White matter microstructural abnormalities in girls with chromosome
22q11.2 deletion syndrome, Fragile X or Turner syndrome as evidenced
by diffusion tensor imaging
Julio Villalon-Reinaa, Neda Jahanshada, Elliott Beatone, Arthur W. Togab,
Paul M. Thompsona,⁎, Tony J. Simonc,d
aImaging Genetics Center, Laboratory of Neuro Imaging, Dept. of Neurology, University of California Los Angeles, School of Medicine, Los Angeles, CA 90095, USA
bLaboratory of Neuro Imaging, Dept. of Neurology, University of California Los Angeles, School of Medicine, Los Angeles, CA 90095, USA
cDept. of Psychiatry and Behavioral Sciences, University of California, Davis, Sacramento, CA, 95618, USA
dMIND Institute, Dept. of Psychiatry and Behavioral Sciences, University of California, Davis, Sacramento, CA, 95618, USA
eStress, Cognition, and Affective Neuroscience Laboratory, Department of Psychology, University of New Orleans, New Orleans, LA, 70148
a b s t r a c ta r t i c l ei n f o
Accepted 10 April 2013
Available online 18 April 2013
Diffusion tensor imaging
Children with chromosome 22q11.2 deletion syndrome (22q11.2DS), Fragile X syndrome (FXS), or Turner syn-
drome (TS) are considered to belong to distinct genetic groups, as each disorder is caused by separate genetic al-
terations. Even so, they have similar cognitive and behavioral dysfunctions, particularly in visuospatial and
ing controls. We scanned 101 female children between 7 and 14 years old: 25 with 22q11.2DS, 18 with FXS, 17
with TS, and 41 aged-matched controls using diffusion tensor imaging (DTI). Anisotropy and diffusivity measures
were calculated and all brain scans were nonlinearly aligned to population and site-specific templates. We
performed voxel-based statistical comparisons of the DTI-derived metrics between each disease group and the
controls, while adjusting for age. Girls with 22q11.2DS showed lower fractional anisotropy (FA) than controls in
the association fibers of the superior and inferior longitudinal fasciculi, the splenium of the corpus callosum, and
capsule and left cerebellar peduncle. Partially overlapping neurodevelopmental anomalies were detected in all
three neurogenetic disorders. Altered white matter integrity in the superior and inferior longitudinal fasciculi
and thalamic to frontal tracts may contribute to the behavioral characteristics of all of these disorders.
© 2013 Elsevier Inc. All rights reserved.
Fragile X syndrome (FXS), or Turner syndrome (TS) are considered to
belong to genetically distinct groups, aseachdisorder has a specific, ap-
parently nonoverlapping, genetic cause. Yet, many of the cognitive dif-
ferences manifested by affected children are common across the three
disorders. Some of the major commonalities involve impairments in
visuospatial and numerical cognitive competence (Simon, 2007, 2011;
Simon et al., 2008a). It is therefore reasonable to hypothesize that
there may be shared patterns of altered brain development that lead
to these cognitive and behavioral changes.
22q11.2DS is the most common microdeletion syndrome with a
prevalence of 1 in 4000–7000 live births (Burn and Goodship, 1996).
In around 90% of cases (Bittel et al., 2009), the deletion on the long
(q) arm of chromosome 22 is 3 Mb in size and encompasses approxi-
mately 30 genes. Children with 22q11.2DS often show mild intellectual
impairment (IQ is typically in the 70–80s) and, in most, non-verbal
skills are affected more than verbal skills (De Smedt et al., 2007;
Wang et al., 2000). Other cognitive impairments involve disrupted
visuospatial attention and working memory, and impaired numerical
ability (Bearden et al., 2001; Simon et al., 2005b, 2008a,b). Individuals
with 22q11.2DS are at increased risk of developing psychiatric disor-
ders, in particular schizophrenia and schizoaffective disorder (Gothelf
et al., 2007a,b; Green et al., 2009; Stoddard et al., 2010). Differences
in brain morphology with respect to healthy subjects include volumet-
ric reductions in brain regions such as the hippocampus, thalamus and
neocortex. There is also reduced cortical thickness and a range of mid-
line anomalies including enlarged lateral ventricles, more prominent
cavum septum pellucidum, and callosal differences (Karayiorgou et al.,
NeuroImage 81 (2013) 441–454
⁎ Corresponding author at: Imaging Genetics Center, Laboratory of Neuro Imaging, UCLA
School of Medicine, 635 Charles Young Drive, Los Angeles, CA 90095, USA. Fax: +1 310 206
E-mail address: email@example.com (P.M. Thompson).
1053-8119/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/ynimg
2010). Animal models and human neuropathological studies of
22q11.2DS have shown that the main developmental disruption occurs
during neuronal migration (Kiehl et al., 2009; Meechan et al., 2009).
Other radiological reports in these subjects have included heterotopia
and polymicrogyria (Robin et al., 2006). Disrupted migration supports
the idea that neuronal connectivity may develop abnormally, making
it a natural target of study using diffusion tensor imaging.
FXS is an X-linked dominant neurodevelopmental disorder caused
by the silencing of the FMR1 (Fragile X Mental Retardation 1 gene),
due to an expansion of a CGG repeat in the 5′-untranslated region. A
“full mutation” exists when there are more than 200 triplet repeats
within the gene. Numerous studies estimate FXS prevalence at 1 in
4000 male births and 1 in 8000 female births for individuals with the
“full mutation” (Crawford et al., 1999). The mutation extinguishes the
expressionoftheFMR1 geneproduct, namely Fragile X MentalRetarda-
tion Protein (FMRP) (Oberle et al., 1991; Verkerk et al., 1991). FMRP is
found mainly in the perikaryon, dendrites and synapses. It is an mRNA
binding protein, transporting mRNAs from the transcription site to the
and coherence (Galvez and Greenough, 2005). These abnormalities in
neuronal development have a phenotypic expression in brain structure
that has been documented in brain imaging studies (Reiss et al., 1995a;
Kates et al., 1997; Lee et al., 2007; Eliez et al., 2001a). In addition to this,
subjects affected with FXS also present weaknesses in visuospatial
relationships among objects, attention and visual–motor coordination
(Cornish et al., 1999; Farzin and Rivera, 2010; Freund and Reiss, 1991;
Kogan et al., 2004; Mazzocco et al., 2006; Murphy et al., 2006; Scerif et
al., 2004, 2007) as well as problems inarithmetic reasoningandcompu-
tation, similar to those in 22q11.2DS (Murphy et al., 1993).
TS occurs in approximately 1 in 2000 live female births (Lippe, 1991).
These individuals lack a complete copy or a portion of one of the X chro-
mosomes. This deletion causes a myriad of physical characteristics and
medical problems, particularly cardiovascular malformations, kidney
malformations and nonfunctional ovaries. Females with TS do not have
notype, which mainly consists of difficulty with visuospatial tasks,
visuomotor control and, as with individuals with 22q11.2DS or FXS, im-
pairments in numerical ability (Beaton et al., 2010; Bruandeta et al.,
2004; Hart et al., 2006; Kesler et al., 2004; Mazzocco et al., 2006;
Murphy et al., 2006; Reiss et al., 1995b; Romans et al., 1998; Ross et al.,
2000; Rovet and Netley, 1980; Rovet and Ireland, 1994; Silbert et al.,
1977; Temple and Carney, 1995). The cognitive characteristics also vary
across subjects affected by TS. This is explained by multiple factors that
have been proven to affect brain development and function in
TS subjects: (1) mosaic or non-mosaic karyotype (Kesler et al., 2003);
(2) X-linked imprinting (Bishop et al., 2000); (3) lack of endogenous
estrogenic influence on brain development (Arnold and Gorski, 1984);
and growth hormones) (Nilsson et al., 1996). Imaging studies show de-
creased regional brain volumes in the parietal and occipital lobes, hippo-
campus, caudate nucleus, and thalamus (Brown et al., 2002, 2004;
Murphy et al., 1993; Reiss et al., 1995b).
Neuroimaging approaches may be used to uncover common pat-
terns of functional and anatomical differences in children with these
disorders to help relate genetic variations to brain organization and be-
havior. This is called a behavioral neurogenetics approach (Reiss and
Dant, 2003). To identify possible common mechanisms and pathways
involved in the etiology of these three disorders, we used diffusion ten-
sor imaging (DTI) to map the 3D profile of white matter abnormalities
in childrenwith22q11.2DS, FXS, and TScompared to typically develop-
ing(TD)children.DTI is a variantof magnetic resonance imaging(MRI)
that enables the study of the white matter microstructure, including
tracts. These tracts are formed by coherent bundles of axons, in which
water diffusion is hindered by cell membranes and the myelin sheaths
that surround them. DTI visualizes the directionality of water diffusion,
Here, we set out to determine whether specific patterns of abnor-
mal white matter microstructure are found within each of the genetic
disorders, and whether common patterns of neuroanatomical abnor-
malities can be observed within the three disorders. In terms of cog-
nition, there are two impairments shared by these 3 neurogenetic
syndromes — impairments in visuospatial processing and numerical
(including arithmetic) ability. Numerical and arithmetic abilities are
thought to share some neural/cognitive processes with visuospatial
tasks (representation of a mental number line, alignment of digits
in calculation, borrowing/carrying concepts). Since numerical abili-
ties develop later than visuospatial abilities it has become widely
accepted that the latter is constructed out of, or at least depends
heavily on, the former (Hubbard et al., 2005; Simon, 1997, 2011;
Walsh, 2003). Visual processing is thought to involve two “streams” of
information processing: the dorsal “where” stream (encoding informa-
tion on spatial position) and the ventral “what” stream (encoding infor-
mation on shape recognition and differentiation). The three syndromes
analyzed here may all involve dorsal visual processing stream impair-
ments (Bearden et al., 2001; Braddick et al., 2003; Farzin et al., 2008;
Hoeft et al., 2007a; Kesler et al., 2004). At least one review reports com-
monneuraldifferences (Walteretal., 2009)—primarilyabnormalstruc-
ture and function of the parietal lobe (important for visuospatial and
arithmetic tasks) and the superior longitudinal fasciculus (SLF).
