal patterns are different. For example, the volumetric warping removes the gross variability existing between the cortical surfaces

Table 2 Maximum likelihood parameter estimates of univariate ACE models for structural variability of both hemispheres. Brackets indicate 95% confidence intervals. Variance components (95% confidence intervals) a2

Frontal

GM WM

Temporal

GM WM

Parietal

GM WM

Occipital

GM WM

Putamen Thalamus Caudate Globus pallidus Lateral ventricle Cerebellum * ** a

p < 0.05. p < 0.01 2 (ACE vs. AE) = 5.24, p = 0.022.

ACE vs. CE

c2

2

e2

Left

Right

Left

Right

Left

Right

0.76 (0.40, 0.84) 0.78 (0.47, 0.86) 0.59 (0.10, 0.80) 0.62 (0.16, 0.83) 0.59 (0.20, 0.73) 0.61 (0.17, 0.74) 0.53 (0.14, 0.68) 0.50 (0.13, 0.67) 0.79 (0.56, 0.86) 0.59 (0.05, 0.76) 0.49 (0.00, 0.74) 0.81 (0.53, 0.88) 0.49 (0.00, 0.76) 0.69 (0.37, 0.94)

0.61 (0.29, 0.75) 0.62 (0.33, 0.75) 0.40 (0.00, 0.76) 0.43 (0.00, 0.77) 0.43 (0.04, 0.61) 0.54 (0.08, 0.69) 0.43 (0.00, 0.67) 0.46 (0.00, 0.63) 0.77 (0.52, 0.85) 0.47 (0.01, 0.80) 0.26 (0.00, 0.70) 0.76 (0.39, 0.84) 0.64 (0.07, 0.76) 0.42 (0.19, 0.82)

0.00 (0.00, 0.33) 0.00 (0.00, 0.30) 0.12 (0.00, 0.55) 0.12 (0.00, 0.54) 0.00 (0.00, 0.33) 0.00 (0.00, 0.37) 0.00 (0.00, 0.31) 0.00 (0.00, 0.29) 0.00 (0.00, 0.21) 0.05 (0.00, 0.53) 0.12 (0.00, 0.59) 0.00 (0.00, 0.26) 0.15 (0.00, 0.60) 0.24 (0.00, 0.55)

0.00 (0.00, 0.26) 0.00 (0.00, 0.23) 0.26 (0.00, 0.66) 0.24 (0.00, 0.65) 0.00 (0.00, 0.30) 0.00 (0.00, 0.38) 0.09 (0.00, 0.55) 0.00 (0.00, 0.48) 0.00 (0.00, 0.24) 0.23 (0.00, 0.63) 0.32 (0.00, 0.65) 0.00 (0.00, 0.34) 0.00 (0.00, 0.50) 0.49a (0.09, 0.72)

0.24 (0.16, 0.37) 0.22 (0.14, 0.34) 0.30 (0.20, 0.45) 0.26 (0.17, 0.39) 0.41 (0.27, 0.60) 0.39 (0.26, 0.57) 0.47 (0.32, 0.68) 0.50 (0.33, 0.72) 0.21 (0.14, 0.34) 0.36 (0.24, 0.53) 0.39 (0.26, 0.58) 0.19 (0.12, 0.30) 0.36 (0.24, 0.53) 0.08 (0.05, 0.12)

0.39 (0.25, 0.58) 0.38 (0.25, 0.57) 0.34 (0.23, 0.51) 0.33 (0.22, 0.49) 0.57 (0.39, 0.80) 0.46 (0.31, 0.66) 0.48 (0.33, 0.70) 0.54 (0.37, 0.76) 0.23 (0.15, 0.35) 0.30 (0.20, 0.45) 0.42 (0.28, 0.61) 0.24 (0.16, 0.37) 0.36 (0.24, 0.54) 0.09 (0.06, 0.14)