Some initial DTI studies on 22q11.2DS report altered white matter
anisotropy in occipital, parietal and frontal regions (Barnea-Goraly et
al., 2003b; Simon et al., 2005a, 2008b). DTI analysis has revealed
lower anisotropy in frontal–striatal and parietal sensorimotor tracts
along the corona radiata and centrumsemiovale in FXS (Barnea-Goraly
et al., 2003a). And in TS, DTI studies reveal lower anisotropy in
fronto-parietal white matter, prefrontal cortex close to the caudate,
orbitofrontal region, and in bilateral internal capsules and higher FA in
the temporoparietal pathways (Holzapfel et al., 2006). We hypothe-
sized thatcommondevelopmentalabnormalities would befound with-
in the white matter structure across all three disorders, compared to
ciation fibers such as the SLF, the inferior longitudinal fasciculus (ILF)
and the inferior fronto-occipital fasciculus (IFO) connect lobes within
the hemispheres. They transfer information from anterior to posterior
regions of the brain or vice versa. These areas comprise the anterior
branch of the internal capsule, the external capsule, and non-specific
white matter in the temporal, parietal and frontal lobes.
We hypothesized that common developmental abnormalities would
be found within the white matter structure between all three disorders
when compared to controls, especially in those areas where long associ-
gitudinalfasciculus (ILF) andtheinferiorfronto-occipitalfasciculus(IFO)
are long association fibers that connect lobes within the hemispheres
transferring information from anterior to posterior portions of the brain
or vice versa. Those areas comprise the anterior branch of the internal
capsule, the external capsule, and non-specific white matter in the tem-
poral, parietal and frontal lobes.
Participants and scanning protocol
Quality control of the data was done to eliminate scans that did
not adhere to our protocol suffered from severe motion artifacts.
Scans were collected at two locations (Tables 1 and 2). One group in-
cluded the children with 22q11.2DS, FXS and corresponding
agematched TD controls (total 57 children) scanned at the University
of California Davis Medical Center, in Davis, California (UCDMC). This
group included 20 TD girls ranging in age from 7 to 14 years (mean:
10.17±2.28 years), 19 girls with 22q11.2DS, ranging in age from 7
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
to 14 years (mean: 10.75±1.87), and 18 girls with FXS, ranging in
age from 7 to 14 years (mean: 11.01±2.12 years). Separate groups
of children were scanned at the Thomas Jefferson University (TJU)
in Philadelphia, PA including 19 TD females ranging in age from 6 to
14 years (mean: 10.6±2.2) as well as 15 girls with TS from 7 to
13 years of age (mean: 10.56±2.67 years). All the TS girls were
nonmosaic (45,X karyotype) and none of them was being treated
with estrogen at the moment of the image acquisition. We only ana-
lyzed girls for 22q11.2DS and FXS since we wanted to exclude gender
as a confounding factor between TS (affected individuals are geneti-
cally determined to be females) and these diseases.
Scans from UCDMC were acquired on a 3.0-Tesla Siemens
MAGNETOM Trio scanner (Siemens Medical Solutions, Erlangen,
Germany) and consisted of a DTI sequence based on single-shot
echo planar imaging, with the following parameters: field of view
(FOV) = 22 cm × 22 cm, matrix size = 128 × 128, TE/TR = 99/
6700 ms, 40 axial slices, slice thickness 3 mm, and in-plane resolu-
tion of 1.72 × 1.72 mm2. We acquired 13 volumes per subject: one
with no diffusion sensitization and 12 diffusion-weighted images
(with a diffusion weighting set to b = 1000 s/mm2), with gradient
directions uniformly distributed on the unit hemisphere, for unbiased
angular sampling of diffusion. T1 weighted structural images were
also acquired for all of the 57 subjects of this group with the following
parameters: MPRAGE, TE = 4.82 ms, TR = 2170 ms IT = 1100 ms,
FOV = 25.6 cm × 25.6 cm, in-plane resolution of 256 × 256 with
192 slices, slice thickness of 1.00 mm.
The scans at TJU were collected using a Philips 3.0-Tesla whole
body clinical MRI system (Achieva, Philips Medical Systems, Best,
The Netherlands) equipped with a Quasar Dual high performance
gradient system. Subjects scanned included the individuals with
Turner syndrome with their corresponding typically developing con-
trols. The acquisition consisted of a single-shot echo planar imaging
and SENSE (sensitivity encoding) sequence to reduce scan time (TE/
TR 99/6700 ms, 15.2 cm FOV, 128 × 128). Each 3D volume consisted
of 40 axial slices (1.19 mm × 1.19 mm × 3 mm). The imaging proto-
col included a single non-diffusion-weighted (b = 0 s/mm2) refer-
ence image followed by 16 diffusion-weighted images with different
gradient directions, with a diffusion weighting of b = 1000 s/mm2,
for a total of 17 volumes. The gradient directions were distributed
uniformly on the unit hemisphere, for unbiased angular sampling of
diffusion. Additionally, each subject had four full acquisitions of the
17 volumes. T1 weighted structural images were acquired with an
MPRAGE protocol for the 34 subjects of this group. The parameters
were: TE = 3.2 ms, TR = 6 ms, FOV = 25.6 cm × 25.6 cm, in-plane
resolution = 256 ×256 with 160 slices, slice thickness of 1 mm.
Once scanning was finished, quality control of the data was done
to eliminate scans that did not adhere to our protocol or that suffered
from severe motion artifacts. Comparisons were made only between
groups imaged on the same scanner, to avoid confounding effects of
scanning site. The methods described below, i.e. DTI preprocessing,
anisotropy and diffusivity computation, template creation and regis-
tration, and statistical analysis, were done separately for each of the
two scanning groups specified in Tables 1 and 2. We analyzed five
groups of participants — 3 of them scanned at UCDMC: 19 girls with
22q11.2DS, 18 girls with FXS and 20 typically developing girls; and
2 of them scanned at TJU: 15 girls with TS and 19 typically developing
girls. The 22q11.2DS and the FXS groups were compared to the 20
typically developing girls at UCDMC. We refer to the typically devel-
oping group from this site as TD1 below. The TS group and the 19 typ-
ically developing girls from TJU were also compared in the statistical
analysis; below, we use TD2 to refer to the typically developing group
scanned at TJU.
Structural T1-images preprocessing
We used the robust, learning-based brain extraction system
(ROBEX) to skullstrip the brain T1-weighted images (Iglesias et al.,
2011). After automatically delineating the brains, we had an expert
delineator manually examine each of these to correct for any inconsis-
tencies, using BrainSuite's visual interface (http://users.loni.ucla.edu/
~shattuck/brainsuite). We then corrected the skull-stripped T1 images
for intensity non-uniformities using N3 (http://www.bic.mni.mcgill.
ca/ServicesSoftware/HomePage). Then, by using FSL's FLIRT tool
(Jenkinson and Smith, 2001), we linearly aligned, with 6 degrees of
atlas (NIHPD) (http://www.bic.mni.mcgill.ca/ServicesAtlases/NIHPD-
obj1) (Fonov et al., 2011) that was previously downsampled to a
110 × 110 × 110 resolution. We used the asymmetric, T1 version of
the NIHPD atlas, age range 7.5–13.5 years old (pre- to mid puberty),
based on 162 subjects.
We automatically removed non-brain regions from the b0
(non-diffusion weighted) images using FSL's BET (Smith, 2002)
(http://fsl.fmrib.ox.ac.uk/fsl/) then manually refined the brain ex-
traction. The 4 scans of each child's data acquired at TJU were
co-registered into a single volume for further processing. Raw diffu-
sion weighted images (DWIs) were corrected for eddy current dis-
tortions using FSL's “eddy_correct” method.
Next, the b0 image of each subject was linearly aligned to the
corresponding T1 structural image that was previously aligned to the
110 × 110 × 110 NIHPD atlas using 9 degrees of freedom so that both
the T1 and DWI images were in the same space, and in performing
Detailed description of the three groups of subjects scanned at the University of Califor-
nia Davis Medical Center (UCDMC). The scanning protocol of the diffusion MRI at this
site is also specified.
SiteUniversity of California Davis Medical Center (UCDMC),
Disease group 22q11.2 deletion
(10.75 ± 1.87)
3 T Siemens MAGNETOM Trio (Siemens Medical Solutions,
Single shot EPI
TE/TR = 99/6700 ms
1.72 × 1.75 × 3 mm3
220 × 220 mm2
1 b0/12 b = 1000 s/mm2
Fragile X syndrome
(11.01 ± 2.12)
(10.17 ± 2.28)
Description of the two groups of subjects scanned at the Thomas Jefferson University.
The scanning protocol of the diffusion MRI at this site is specified as well.
SiteThomas Jefferson University (TJU) in Philadelphia,
Turner syndrome (TS)
7–14 years (10.56 ± 2.67)
3 T Philips Achieva (Philips Medical Systems, Best,
Single Shot EPI
TE/TR = 99/6700 ms
1.19 × 1.19 × 3 mm3
152 × 152 mm2
1 b0/16 b = 1000 s/mm2
Typically developing (TD2)
7–14 years (10.6 ± 2.2)
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
this alignment, generated subject-specific transformation matrices.
These transformation matrices were then applied to the rest of the sub-
jects' volumes and were used to rotate the original scanner gradient
In order to adjust for any echo planar imaging (EPI)-induced sus-
ceptibility artifacts, we performed a 3D non-linear inverse-consistent
elastic intensity-based warping technique with a mutual information
cost function (Leow et al., 2005) of the b0image (previously linearly
aligned to the T1 structural image) to each subject's T1 image. The
transformation derived from this registration was then applied to all
of the DWI volumes.
Computing anisotropy and diffusivity
Once we rotated the original gradient vectors using the rotation
matrix from the linear transformation to the T1 structural image
and the DWI volumes that were corrected for EPI-induced distortions,
we computed diffusion tensors and derived scalar measures, using
FSL. Fractional anisotropy (FA) is the most common measure of
fiber integrity derived from DTI. It represents at a microstructural
level the integrity of myelinated neuronal fiber tracts as well as
fiber diameter and density. At a macrostructural level, it represents
the fiber-tract coherence. We compared FA values at each voxel be-
tween the 3 diagnostic groups and the typically developing control
children. FA was calculated from the tensor eigenvalues (λ1, λ2, λ3),
according to the standard formula:
bλ>¼λ1þ λ2þ λ3
ðÞ2 þ λ3−bλ>
where bλ> is the mean diffusivity. Mean diffusivity (MD) is an overall
summary measure of the diffusion in all directions, in a voxel or re-
gion. It can be interpreted as the mean displacement of the water
molecules within the voxel and also characterizes the overall pres-
ence of obstacles to diffusion.
Axial diffusivity (AD) represents the principal eigenvalue (λ1) of
the diffusion tensor — that is, the diffusion along the dominant direc-
tion within the voxel (i.e., along the dominant fiber). AD has been as-
sociated with axonal integrity (Assaf et al., 2008). Radial diffusivity
(RD) characterizes the other two eigenvalues of the tensor (λ2, λ3)
— measuring diffusivity along the two axes orthogonal to the princi-
pal one. RD is defined as:
It has been shown that RD is decreased when there is myelin loss
(Assaf et al., 2008; Tournier et al., 2011) and is also associated with
myelin content in the white matter, although lower RD values do
not necessarily indicate demyelination. We hypothesized that we
might find higher MD, AD, and RD in atypically developing girls com-
pared to TD girls.
Template creation and registration
As we analyzed data from two different scanning locations, we
kept the two data sets (acquired on the different scanners) separate
for processing and analysis. In particular, starting from this point of
the analysis; a minimal deformation template (MDT) is created for
each group and all subjects are registered to each group's MDT.
First we created an FA-based minimal deformation template
anatomical features for a group of participants. From the set of scans
acquired at UCDMC (22q11.2DS, FXS and TD1), we took the FA maps of
all the typically developing girls (TD1 group) to create the MDT, which
we will call MDT-1. In the same way, all the FA maps from the typically
developing girls scanned at TJU (TD2) were used to create their corre-
sponding MDT, which we will call MDT-2.
To compute both MDTs (MDT-1 and MDT-2), the first step was to
images affinely (Woods et al., 1998). For group-wise registration,
Nx(N-1) transformations were computed and composed using the AIR
‘reconcile’ tool to bring all images into a common 12-parameter space.
The second step was to create a non-linear average. For this, all individ-
(mean of the affinely aligned images) using a non-linear inverse-
consistent fluid intensity-based registration algorithm (Leow et al.,
2007). The non-linear average template was computed as a voxel-wise
average of the intensities of the FA maps that had been non-linearly
registered to the affine average template. Finally, we created the MDT
by applying the inverse of the average deformation field to the mean of
nonlinearly registered images.
FA maps of all children with each of the three genetic disorders
and all typical developing controls were registered to their corre-
sponding FA-based MDT using a 3D non-linear inverse-consistent
elastic intensity-based warping technique with a mutual information
cost function (Leow et al., 2005). Thus, 22q11.2DS, FXS and TD1
groups were registered to MDT-1, and TS and TD2 groups were regis-
tered to MDT-2.
To better align white matter regions of interest, the FA-MDTs
(MDT-1, and MDT-2) and all whole-brain registered FA maps from
each individual subject were thresholded at FA > 0.2 to exclude
contributions from non-white matter. Thresholded FA maps were
re-registered to the corresponding thresholded MDT. Additionally,
we applied the deformation fields computed from the first round regis-
tration (whole FA volume to whole MDT) and last round registration
(from the thresholded-FA to the thresholded MDT) to each subject's
previously computed mean, axial and radial diffusivity maps.
Tissue specific smoothing compensation
We implemented the tissue-specific-smoothing compensation
method (T-SPOON) as described in Lee et al. (2009). Once each
subject's FA map was non-linearly registered to the minimal deforma-
tion template, we thresholded these FA maps to extract the white
matter “skeletons” by taking only the voxels with values higher than
0.2, and then binarize them. These skeletons are considered to repre-
sent the white matter (WM), thus we will refer to these as WM
masks. We also kept a non-binarized version of the thresholded FA.
The second step consists of spatial smoothing of both the thresholded
FA maps and to the WM masks with a Gaussian smoothing kernel of
2 × 2 × 2 mm. Finally, we divided the smoothed thresholded FA
masks by the smoothed white matter masks. This method was pro-
posed by Lee et al. to compensate for the spatial smoothing effects
by dividing the smoothed FA maps and the smoothed tissue specific
linear regression model was used to adjust for any confounding effects
of age in the different diagnostic groups (p b 0.05 significance level). In
this regression analysis the corresponding FA, AD, RD and MD maps for
each child were considered as the outcome variable, while diagnostic
group, age and total brain volume were used as predictors. To correct
for the increased risk of Type I (false positive) errors due to the thou-
sands of voxel-wise association tests, we used the false discovery rate
(FDR) method (Benjamini and Hochberg, 1995) to correct for multiple
comparisons. All maps are shown thresholded at the appropriate FDR
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
critical p-value, where such a threshold exists, to show only regions of
significance with an expected 5% false posi tive rate (q = 0.05). Impor-
fect on the brain, not an effect size at a particular location. The “critical”
p-values used in FDR analysis are not identical to the usual meaning for
bychance. Instead, the criticalp-values showthe highestthreshold that
can be applied to the maps, if there is one, while still being able to say
that only 5% of the voxels shown are expected to be false positives.
There is no notion of effect size directly corresponding to this p-value,
although there is an effect size at each location in the brain. Essentially
the critical p-value comes from a cumulative distribution function (his-
effects or a smaller number of large effects would be counted as signif-
icant. As such an impression of the effect size in the map could be in-
ferred from the p-value CDF plot, as we have done in some of our
prior papers (Chou et al., 2008).
Quantitative assessment of overlap
We used the Dice coefficient to measure the overlap of the signifi-
cant areas in the p-maps obtained after the statistical analysis. In
doing so, we have to bear in mind that the extent of the significant re-
gion depends on the sample size, and with an expanded sample group
regarding overlap may be conservative estimates of the regions impli-
cated across multiple disorders. We compared each p-map with signif-
icantvoxels in it to each ofthe other p-maps. Theresults of this analysis
are detailed in the Results and Discussion sections below. To ensure a
fair comparison between the maps calculated in the both MDTs
(MDT-1 and MDT-2), we non-linearly registered MDT-2 to MDT-1.
We then applied this transformation to each of the maps in the
MDT-2 space with a nearest neighbor interpolation.
Given two sets, the Dice coefficient measures the agreement be-
tween them. If A and B are the two sets, the Dice coefficient is given by:
ð Þ ¼2 A∩B
j j þ B j j:
A value of 0 indicates no overlap; a value of 1 indicates perfect
Tractography and anatomical location of the areas of significance
We performed whole-brain tractography with a global probabilis-
tic approach based on the voting procedure provided by the Hough
transform (Aganj et al., 2011). This algorithm tests candidate 3D
curves in the volume, assigning a score to each of them. It then
returns the curves with the highest scores as the potential anatomical
connections. The score is accordingly derived from the DWI data (for
details, please see Aganj et al., 2011). We calculated a total of 10,000
fibers for each subject.
ysis were transformed back to the each subject's space, where the
tractography was calculated. By doing this, we were able to overlay the
significant areas in the space of the subject's tractography and were able
to set them as ROIs to extract and identify the tracts crossing those ROIs.
Girls with 22q11.2 had abnormally lower FA values, relative to
corresponding typically developing controls (TD1), in the following
regions bilaterally: superior temporal gyri and in the superior corona
radiata (p b 0.00026), (Fig. 1). This pattern of significance was repli-
cated in the AD maps, where 22q11.2 girls had lower AD and MD in
the same regions mentioned above in addition to the middle temporal
Fig. 1. Superior row: Hot colors depicted in the color bar represent the significant areas with higher FA in typically developing girls than in girls with 22q11.2 (TD > 22q11.2DS).
Middle row: Green colors depicted in the color bar represent the significant areas with higher RD in girls with 22q11.2 (22q11.2DS > TD) than in typically developing girls. Bottom
row: Blue colors depicted in the color bar represent the significant areas with higher MD in girls with 22q11.2 (22q11.2DS > TD) than in typically developing girls 0.00026, 0.0016
and 0.0011 represent the critical p-values for each map as determined by the false discovery rate procedure (Benjamini and Hochberg, 1995).
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
gyri and in the right posterior internal capsule (p b 0.0010 and
p b 0.0011, respectively) (Fig. 1).
Bycontrast, girlswith22q11.2DS hadhigher AD, RDand MDvalues in
regions including clusters in the corpus callosum (genu and splenium),
the anterior limb of the internal capsules, external capsules, inferior
and posterior parts of the thalami (p b 0.0010, 0.0016 and p b 0.0011)
(Fig. 1). In addition to this, RD and MD maps also showed higher diffusiv-
ity values in 22q11.2DS where TD1 children have higher FA and AD
valuesas describedabove(superiorcorona radiata andsuperiorandinfe-
rior temporal lobes).
and MD areas of significance (p b 0.0036, p b 0.0096 and p b 0.0086
respectively). These maps showed higher diffusivity values, areas in-
terior corona radiata, anterior limb of the internal capsules, the external
capsules, extended areas of the temporal lobes, occipital lobes, inferior
portions of the thalami and central portions of the right thalamus
(Fig. 2). No p-values survived the FDR correction when analyzing the
FA maps of FXS vs. TD1 girls.
Relative to TD2 controls, girls with TS had lower FA in clusters in the
superior corona radiata, left superior frontal gyrus, corpus callosum
(splenium and genu), the anterior and posterior limbs of the internal
capsules, cerebral peduncles, and the occipital and temporal lobes
(p b 0.0046) (Fig. 3). 22q11.2 girls showed significant higher diffusivity
less involvement of the internal capsules and the cerebral peduncles
(p b 0.0025 p b 0.00068) (Fig. 3). AD maps of TS vs. TD2 girls were
not significantly different.
When analyzing the dice coefficients (DC) resulting from comparing
the maps (see Table 3) with statistically significant differences, we
found the highest score when comparing the maps where FXS girls had
higher AD and RD values than the TD1 controls (DC: 0.8569) (Fig. 2),
and when comparing the maps where AD, RD and MD were higher in
22q11.2DS than in TD1 (also see Table 3), where all DC were above 0.6
(better agreement the closer to 1). All of the other comparisons made
were below 0.2, where it is worth mentioning the across-syndrome sim-
higher FA in TD2 over TS and the areas of higher AD in FXS than in TD1
(DC: 0.1524). Additionally, the areas where TS had higher RD than TD2
controls overlapped with areas where AD was higher in FXS than in TD1.
There were interesting findings when using the significant clus-
ters from the maps as ROIs to parcellate the tractography results.
We used tractographies from the typically developing groups (TD1
and TD2) to better identify the tracts involved. For clarity, we also dis-
tinguished the three major types of fibers: long association fibers,
projection fibers and commissural fibers (corpus callosum).