Left

Right

14.28**

9.12**

17.31**

10.41**

5.71*

2.53

7.52**

3.00

**

7.00

4.36*

6.57**

4.91*

5.97*

1.76

*

5.79

2.08

21.72**

19.14**

4.54*

4.08*

2.93

0.87

**

21.62

13.86**

3.33

4.91*

31.80**

16.86**

12

U. Yoon et al. / Neuroscience Letters 493 (2011) 8–13

of each subject, and provides a good initialization for the subsequent surface-based registration which improves the accuracy of surface registration [16,17]. Global brain structures such as the cerebrum, total GM and WM showed statistically significant heritability for volumetric variance, consistent with previous studies [9,19]. Intriguingly, the volumetric variance for left hemisphere and its GM and WM components were significantly influenced by genetic factors, whereas this was not the case for the right hemisphere. In addition, structural variability of left-side lobar volumes showed a significant heritability. This was quite consistent with our previous heritability study of cortical thickness from the same dataset that revealed significant genetic effects in several cortical regions of the left hemisphere, those mostly involved in language processing and the somatosensory system [27]. The cerebral hemispheres might experience different genetic influences on cortical morphogenesis, with the languagedominant left cerebral cortex under stronger genetic control than the right [25]. Therefore, our finding is an intriguing association that may shed light on the neurobiological basis of human evolution. The bilateral putamen and globus pallidus were found to be significantly heritable in both voxel-wise and ROI-based results, again consistent with a previous MZ twin study [26]. The putamen is multiply connected to the globus pallidus and has as its main functions the regulation of movement and the modulation of various forms of learning. The putamen has also been implicated in several heritable psychiatric disorders such as schizophrenia and attention deficit hyper activity disorder (ADHD) [4,11]. The relatively high, but not significant, heritability of lateral ventricular volume is consistent with previous reports in older children and adults [9,19], but contrasts with the significant genetic effects found in the neonatal period [10]. It has been suggested that the lateral ventricle is significantly influenced by genetic factors in early development but environmental influences become progressively stronger during postnatal development [10]. Genetic influence was statistically significant for the midsagittal area of the corpus callosum, consistent with previous results [19]. The size of the corpus callosum has been known to be highly heritable, but there is also evidence of epigenetic effects from animal studies that variations in environment affect corpus callosum anatomy [12]. In our voxel-wise analysis, the anterior portion of the corpus callosum (genu), was shown to be highly influenced by genetic factors (Fig. 2), but the splenium, the posterior portion, showed a considerable common environmental influence (Supplement Fig. 2). This profile of genetic influences might be related to a rostro-caudal wave of growth pattern that was detected at the corpus callosum of young normal subjects aged 3–15 years [23]. This is consistent with the hypothesis that the earliest-maturing brain regions are more genetically influenced while environmental influences are progressively greater for laterdeveloping regions [3,14,15]. Common environmental effects on the volumetric variances of the bilateral insula, posterior cingulate region, and lateral parts of right cerebellum were considerable in both voxel-wise and ROI-based results. Although the classical twin method is generally insufficient for detecting common environmental influences on neuroanatomical variability due to subject heterogeneity, the present study provides a better estimate of environmental influence by virtue of the homogeneity in the ages and growing environment of the subjects [27]. In conclusion, we investigated the influence of genetic and environmental effects on the human brain structural variations in a cohort of healthy homogeneously aged pediatric twins. Gross brain volumes including total GM and WM were significantly influenced by genes and were left-lateralized. In particular, the structural variability of left temporal and parietal lobes showed significant