Long association fibers
We found the superior longitudinal fasciculus (SLF) to be involved
in all three syndromes studied here, as the significant areas in parietal
and temporal lobes where FA was lower in 22q11.2DS and TS com-
pared to TD intersected it (Figs. 4, 7). The SLF also passed through
areas where the AD, RD, and MD were higher in FXS (Fig. 6) and
where RD was higher in 22q11.2DS (Fig. 8).
The inferior longitudinal fasciculus (ILF) also intersected areas in
the temporal and occipital lobes where FA was lower in 22q11.2DS
and TS (Figs. 4 and 7), especially in the right hemisphere. The ILF
could also be identified crossing regions where the AD, RD and MD
were higher in 22q11.2DS and FXS (Figs. 6 and 8), as well as where
RD and MD were higher in TS. In all these maps, tractography showed
that the ILF was more clearly involved on the right hemisphere.
The inferior fronto-occipital fasciculus (IFO) was found to be
intersecting significant clusters in the right hemisphere where FA was
lower in TS (Fig. 4) and where AD, RD and MD were higher in FXS
girls (Fig. 6). Also some IFO fibers were identified intersecting small
clusters with lower RD and MD in TS girls (Fig. 4). The intersecting
Fig. 2. Superior row: Yellow colors depicted in the color bar represent the significant areas with higher AD in girls with Fragile-X syndrome (FXS > TD) than in typically developing
children. Middle row: Green colors depicted in the color bar represent the significant areas with higher RD in girls with Fragile-X syndrome (FXS > TD) than in typically developing
girls. Bottom row: Blue colors depicted in the color bar represent the significant areas with higher MD in girls with Fragile-X syndrome (FXS > TD) than in typically developing girls.
0.0036, 0.0096, 0.0086 represent the critical p-values for each map as determined by the false discovery rate procedure (Benjamini and Hochberg, 1995).
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
clusters were mainly seen in thetemporaland occipitallobes and inthe
anterior limbs of the external capsules.
Projection fibers and corpus callosum
In general, the significant scattered clusters that were found within
the central and posterior corona radiata were intersected by a myriad of
projection fibers (cortico-thalamic, cortico-spinal, cortico-cerebellar).
We noticed a pattern where FXS girls had higher AD, RD, and MD than
TD1 controls (Fig. 6), where TS had lower FA and higher MD and RD
than TD2 (Fig. 5) and where 22q11.2DS girls had higher AD and RD
than TD1 (Fig. 7). These areas were also intersected by some fibers
from the body of the corpus callosum.
statistically significant clusters of the postero-inferior sections of the
thalami to the occipital lobe. This pattern was seen where FXS had
higher AD, RD, and MD than TD1 and in areas where 22q11.2DS had
higher AD and RD than TD2 controls.
As mentioned above, the anterior limb of the internal capsule had
higher AD, RD and MD in FXS and lower FA and higher RD and MD in
TS. This area was crossed by projection fibers to the frontal lobe (Figs. 4,
6 and 8).
The current study aimed to test the hypothesis that, given the phe-
notypic overlap in three different genetic syndromes (22q11.2DS, FXS,
TS), there would be broad white matter structural differences and
similarities determined across these groups compared to typically de-
We found patterns of statistical differences of typical developing
children having significantly higher FA values than atypically develop-
ing ones, which were also replicated in the diffusivity maps where
non-typical developing children had higher diffusivity values than con-
trols. This pattern of findings was evident when measuring the overlap
of these maps (Table 3). FA is said to account for white matter fiber in-
tegrity and fascicle coherence. In general, we expect higher diffusivities
in areas where FA is diminished (Thomason and Thompson, 2011). In
one case, our results for FXS compared to TD did not show significant
ity measures within the white matter.
Clearly, differences in FA and diffusivity measures between typical
and non-typical children do not necessarily have the same causes in
each of the three disorders studied here. Even so, pathological evidence
supports some similarities among the three conditions. In general, FA,
AD, RD and MD depend mainly on structural variables such as axonal
packing, myelination, membrane permeability to water, internal axonal
structure, and intra-axonal space (Assaf et al., 2008). For example patho-
logical findings in TS point out the prevalence of cortical organization
In this table we show the highest scores after calculating the dice coefficient when
comparing the maps where FA, AD, RD, and MD were significantly higher in 22q11.2
deletion syndrome (22Q), Fragile X syndrome (FXS), and Turner syndrome (TS) over
their corresponding typical developing group (TD1 and TD2) or vice versa. In bold:
The comparisons across syndromes that resulted with a relatively higher score.
Fig. 3. Superior row: Hot colors depicted in the color bar represent the significant areas with higher AD in girls with Turner syndrome (TS > TD) than in typically developing chil-
dren. Middle row: Green colors depicted in the color bar represent the significant areas with higher RD in girls with Turner syndrome (TS > TD) than in typically developing girls.
Bottom row: Blue colors depicted in the color bar represent the significant areas with higher MD in girls with Turner (TS > TD) than in typically developing girls 0.0046, 0.0025 and
0.00068 represent the critical p-values in each map as determined by the false discovery rate procedure (Benjamini and Hochberg, 1995).
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
abnormalities such as cortical dysplasia, pachygyria and polymicrogyria,
ing in the white matter (Della Giustina et al., 1985; Terao et al., 1996;
Tombini et al., 2003). Myelin has been reported as normal in TS though
(Palo and Savolainen, 1973; Sy et al., 2010). In 22q11.2DS, there have
been radiological and pathological reports of cortical dysgenesis and ab-
normal gyrification patterns in some patients (Kiehl et al., 2009; Robin
et al., 2006; Schaer et al., 2006; Sztriha et al., 2004). There is also experi-
mental evidence of abnormal neuronal proliferation and migration in
early stages of development resulting in an altered cortical connectivity
pattern (Meechan et al., 2009). Focal myelin damage in 22q11.2DS has
been related to post-developmental vascular events and microvascular
anomalies caused by dysregulated angiogenesis. Experimental evidence
in FXS has revealed abnormal dendritic spine lengths and shapes on
neocortical pyramidal cells, but no other major neuropathological abnor-
malities have been reported (Irwin et al., 2000; Rudelli et al., 1985;
Wisniewski et al., 1991). These findings suggest the failure of normal
dendritic spine maturation, and/or pruning, but not neurogenesis or mi-
gration problems. Thus, in the three diseases, the neuropathogenesis is
ment. Abnormal neuronal connectivity, either locally or throughout the
entire brain, may ultimately affect axonal integrity and adequate packing
Our findings show lower FA and AD plus higher MD and RD values in
22q11.2DS children especially in posterior regions of the brain (parietal,
occipital and posterior temporal lobes), mainly in areas intersecting the
inferior longitudinal fasciculus (ILF), superior longitudinal fasciculus
Fig. 4. a),b)ande)Tractographyshows fibersfromlongassociationfasciculi(SLF,ILF andIFO)afterbeingseededwithclusters oflowerFAinTurnersyndrome.c),d)andf)Tractography
after being seeded with clusters of higher RD in Turner syndrome. Subject: 13.16 year old typically developing girl.
Fig. 5. The four figures show extensive projection fibers (corticospinal and corticothalamic). Seeds on the left side represent areas of lower FA in Turner syndrome and on the right
side higher RD in Turner syndrome. Subject: 13.16 year old typically developing girl.
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
(SLF) (Figs. 1, 7 and 8). Prior studies also reported lower anisotropy in
areas corresponding to the SLF (Barnea-Goraly et al., 2003b; Simon et
al., 2008b). Simon et al. (2008b) also found significant lower anisotropy
in the external capsules, which they considered to be part of the inferior
longitudinal fasciculus (IFO) and in a similar fashion our analysis also re-
vealed increased RD and MD in the same location. Other regions of inter-
est in participants with 22q11.2DS included significantly lower FA values
and higher AD, RD and MD in bilateral clusters in the inferior and poste-
dle of fibers projecting to the occipital cortex is originating. When
analyzing FA values in the parietal white matter, Barnea-Goraly et al.
found a positive correlation between FA and arithmetic scores achieved
by subjects with 22q11.2DS in the left parietal lobe. In addition to this,
tions and a coarser in-plane matrix, Simon et al. reported differences in
the parietal lobes of 22q11.2DS children (girls and boys together)
consisting of significant direct correlation of higher FA and higher AD
to poorer performance on a visuospatial attention task that would be
expected to involve contiguous cortex. According to the authors, this
suggests a lower degree of connectivity of the parietal cortex to sur-
rounding cortex (Simon et al., 2005b, Simon et al., 2008b). Although
we were not able to correlate our neuroimaging measures with visuo-
spatial and arithmetic performance, we did find higher AD (as well as
RD) but not FA in the parietal white matter of 22q11.2DS girls. This
may be due to differences in the acquisition parameters for the DTI
data, although its cause is not entirely clear.
Our findings of22q11.2DS whitematter abnormalities are relatedto
previous volumetric morphometry studies. Besides having lower brain
volumes (by 8–11%) (Eliez et al., 2000; Kates et al., 2001; Simon et al.,
2005b), in both gray and white matter, reductions are more localized
towards the cerebellum and parietal, temporal and occipital lobes,
which are in the posterior regions of the brain. ROI-based studies of
structural MRI found reduced white matter volumes in areas in the pa-
rietal, occipital and temporal regions (van Amelsvoort et al., 2001). Re-
associated with the increase psychosis risk in this population (Chow et
al., 2011; Gothelf et al., 2007a,b). Decreased volumes of posterior tha-
lamic nuclei in children with 22q11.2DS (Bish et al., 2004) have been
reported as well.
We found areas of significant differences of FA and diffusivity mea-
sures at the edges of the corpus callosum in children with 22q11.2DS,
more specifically at the genu and the splenium (Fig. 1). There is consid-
erableevidenceshowingvolumetric differences of the corpus callosum,
the lateral ventricles and thalami between typical developing children
found a generally increased volume in individuals with 22q11.2DS,
particularly in the isthmus (Shashi et al., 2004) and rostrum
(Machado et al., 2007) but also in the splenium (van Amelsvoort et al.,
2001). Machado et al. also found a significant correlation between ven-
tricular enlargement and corpus callosum curvature along the rostral
body, posterior midbody, isthmus and splenium of the 22q11.2DS indi-
viduals (Machado et al., 2007). In addition to this, Simon et al. reported
higher FA values within the splenium in typically developing children
(Barnea-Goraly et al., 2003b; Simon et al., 2005b), which is consistent
with our findings, and they concluded after a voxel-based morphome-
tryanalysiscomparingTD and22q11.2DSthatthesefindings inthecor-
this structure due to enlarged lateral ventricles in the 22q11.2DS popu-
lation. Additionally, there is evidence of decreased volume in the thala-
mus — particularly in its posterior nuclei (Bish et al., 2004), which may
also add to the structural differences surrounding the lateral ventricles.
Further analysis of this area is required to identifyareas of thesplenium
with significant FA differences that are not merely due to volumetric
differences between populations.
Prior fMRI studies have suggested disrupted fronto-parietal connec-
functional abnormalities within this network, for example abnormally
increased activations in prefrontal, precentral gyrus and parietal cortex
when solving arithmetical tasks. Additionally, there was increased
activity in supramarginal gyrus when solving numerical computation
tasks of greater difficulty (Eliez et al., 2001b) as opposed to a steady
normal frontal and parietal activation in control subjects. The authors
argued this to be related to the need to recruit more resources to per-
form the tasks. In another study, 22q11.2DS children and adolescents
were compared to TD controls during a spatial working memory task
(Azuma et al., 2009), researchers found a greater activation in the TD
group than in children with 22q11DS in the parietal and occipital
regions with no significant difference in activation of frontal regions
(dorsolateral prefrontal cortex).
Although we did not find significant differences in FA between FXS
girls and controls, our study did find an increase in all three diffusivity
measures (AD, RD and MD) in FXS girls in a very similar fashion (Fig. 2
Fig. 6. a) c) and e) show dissected long association fibers (SLF, ILF and IFO) after seeding the tractography with areas of high RD and MD in Fragile-X girls. b) and d) show the same
tractography but the projection fibers are dissected here. Notice the optic radiation fibers in posterior section of the brain. Subject: 13 years old typically developing girl.
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
and Table3). Thesesignificantclusters wereconsistent withprevious DTI
findings of lower FA in the corona radiata and centrum semiovale
(Barnea-Goraly et al., 2003a) and our findings do also extend to the tem-
poral and occipital lobes, internal and external capsules, corpus callosum
and thalami. These areas involve SLF, ILF, and IFO fibers. Previous volu-
metric studies have found increased white matter in temporal and parie-
tal lobes of FXS children (Lee et al., 2007), areas where these tracts travel
through. Also some fMRI studies that have investigated arithmetic abili-
ties and working memory function in FXS children have pointed out a
dysfunction in the fronto-parietal networks. Rivera et al. (2002) investi-
ing the parietal lobe. That study showed that individuals with FXS were
unable to modulate activation in prefrontal and parietal cortex, as
evidenced by an inability to recruit new neural resources in response to
increasing cognitive load. Studies evaluating attention in subjects with
heavily modulated by the FMRP levels and, in general, FXS shows im-
controls (Menon et al., 2004; Hoeft et al., 2007b).
age-matched FXS subjects (mean age 14 years) with normal controls.
There, we found increased volume in periventricular white matter,
Fig. 7. a) and c) show right ILF and SLF and left SLF in a typically developing girl of 13 years old. Seeds correspond to regions of lower FA in 22q11.2DS. b) and d) show same subject's
tractography seeded with clusters of lower AD in 22q11.2DS (TD > 22q11.2DS). Notice the ILF in both hemispheres.
Fig. 8. a) b) and c) show a tractography of a 13 year old typical developing girl seeded with areas of higher RD and MD in 22q11.2DS girls. Notice the SLF and ILF in both hemispheres
and the IFO in the right hemisphere.
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
is due to alteration in the corpus callosum fiber structure. There is a
paucity of DTI studies of FXS, so our findings in the corpus callosum
need further validation and replication.
TS is a heterogeneous disorder that includes women with different
genotypes such as complete monosomy X (45,X karyotype), isochro-
mosomes (46,X,i[Xq]), rings (46,X,r[X]), deletions (46,X,del[Xp] or
46,X,del[Xq]), and mosaicism (Held et al., 1992). Additionally, many
women with TS receive estrogen replacement therapy due to ovarian
failure, which adds another variable that may influence the brain's
developmental trajectory. Three studies have investigated white mat-
ter structure with DTI in TS. Each of these studies has examined dif-
ferent populations with TS in terms of their genotype and onset of
hormonal replacement therapy at the time of data acquisition. The
first study using DTI in TS included a mixture of mosaic and
nonmosaic TS young adult women, and all but one of whom were
being treated with estrogens (Molko et al., 2004). They found higher
FA values in TS women in both superior temporal sulci and the right
centrum semiovale and right external capsule, and no there were sig-
nificant regions with lower FA. Nonetheless, MD was higher in TS
women in the right fusiform gyrus and in the occipitotemporal re-
gion, bilaterally. Subsequently, Holzapfel et al. (2006) examined a
sample of only nonmosaic children and young adults–all but two
were receiving hormonal treatment. Their results showed lower FA
in TS in the deep left frontoparietal white matter (SLF), in the right
prefrontal cortex close to the caudate, and in the internal capsules, bi-
laterally. These locations were also significant in the same study for
reduced white matter density, as seen with VBM, and TBM. Addition-
ally, the TS group exhibited higher FA values in the right precentral
gyrus and in the temporal gyri.
The latest of these studies (Yagamata et al., 2011) investigated the
white matter of a very similar cohort to ours, namely only 45, X kar-
yotype, nonmosaic girls, none of them with estrogen treatment. Using
TBSS (Smith et al., 2006) and an atlasbased approach they found
lower FA in these girls in regions of the white matter very similar to
ours: the SLF bilaterally (more prominent on the left side), the ILF
(more prominent on the left) and the IFO, although we found the
right ILF to have more areas with significant differences. They also
found lower FA in extended areas of the corpus callosum, including
the splenium, body, genu and tapetum, as well as in the corticospinal
tract, anterior and posterior thalamic radiations. We consistently
found lower FA in TS in these areas as well. They report finding
areas of lower FA in the left external capsule, where we did not find
any significant difference between TS and control girls.
Our study has the limitation of not having the cognitive tests avail-
able of the TS girls, which makes the results more difficult to interpret
in terms of their relevance for cognition and behavior, and also im-
plies that we cannot make strong generalizations to other cohorts
who may differ cognitively or behaviorally. Even so, our cohort of TS
girls is genetically homogeneous as well as clinically in terms of the
estrogen replacement therapy and many of our findings were highly
consistent with findings reported in prior studies. Some white matter
regions have been consistently found to be altered in all the studies –
in particular, the occipitotemporal region, which contains fibers of the
ILF and IFO. The SLF, especially on the left side, and the internal cap-
sules bilaterally – which are crossed by a myriad of projection fibers
have been reported to be altered in two of the studies mentioned
above that analyzed only nonmosaic girls, as well as in ours. This sug-
gests that there may be white matter structural abnormalities in TS
that are not affected by hormonal replacement therapy and that de-
pend more on the specific genetic load given by the sex chromosomes
in earlier stages of brain development. SLF, ILF and IFO are association
tracts that connect the frontal lobe to the more posterior regions of
the parietal, occipital and temporal lobes. They are thought to play
an important role in the transport of information between these
areas, particularly in visuospatial tasks, attention and numerical rea-
soning. Behavioral studies support the view that visuospatial and per-
ceptual deficits in TS (mosaic and nonmosaic) persist from childhood
into adulthood and do not improve with estrogen replacement,
whereas motor and certain memory functions respond to this kind
of therapy (Ross et al. 2000; Ross et al., 2002). In addition to this,
fMRI studies on nonmosaic and mosaic adult women with TS have
shown impaired frontoparietal activation during numerical tasks
and visuospatial paradigms while being on estrogen therapy (Molko
et al., 2003; Hart et al. 2006). In the future, efforts should be directed
towards imaging studies with more homogeneous populations with
TS, while taking into account important cognitive and therapeutic as-
pects in these women.
We did find common abnormalities shared by the 3 syndromes.
fusivity of the SLF in the three syndromes. In 22q11.2DS this region
showed lower FA and higher AD, RD and MD, most prominently on
the left side (Figs. 7 and 8). The SLF showed higher AD, RD and MD in
FXS (Fig. 6) and in TS the same region showed higher RD and MD and
lower FAin both hemispheres(Fig.4).Prior DTIstudies on the anatomy
its counterpart in non-human primates. It has been divided into three
three sections: a fronto-parietal component that runs lateral to the pro-
jection fibers of the corona radiata (SLF I), a temporo-parietal part that
runs around the Sylvian fissure (SLFII) and a more lateral one that pen-
2008). Interestingly, we found that the altered section in TS is the
temporo-parietal section (SLF II), whereas for 22q11.2DS and for FXS
the fronto-parietal section is clearly involved (SLF I). The SLF is part of
brain networks processing language, spatial working memory and
numerical tasks on the left side (van Eimeren et al., 2010; Vestergaard
et al., 2010), to visuospatial attention in the right side (de Schotten
et al., 2011). Interestingly, numerical ability tends to show positive
correlations with FA values in the left SLF (van Eimeren et al., 2010).
As we noted before, for all three syndromes there is substantial
evidence that fronto-parietal connectivity is disrupted, and this may
contribute to poorer performance on tasks involving attention, visuo-
spatial working memory and arithmetic. Despite the age range of our
cohort, and that the SLF anisotropy and diffusivity signals are not fully
mature until the mid-teen years (Lebel et al., 2008, Qiu et al., 2008),
we were still able to see differences. In terms of known functions of
the SLF, we could not find an explanation to account for different sec-
tions of this tract being involved TS vs. FXS/22q11.2DS.
We found the ILF to be altered in the three syndromes, but where-
as the differences could be seen bilaterally in the TS and FXS subjects
(lower FA and higher diffusivity values), the findings were unilateral
in 22q11.2DS children. FA was clearly lower and MD and RD higher in
the right ILF in these subjects and TD controls had higher AD values
bilaterally. Very similar effects were also found for the IFO, which
intersected significant clusters in the external capsules and frontal
lobes in TS and FXS, but not in 22q11.2DS where its fibers showed dif-
ferences mainly on the right side. The ILF connects the occipital and
temporal lobes and is involved in visual perception (Ffytche, 2008)
and face processing (Fox et al., 2008). Different functional studies
have found impairments in the occipito-temporal network in facial
recognition and verbal processing in FXS and 22q11.2DS (van
Amelsvoort et al., 2006; Holsen et al., 2008). The role of the IFO in
these syndromes' cognition and executive functions is less clear, but
there is some evidence linking this tract to visual processing and at-
tention (Fox et al., 2008; Doricchi et al., 2008).