heritability. Several subcortical structures such as thalamus, putamen and globus pallidus were also significantly heritable. Even failing to reach statistical significance, lateral parts of the right cerebellum, especially cerebrocerebellum and the posterior portion of the corpus callosum, splenium, were rather environmentally determined. Pediatric twin studies are useful to discern several influences on developmental brain trajectories and to identify relationships between gene and behavior. Several brain morphometric measures showed significant heritability and can therefore serve as biological markers for inherited traits or targets for genetic linkage and association studies. Acknowledgements This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2006214-D00208), the National Research Foundation of Korea (NRF) grant funded by the Korea government (No.2010-0002136) and grants from the Fonds de la recherche en santé du Québec and Hôpital Sainte-Justine to Daniel Pérusse, and this project has been funded in whole or in part with Federal funds from the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, the National Institute of Mental Health, and the National Institute of Neurological Disorders and Stroke (Contract #s N01-HD02-3343, N01-MH9-0002, and N01-NS-9-2314, -2315, -2316, -2317, -2319 and -2320). Special thanks to the NIH contracting officers for their support. We also acknowledge the important contribution and remarkable spirit of John Haselgrove, Ph.D. (deceased). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.neulet.2011.01.070. References [1] J. Ashburner, K.J. Friston, Voxel-based morphometry – the methods, Neuroimage 11 (2000) 805–821. [2] W.F. Baare, H.E. Hulshoff Pol, D.I. Boomsma, D. Posthuma, E.J. de Geus, H.G. Schnack, N.E. van Haren, C.J. van Oel, R.S. Kahn, Quantitative genetic modeling of variation in human brain morphology, Cereb. Cortex 11 (2001) 816–824. [3] C.C. Brun, N. Lepore, X. Pennec, A.D. Lee, M. Barysheva, S.K. Madsen, C. Avedissian, Y.Y. Chou, G.I. de Zubicaray, K.L. McMahon, M.J. Wright, A.W. Toga, P.M. Thompson, Mapping the regional influence of genetics on brain structure variability – a tensor-based morphometry study, Neuroimage 48 (2009) 37–49. [4] F.X. Castellanos, W.S. Sharp, R.F. Gottesman, D.K. Greenstein, J.N. Giedd, J.L. Rapoport, Anatomic brain abnormalities in monozygotic twins discordant for attention deficit hyperactivity disorder, Am. J. Psychiatry 160 (2003) 1693–1696. [5] D.L. Collins, C.J. Holmes, T.M. Peters, A.C. Evans, Automatic 3-D model-based neuroanatomical segmentation, Hum. Brain Mapp. 3 (1995) 190–208. [6] D.L. Collins, P. Neelin, T.M. Peters, A.C. Evans, Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space, J. Comput. Assist. Tomogr. 18 (1994) 192–205. [7] D.S. Falconer, T.F.C. Mackay, Introduction to Quantitative Genetics, 4th ed., Longman Group, London, 1996. [8] D.H. Geschwind, B.L. Miller, C. DeCarli, D. Carmelli, Heritability of lobar brain volumes in twins supports genetic models of cerebral laterality and handedness, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 3176–3181. [9] J.N. Giedd, J.E. Schmitt, M.C. Neale, Structural brain magnetic resonance imaging of pediatric twins, Hum. Brain Mapp. 28 (2007) 474–481. [10] J.H. Gilmore, J.E. Schmitt, R.C. Knickmeyer, J.K. Smith, W. Lin, M. Styner, G. Gerig, M.C. Neale, Genetic and environmental contributions to neonatal brain structure: a twin study, Hum. Brain Mapp. 31 (2010) 1174–1182. [11] A.L. Goldman, L. Pezawas, V.S. Mattay, B. Fischl, B.A. Verchinski, B. Zoltick, D.R. Weinberger, A. Meyer-Lindenberg, Heritability of brain morphology related to schizophrenia: a large-scale automated magnetic resonance imaging segmentation study, Biol. Psychiatry 63 (2008) 475–483. [12] J.M. Juraska, J.R. Kopcik, Sex and environmental influences on the size and ultrastructure of the rat corpus callosum, Brain Res. 450 (1988) 1–8. [13] J.C. Lau, J.P. Lerch, J.G. Sled, R.M. Henkelman, A.C. Evans, B.J. Bedell, Longitudinal neuroanatomical changes determined by deformation-based morphometry in a mouse model of Alzheimer’s disease, Neuroimage 42 (2008) 19–27.