The role of the different projection fibers connecting the thalamus
and basal ganglia to the frontal lobes through the internal capsules
and the optic radiations towards the occipital lobe is less clear for
visuospatial attention and executive functions, as well as numerical
abilities. Further evidence of the role of these fibers in visuospatial
attention and executive function is still to be found.
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
In summary, we investigated the white matter integrity profile
across females with three defined genetic disorders. Many regions
with significant differences across the three diseases do not completely
normalities were evident in 22q11.2DS and TS, where SLF and ILF had
lower anisotropy and diffusivity and FXS had higher diffusivities. Addi-
tionally we found thalamic-frontal and thalamic-occipital fibers to be
altered in 22q11.2DS, FXS, and TS.
Conflict of interest
This work was supported by grants NIH HD46159 and NIH HD42974
(to T.J.S.). Additional support was provided by R01 grants EB008432,
EB008281, and EB007813 (to P.T.) and by the National Library of
Medicine (T15 LM07356; to N.J.). The funding source had no role in the
study design, collection, analysis and interpretation of data, writing of
the report, or in the decision to submit the article for publication.
Aganj, I., Lenglet, C., Jahanshad, N., Yacoub, E., Harel, N., Thompson, P.M., Sapiro, G.,
2011. A Hough transform global probabilistic approach to multiple-subject diffusion
MRI tractography. Med. Image Anal. 15, 414–425.
Arnold, A.P., Gorski, R.A., 1984. Gonadal steroid induction of structural sex differences
in the central nervous system. Annu. Rev. Neurosci. 7, 413–442.
Assaf, Y., Blumenfeld-Katzir, T., Yovel, Y., Basser, P.J., 2008. Axcaliber: a method for
measuring axon diameter distribution from diffusion MRI. Magn. Reson. Med. 59,
Azuma, R., Daly, E., Campbell, L., Stevens, A., Deeley, Q., Giampietro, V., Brammer, M.,
Glaser, B., Ambery, F., Morris, R., Williams, S., Owen, M., Murphy, D., Murphy, K.,
2009. Visuospatial working memory in children and adolescents with 22q11.2
deletion syndrome; an fMRI study. J. Neurodev. Disord. 1, 46–60.
Barnea-Goraly, N., Eliez, S., Hedeus, M., Menon, V., White, C.D., Moseley, M., Reiss, A.L.,
2003a. White matter tract alterations in Fragile X syndrome: preliminary evidence
from diffusion tensor imaging. Am. J. Med. Genet. B Neuropsychiatr. Genet. 118B,
Barnea-Goraly, N., Menon, V., Krasnow, B., Ko, A., Reiss, A., Eliez, S., 2003b. Investigation
of white matter structure in velocardiofacial syndrome: a diffusion tensor imaging
study. Am. J. Psychiatry 160, 1863–1869.
Beaton, E.A., Qin, Y., Nguyen, V., Stoddard, J., Pinter, J.D., Simon, T.J., 2010. Increased in-
cidence and size of cavum septum pellicidum in children with chromosome
22q11.2 deletion syndrome. Psychiatry Res. Neuroimaging 181, 108113.
Bearden, C.E., Woodin, M.F., Wang, P.P., Moss, E., McDonald-McGinn, D., Zackai, E.,
Emannuel, B., Cannon, T.D., 2001. The neurocognitive phenotype of the 22q11.2 dele-
tion syndrome: selective deficit in visual–spatial memory. J. Clin. Exp. Neuropsychol.
Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and
powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Methodol. 57, 289–300.
with chromosome 22q11.2 deletion syndrome. Neuroreport 15, 1413–1415.
Bishop, D.V., Canning, E., Elgar, K., Morris, E., Jacobs, P.A., Skuse, D.H., 2000. Distinctive
patterns of memory function in subgroups of females with Turner syndrome:
evidence for imprinted loci on the X-chromosome affecting neurodevelopment.
Neuropsychologia 38, 712–721.
Bittel, D.C., Yu, S., Newkirk, H., Kibiryeva,N.,Holt, A., Butler, M.G., Cooley, L.D., 2009. Refin-
ing the 22q11.2 deletion breakpoints in DiGeorge syndrome by aCGH. Cytogenet.
Genome Res. 124, 113–120.
Braddick, O., Atkinson, J., Wattam-Bell, J., 2003. Normal and anomalous development of
visual motion processing: motion coherence and ‘dorsal-stream vulnerability’.
Neuropsychologia 41, 1769–1784.
Brown, W.E., Kesler, S.R., Eliez, S., Warsofsky, I.S., Haberecht, M., Reiss, A.L., 2004. A
volumetric study of parietal lobe subregions in Turner syndrome 46 (9), 607–609.
Brown, W.E., Kesler, S.R., Eliez, S., Warsofsky, I.S., Haberecht, M., Patwardhan, A., Ross, J.L.,
Neely, J.L., Zeng, S.M., Yankowitz, J., Reiss, A.L., 2002. Brain development in Turner
syndrome: a magnetic resonance imaging study. Psychiatry Res. Neuroimaging. 116
Bruandeta, M., Molkoa, N., Cohena, L., Dehaenea, S., 2004. A cognitive characterization of
dyscalculia in Turner syndrome. Neuropsychol. 42 (3), 288–298.
Burn, J., Goodship, J., 1996. Developmental genetics of the heart. Curr. Opin. Genet. Dev.
Catani, M., Thiebaut de Schotten, M., 2008. A diffusion tensor imaging tractography
atlas for virtual in vivo dissections. Cortex 44, 1105–1132.
Chou, Y.-Y., Lepore, N., de Zubicaray, G.I., Carmichael, O.T., Becker, J.T., Toga, A.W.,
Thompson, P.M., 2008. Automated ventricular mapping with multi-atlas fluid
image alignment reveals genetic effects in Alzheimer's disease. Neuroimage 40,
Chow, E.W.C., Ho, A., Wei, C., Voormolen, E.H.J., Crawley, A.P., Bassett, A.S., 2011. Asso-
ciation of schizophrenia in 22q11. 2 deletion syndrome and gray matter volumetric
deficits in the superior temporal gyrus. Am. J. Psychiatry 168, 522.
Cornish, K.M., Munir, F., Cross, G., 1999. Spatial cognition in males with Fragile-X
syndrome: evidence for a neuropsychological phenotype. Cortex 35, 263–271.
Crawford, D.C., Meadows, K.L., Newman, J.L., Taft, L.F., Pettay, D.L., Gold, L.B., Hersey, S.J.,
Hinkle, E.F., Stanfield, M.L., Holmgreen, P., Yeargin-Allsopp, M., Boyle, C., Sherman, S.L.,
1999. Prevalence and phenotype consequence of FRAXA and FRAXE alleles in a large,
ethnically diverse, special education-needs population. Am. J. Hum. Genet. 64, 495–507.
de Schotten, M.T., Dell'Acqua, F., Forkel, S.J., Simmons, A., Vergani, F., Murphy, D.G.M.,
Catani, M., 2011. A lateralized brain network for visuospatial attention. Nat. Neurosci.
De Smedt, B., Devriendt, K., Fryns, J.P., Vogels, A., Gewillig, M., Swillen, A., 2007. Intel-
lectual abilities in a large sample of children with velo-cardio-facial syndrome:
an update. J. Intellect. Disabil. Res. 51, 666–670.
Della Giustina, E., Forabosco, A., Botticelli, A., Pace, P., 1985. Neuropathology of the
Turner syndrome. Pediatr. Med Chir. 7, 49.
Doricchi, F., Thiebaut de Schotten, M., Tomaiuolo, F., Bartolomeo, P., 2008. White
matter (dis) connections and gray matter (dys) functions in visual neglect: gaining
insights into the brain networks of spatial awareness. Cortex 44, 983–995.
Eliez, S., Schmitt, J.E., White, C.D., Reiss, A.L., 2000. Children and adolescents with
velocardiofacial syndrome: a volumetric MRI study. Am. J. Psychiatry 157, 409–415.
Eliez, S., Blasey, C.M., Freund, L.S., Hastie, T., Reiss, A.L., 2001a. Brain anatomy, gender and
IQ in children and adolescents with Fragile X syndrome. Brain 124, 1610–1618.
Eliez, S., Blasey, C.M., Menon,V., White, C.D., Schmitt, J.E., Reiss,A.L., 2001b. Functional brain
imaging study of mathematical reasoning abilities in velocardiofacial syndrome
(del22q11.2). Genet. Med. 3, 49–55.
Farzin, F., Rivera, S.M., 2010. Dynamic object representations in infants with and without
Fragile X syndrome. Front. Hum. Neurosci. 4.
Farzin, F., Whitney, D., Hagerman, R.J., Rivera, S.M., 2008. Contrast detection in infants
with Fragile X syndrome. Vision Res. 48, 1471–1478.
ffytche, D.H., 2008. The hodology of hallucinations. Cortex 44, 1067–1083.
Fonov,V., Evans, A.C., Botteron,K., Almli, C.R., McKinstry,R.C., Collins,D.L., 2011. Unbiased
average age-appropriate atlases for pediatric studies. Neuroimage 54, 313–327.
Fox, C.J., Iaria, G., Barton, J.J.S., 2008. Disconnection in prosopagnosia and face processing.
Cortex 44, 996–1009.
Freund, L.S., Reiss, A.L., 1991. Cognitive profiles associated with the fra(X) syndrome in
males and females. Am. J. Med. Genet. 38, 542–547.
Galvez, R., Greenough, W.T., 2005. Sequence of abnormal dendritic spine development
in primary somatosensory cortex of a mouse model of the Fragile X mental retar-
dation syndrome. Am. J. Med. Genet. A 135A, 155–160.
Gothelf, D., Feinstein, C., Thompson, T., Gu, E., Penniman, L., Van Stone, E., Kwon, H.,
Eliez, S., Reiss, A.L., 2007a. Risk factors for the emergence of psychotic disorders
in adolescents with 22q11.2 deletion syndrome. Am. J. Psychiatry 164, 663–669.
Gothelf, D., Penniman, L., Gu, E., Eliez, S., Reiss, A.L., 2007b. Developmental trajectories
of brain structure in adolescents with 22q11.2 deletion syndrome: a longitudinal
study. Schizophr. Res. 96, 72–81.
Green, T., Gothelf, D., Glaser, B., Debbane, M., Frisch, A., Kotler, M., Weizman, A., Eliez,
S., 2009. Psychiatric disorders and intellectual functioning throughout develop-
ment in velocardiofacial (22q11.2 deletion) syndrome. J. Am. Acad. Child Adolesc.
Psychiatry 48, 1060–1068.
Hart, S.J., Davenport, M.L., Hooper, S.R., Belger, A., 2006. Visuospatial executive function in
Turner syndrome: functional MRI and neurocognitive findings. Brain 129, 1125–1136.
Held, K.R., et al., 1992. Mosaicism in 45,X Turner syndrome: does survival in early preg-
nancy depend on the presence of two sex chromosomes? Brain 88 (3), 288–294.
Hoeft, F., Barnea-Goraly, N., Haas, B.W., Golarai, G., Ng, D., Mills, D., Korenberg, J.,
Bellugi, U., Galaburda, A., Reiss, A.L., 2007a. More is not always better: increased
fractional anisotropy of superior longitudinal fasciculus associated with poor
visuospatial abilities in Williams syndrome. J. Neurosci. 27, 11960–11965.
Hoeft, F., Hernandez, A., Parthasarathy, S., Watson, C.L., Hall, S.S., Reiss, A.L., 2007b. Fronto-
striatal dysfunction and potential compensatory mechanisms in male adolescents
with Fragile X syndrome. Hum. Brain Mapp. 28, 543–554.
Holsen, L.M., Dalton, K.M., Johnstone, T., Davidson, R.J., 2008. Prefrontal social cognition
network dysfunction underlying face encoding and social anxiety in Fragile X
syndrome. Neuroimage 43, 592–604.
Holzapfel, M., Barnea-Goraly, N., Eckert, M.A., Kesler, S.R., Reiss, A.L., 2006. Selective
alterations of white matter associated with visuospatial and sensorimotor dysfunc-
tion in Turner syndrome. J. Neurosci. 26, 7007–7013.
Hubbard, E.M., Piazza, M., Pinel, P., Dehaene, S., 2005. Interactions between number
and space in parietal cortex. Nat. Rev. Neurosci. 6, 435–448.
Iglesias, J.E., Cheng-Yi, L., Thompson, P.M., Zhuowen, T., 2011. Robust brain extraction across
datasets and comparison with publicly available methods. IEEE Trans. Med. Imaging 30,
Irwin, S.A., Galvez, R., Greenough, W.T., 2000. Dendritic spine structural anomalies in
Fragile-x mental retardation syndrome. Cereb. Cortex 10, 1038–1044.
Jenkinson, M., Smith, S., 2001. A global optimisation method for robust affine registration
of brain images. Med. Image Anal. 5, 143–156.
variation to brain dysfunction and schizophrenia. Nat. Rev. Neurosci. 11, 402–416.
Kates, W.R., Abrams, M.T., Kaufmann, W.E., Breiter, S.N., Reiss, A.L., 1997. Reliability and
validity of MRI measurement of the amygdala and hippocampus in children with
Fragile X syndrome. Psychiatry Res. 75, 31–48.
Kates, W.R., Burnette, C.P., Jabs, E.W., Rutberg, J., Murphy, A.M., Grados, M., Geraghty,
M., Kaufmann, W.E., Pearlson, G.D., 2001. Regional cortical white matter reductions
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
in velocardiofacial syndrome: a volumetric MRI analysis. Biol. Psychiatry 49,
Kesler, S.R., Blasey, C.M., Brown, W.E., Yankowitz, J., Zeng, S.M., Bender, B.G., Reiss, A.L.,
2003. Effects of X-monosomy and X-linked imprinting on superior temporal gyrus
morphology in Turner syndrome. Biol. Psychiatry 54, 636–646.
Kesler, S.R., Haberecht, M.F., Menon, V., Warsofsky, I.S., Dyer-Friedman, J., Neely, E.K.,
Reiss, A.L., 2004. Functional neuroanatomy of spatial orientation processing in
Turner syndrome. Cereb. Cortex 14, 174–180.
Kiehl, T.R., Chow, E.W., Mikulis, D.J., George, S.R., Bassett, A.S., 2009. Neuropatho-
logic features in adults with 22q11.2 deletion syndrome. Cereb. Cortex 19,
Kogan, C.S., Bertone, A., Cornish, K., Boutet, I., Der Kaloustian, V.M., Andermann, E.,
Faubert, J., Chaudhuri, A., 2004. Integrative cortical dysfunction and pervasive mo-
tion perception deficit in Fragile X syndrome. Neurology 63, 1634–1639.
of the human brain from childhood to adulthood. Neuroimage 40, 1044–1055.
Lee, A.D., Leow, A.D., Lu, A., Reiss, A.L., Hall, S., Chiang, M.-C., Toga, A.W., Thompson,
P.M., 2007. 3D pattern of brain abnormalities in Fragile X syndrome visualized
using tensor-based morphometry. Neuroimage 34, 924–938.
2009. A study of diffusion tensor imaging by tissue-specific, smoothing-compensated
voxel-based analysis. Neuroimage 44 (3), 870–883.
Leow, A., Huang, S.-C., Geng, A., Becker, J., Davis, S., Toga, A., Thompson, P., 2005. Inverse
consistent mapping in 3D deformable image registration: its construction and statis-
tical properties. In: Christensen, G.E., Sonka, M. (Eds.), Inf. Process. Med. Imaging, vol.
3565. Springer, Berlin/Heidelberg, pp. 493–503.
Leow, A.D., Yanovsky, I., Ming-Chang, C., Lee, A.D., Klunder, A.D., Lu, A., Becker, J.T.,
Davis, S.W., Toga, A.W., Thompson, P.M., 2007. Statistical properties of Jacobian
maps and the realization of unbiased large-deformation nonlinear image registra-
tion. IEEE Trans. Med. Imaging 26, 822–832.
Lippe, B., 1991. Turner syndrome. Endocrinol. Metab. Clin. North Am. 20, 121–152.
Machado, A.M., Simon, T.J., Nguyen, V., McDonald-McGinn, D.M., Zackai, E.H., Gee, J.C.,
2007. Corpus callosum morphology and ventricular size in chromosome 22q11.2
deletion syndrome. Brain Res. 1131, 197–210.
Mazzocco, M.M., Bhatia, N.S., Lesniak-Karpiak, K., 2006. Visuospatial skills and their
association with math performance in girls with Fragile X or Turner syndrome. Child
Neuropsychol. 12, 87–110.
Meechan, D.W., Tucker, E.S., Maynard, T.M., LaMantia, A.S., 2009. Diminished dosage
of 22q11 genes disrupts neurogenesis and cortical development in a mouse
model of 22q11 deletion/DiGeorge syndrome. Proc. Natl. Acad. Sci. U. S. A. 106,
Menon, V., Leroux, J., White, C., Reiss, A., 2004. Frontostriatal deficits in Fragile X syn-
drome: relation to FMR1 gene expression. Proc. Natl. Acad. Sci. U. S. A. 101, 3615.
Molko, N., Cachia, A., Riviere, D., J-Fo, Mangin, Bruandet, M., Le Bihan, D., Cohen, L.,
Dehaene, S., 2003. Functional and structural alterations of the intraparietal sulcus
in a developmental dyscalculia of genetic origin. Neuron 40, 847–858.
Molko, N., Cachia, A., Riviere, D., Mangin, J.F., Bruandet, M., LeBihan, D., Cohen, L.,
Dehaene, S., 2004. Brain anatomy in Turner syndrome: evidence for impaired
social and spatial-numerical networks. Cereb. Cortex 14, 840–850.
Murphy, D.G., DeCarli, C., Daly, E., Haxby, J.V., Allen, G., White, B.J., McIntosh, A.R.,
Powell, C.M., Horwitz, B., Rapoport, S.I., et al., 1993. X-chromosome effects on
female brain: a magnetic resonance imaging study of Turner's syndrome. Lancet
Murphy, M.M., Mazzocco, M.M.M., Gerner, G., Henry, A.E., 2006. Mathematics learn-
ing disability in girls with Turner syndrome or Fragile X syndrome. Brain Cogn.
Nilsson, K.O., Albertsson-Wikland, K., Alm, J., Aronson, S., Gustafsson, J., Hagenas, L.,
Hager, A., Ivarsson, S.A., Karlberg, J., Kristrom, B., Marcus, C., Moell, C., Ritzen, M.,
Tuvemo, T., Wattsgard, C., Westgren, U., Westphal, O., Aman, J., 1996. Improved
final height in girls with Turner's syndrome treated with growth hormone and
oxandrolone. J. Clin. Endocrinol. Metab. 81, 635–640.
Oberle, I., Rousseau, F., Heitz, D., Kretz, C., Devys, D., Hanauer, A., Boue, J., Bertheas, M.,
Mandel, J., 1991. Instability of a 550-base pair DNA segment and abnormal meth-
ylation in Fragile X syndrome. Science 252, 1097–1102.
Palo, J., Savolainen, H., 1973. The proteins of human myelin in inborn errors of metab-
olism and in chromosomal anomalies. Acta Neuropathol. 24, 56–61.
Qiu, D., Tan, L.-H., Zhou, K., Khong, P.-L., 2008. Diffusion tensor imaging of normal white
matter maturation from late childhood to young adulthood: voxel-wise evaluation
of mean diffusivity, fractional anisotropy, radial and axial diffusivities, and correla-
tion with reading development. Neuroimage 41, 223–232.
Reiss, A.L., Dant, C.C., 2003. The behavioral neurogenetics of Fragile X syndrome:
analyzing gene–brain–behavior relationships in child developmental psychopa-
thologies. Dev. Psychopathol. 15, 927–968.
effects of the FMR-1 full mutation in humans. Nat. Med. 1, 159–167.
Reiss, A.L., Mazzocco, M.M., Greenlaw, R., Freund, L.S., Ross, J.L., 1995b. Neurodevelopmental
effects of X monosomy: a volumetric imaging study. Ann. Neurol. 38, 731–738.
Rivera, S.M., Menon, V., White, C.D., Glaser, B., Reiss, A.L., 2002. Functional brain activa-
tion during arithmetic processing in females with Fragile X syndrome is related to
FMR1 protein expression. Hum. Brain Mapp. 16, 206–218.