U. Yoon et al. / Neuroscience Letters 493 (2011) 8–13 [14] R.K. Lenroot, J.N. Giedd, The changing impact of genes and environment on brain development during childhood and adolescence: initial findings from a neuroimaging study of pediatric twins, Dev. Psychopathol. 20 (2008) 1161–1175. [15] R.K. Lenroot, J.E. Schmitt, S.J. Ordaz, G.L. Wallace, M.C. Neale, J.P. Lerch, K.S. Kendler, A.C. Evans, J.N. Giedd, Differences in genetic and environmental influences on the human cerebral cortex associated with development during childhood and adolescence, Hum. Brain Mapp. 30 (2009) 163–174. [16] T. Liu, D. Shen, C. Davatzikos, Deformable registration of cortical structures via hybrid volumetric and surface warping, Neuroimage 22 (2004) 1790–1801. [17] O.C. Lyttelton, S. Karama, Y. Ad-Dab’bagh, R.J. Zatorre, F. Carbonell, K. Worsley, A.C. Evans, Positional and surface area asymmetry of the human cerebral cortex, Neuroimage 46 (2009) 895–903. [18] M.C. Neale, S.M. Boker, G. Xie, H.H. Maes, Mx: Statistical Modeling, 6th ed., Department of Psychiatry, Virginia Commonwealth University, Richmond, VA, 2003. [19] J.S. Peper, R.M. Brouwer, D.I. Boomsma, R.S. Kahn, H.E. Hulshoff Pol, Genetic influences on human brain structure: a review of brain imaging studies in twins, Hum. Brain Mapp. 28 (2007) 464–473. [20] J.S. Peper, H.G. Schnack, R.M. Brouwer, G.C. Van Baal, E. Pjetri, E. Szekely, M. van Leeuwen, S.M. van den Berg, D.L. Collins, A.C. Evans, D.I. Boomsma, R.S. Kahn, H.E. Hulshoff Pol, Heritability of regional and global brain structure at the onset of puberty: a magnetic resonance imaging study in 9-year-old twin pairs, Hum. Brain Mapp. 30 (2009) 2184–2196.

13

[21] J.E. Schmitt, R.K. Lenroot, G.L. Wallace, S. Ordaz, K.N. Taylor, N. Kabani, D. Greenstein, J.P. Lerch, K.S. Kendler, M.C. Neale, J.N. Giedd, Identification of genetically mediated cortical networks: a multivariate study of pediatric twins and siblings, Cereb. Cortex 18 (2008) 1737–1747. [22] J.G. Sled, A.P. Zijdenbos, A.C. Evans, A nonparametric method for automatic correction of intensity nonuniformity in MRI data, IEEE Trans. Med. Imaging 17 (1998) 87–97. [23] P.M. Thompson, J.N. Giedd, R.P. Woods, D. MacDonald, A.C. Evans, A.W. Toga, Growth patterns in the developing brain detected by using continuum mechanical tensor maps, Nature 404 (2000) 190–193. [24] J. Tohka, A. Zijdenbos, A. Evans, Fast and robust parameter estimation for statistical partial volume models in brain MRI, Neuroimage 23 (2004) 84–97. [25] M.J. Tramo, W.C. Loftus, C.E. Thomas, R.L. Green, L.A. Mott, M.S. Gazzaniga, Surface-area of human cerebral-cortex and its gross morphological subdivisions – in-vivo measurements in monozygotic twins suggest differential hemisphere effects of genetic-factors, J. Cogn. Neurosci. 7 (1995) 292–302. [26] T. White, N.C. Andreasen, P. Nopoulos, Brain volumes and surface morphology in monozygotic twins, Cereb. Cortex 12 (2002) 486–493. [27] U. Yoon, C. Fahim, D. Perusse, A.C. Evans, Lateralized genetic and environmental influences on human brain morphology of 8-year-old twins, Neuroimage 53 (2010) 1117–1125. [28] A.P. Zijdenbos, R. Forghani, A.C. Evans, Automatic pipeline analysis of 3-D MRI data for clinical trials: application to multiple sclerosis, IEEE Trans. Med. Imaging 21 (2002) 1280–1291.