Robin, N.H., Taylor, C.J., McDonald-McGinn, D.M., Zackai, E.H., Bingham, P., Collins, K.J.,
Earl, D., Gill, D., Granata, T., Guerrini, R., Katz, N., Kimonis, V., Lin, J.P., Lynch, D.R.,
Mohammed, S.N., Massey, R.F., McDonald, M., Rogers, R.C., Splitt, M., Stevens,
C.A., Tischkowitz, M.D., Stoodley, N., Leventer, R.J., Pilz, D.T., Dobyns, W.B., 2006.
cortical malformation. Am. J. Med. Genet. A 140, 2416–2425.
Romans, S.M., Stefanatos, G., Roeltgen, D.P., Kushner, H., Ross, J.L., 1998. Transition to
young adulthood in Ullrich–Turner syndrome: neurodevelopmental changes. Am.
J. Med. Genet. 79, 140–147.
Ross, J., Stefanatos, G., Kushner, H., Zinn, A., Bondy, C., Roeltgen, D., 2002. Persistent
cognitive deficits in adult women with Turner syndrome. Neurology 58, 218–225.
Ross, J., Zinn, A., McCauley, E., 2000. Neurodevelopmental and psychosocial aspects of
Turner syndrome. Ment. Retard. Dev. Disabil. Res. Rev. 6, 135–141.
Rovet, J., Netley, C., 1980. The mental rotation task performance of Turner syndrome
subjects. Behav. Genet. 10 (5), 437–443.
Rovet, J., Ireland, L., 1994. Behavioral Phenotype in Children with Turner Syndrome.
J. Pediatr. Psychol. 19 (6), 779–790.
Rudelli, R.D., Brown, W.T., Wisniewski, K., Jenkins, E.C., Laure-Kamionowska, M.,
Connell, F., Wisniewski, H.M., 1985. Adult Fragile X syndrome. Acta Neuropathol.
Scerif, G., Cornish, K., Wilding, J., Driver, J., Karmiloff-Smith, A., 2004. Visual search in
typically developing toddlers and toddlers with Fragile X or Williams syndrome.
Dev. Sci. 7, 116–130.
Scerif, G., Cornish, K., Wilding, J., Driver, J., Karmiloff-Smith, A., 2007. Delineation of early
attentional control difficulties in Fragile X syndrome: focus on neurocomputational
changes. Neuropsychologia 45, 1889–1898.
Schaer, M., Eric Schmitt, J., Glaser, B., Fo, Lazeyras, Delavelle, J., Eliez, S., 2006. Abnormal
patterns of cortical gyrification in velo-cardio-facial syndrome (deletion 22q11.2):
an MRI study. Psychiatry Res. Neuroimaging 146, 1–11.
Shashi, V., Muddasani, S., Santos, C.C., Berry, M.N., Kwapil, T.R., Lewandowski, E.,
Keshavan, M.S., 2004. Abnormalities of the corpus callosum in nonpsychotic chil-
dren with chromosome 22q11 deletion syndrome. Neuroimage 21, 1399–1406.
Silbert, A., Wolff, P.H., Lilienthal, J., 1977. Spatial and temporal processing in patients
with Turner's syndrome. Behav. Genet. 7 (1), 11–21.
Simon, T.J., 1997. Reconceptualizing the origins of number knowledge: a “non-numer-
ical” account. Cogn. Dev. 12, 349–372.
Simon, T.J., 2007. Cognitive characteristics of children with genetic syndromes. Child
Adolesc. Psychiatr. Clin. N. Am. 16, 599–616.
Simon, T.J., 2011. Clues to the foundations of numerical cognitive impairments: evidence
from genetic disorders. Dev. Neuropsychol. 36, 788–805.
Simon, T.J., Bearden, C.E., Mc-Ginn, D.M., Zackai, E., 2005a. Visuospatial and numerical
cognitive deficits in children with chromosome 22q11.2 deletion syndrome. Cortex
Simon, T.J., Ding, L., Bish, J.P., McDonald-McGinn, D.M., Zackai, E.H., Gee, J., 2005b.
Volumetric, connective, and morphologic changes in the brains of children with
chromosome 22q11.2 deletion syndrome: an integrative study. Neuroimage 25,
Simon, T.J., Takarae, Y., DeBoer, T., McDonald-McGinn, D.M., Zackai, E.H., Ross, J.L.,
2008a. Overlapping numerical cognition impairments in children with chromo-
some 22q11.2 deletion or Turner syndromes. Neuropsychologia 46, 82–94.
Simon, T.J., Wu, Z., Avants, B., Zhang, H., Gee, J.C., Stebbins, G.T., 2008b. Atypical cortical
connectivity and visuospatial cognitive impairments are related in children with
chromosome 22q11.2 deletion syndrome. Behav. Brain Funct. 4, 25.
Smith, S.M., Jenkinson, M., Johansen-Berg, H., Rueckert, D., Nichols, T.E., Mackay, C.E.,
Watkins, K.E., Ciccarelli, O., Cader, M.Z., Matthews, P.M., Behrens, T.E.J., 2006.
Tractbased spatial statistics: voxelwise analysis of multisubject diffusion data.
NeuroImage 31, 1487–1505.
Stoddard, J., Niendam, T., Hendren, R., Carter, C., Simon, T.J., 2010. Attenuated positive
symptoms of psychosis inadolescentswithchromosome22q11.2deletionsyndrome.
Schizophr. Res. 118, 118–121.
Sy, J., Alturkustani, M., Ang, L.-C., 2010. Intracallosal longitudinal fiber bundle:
an unexpected finding mimicking demyelination in a patient with Turner syndrome.
Acta Neuropathol. 120, 545–547.
Sztriha, L., Guerrini, R., Harding, B., Stewart, F., Chelloug, N., Johansen, J.G., 2004.
Clinical, MRI, and pathological features of polymicrogyria in chromosome 22q11
deletion syndrome. Am. J. Med. Genet. A 127A, 313–317.
Temple, C.M., Carney, R.A., 1995. Patterns of spatial functioning in Turner's syndrome.
Cortex 31 (1), 109–118.
Terao, Y., Hashimoto, K., Nukina, N., Mannen, T., Shinohara, T., 1996. Cortical dysgenesis
in a patient with Turner mosaicism. Dev. Med. Child Neurol. 38, 455–460.
Thomason, M.E., Thompson, P.M., 2011. Diffusion imaging, white matter, and psycho-
pathology. Annu. Rev. Clin. Psychol. 7, 63–85.
Tombini, M., Marciani, M.G., Romigi, A., Izzi, F., Sperli, F., Bozzao, A., Floris, R., De Simone,
R., Placidi, F., 2003. Bilateral frontal polymicrogyria and epilepsy in a patient with
Turner mosaicism: a case report. J. Neurol. Sci. 213, 83–86.
Tournier, J.-D., Mori, S., Leemans, A., 2011. Diffusion tensor imaging and beyond. Magn.
Reson. Med. 65, 1532–1556.
van Amelsvoort, T., Daly, E., Robertson, D., Suckling, J., Ng, V., Critchley, H., Owen, M.J.,
Henry, J., Murphy, K.C., Murphy, D.G., 2001. Structural brain abnormalities associ-
ated with deletion at chromosome 22q11: quantitative neuroimaging study of
adults with velo-cardio-facial syndrome. Br. J. Psychiatry 178, 412–419.
van Amelsvoort, T., Schmitz, N., Daly, E., Deeley, Q., Critchley, H., Henry, J., Robertson,
D., Owen, M., Murphy, K.C., Murphy, D.G., 2006. Processing facial emotions in
adults with velo-cardio-facial syndrome: functional magnetic resonance imaging.
Br. J. Psychiatry 189, 560–561.
van Eimeren, L., Grabner, R.H., Koschutnig, K., Reishofer, G., Ebner, F., Ansari, D., 2010.
Structure–function relationships underlying calculation: a combined diffusion tensor
imaging and fMRI study. Neuroimage 52, 358–363.
Verkerk, A.J.M.H., Pieretti, M., Sutcliffe, J.S., Fu, Y.-H., Kuhl, D.P.A., Pizzuti, A., Reiner, O.,
Richards, S., Victoria, M.F., Zhang, F., Eussen, B.E., van Ommen, G.-J.B., Blonden, L.A.J.,
Riggins, G.J., Chastain, J.L., Kunst, C.B., Galjaard, H., Thomas Caskey, C., Nelson, D.L.,
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454
Oostra, B.A., Warren, S.T., 1991. Identification of a gene (FMR-1) containing a CGG re- Download full-text
peat coincident with a breakpoint cluster region exhibiting length variation in Fragile
X syndrome. Cell 65, 905–914.
Vestergaard, M., Madsen, K.S., Baar√©, W.F.C., Skimminge, A., Ejersbo, L.R., Rams√∏y,
T.Z., Gerlach, C., √Ökeson, P., Paulson, O.B., Jernigan, T.L., 2010. White matter mi-
crostructure in superior longitudinal fasciculus associated with spatial working
memory performance in children. J. Cogn. Neurosci. 23, 2135–2146.
Walsh, V., 2003. A theory of magnitude: common cortical metrics of time, space and
quantity. Trends Cogn. Sci. 7, 483–488.
Walter, E., Mazaika, P.K., Reiss, A.L., 2009. Insights into brain development from
neurogenetic syndromes: evidence from Fragile X syndrome, Williams syndrome,
Turner syndrome and velocardiofacial syndrome. Neuroscience 164, 257–271.
Wang, P.P., Woodin, M.F., Kreps-Falk, R., Moss, E.M., 2000. Research on behavioral
phenotypes: velocardiofacial syndrome (deletion 22q11.2). Dev. Med. Child Neurol.
Wisniewski, K.E., Segan, S.M., Miezejeski, C.M., Sersen, E.A., Rudelli, R.D., 1991. The fra(X)
syndrome: neurological, electrophysiological, and neuropathological abnormalities.
Am. J. Med. Genet. 38, 476–480.
Woods, R.P., Grafton, S.T., Watson, J.D.G., Sicotte, N.L., Mazziotta, J.C., 1998. Automated
image registration: II. Intersubject validation of linear and nonlinear models. J. Comput.
Assist. Tomogr. 22, 153–165.
Yamagata, B., Barnea–Goraly, N., Marzelli, M.J., Park, Y., Hong, D.S., Mimura, M., Reiss,
A.L., 2011. White Matter Aberrations in Prepubertal EstrogenNaive Girls with
Monosomic Turner Syndrome. Cereb. Cortex.
J. Villalon-Reina et al. / NeuroImage 81 (2013) 441–454