Diagnostic, carrier and prenatal genetic testing for fragile X syndrome and other FMR-1-related disorders in Johannesburg, South Africa: A 20-year review
1 Division of Human Genetics, School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
Division of Human Genetics,
National Health Laboratory Service, Johannesburg, South Africa
Background. Fragile X syndrome (FXS), the most common inherited cause of intellectual disability (ID) worldwide, is caused by the expansion of a CGG repeat in the fragile X mental retardation gene (FMR-1) gene.
Objectives. To review, retrospectively, the genetic services for FXS and other FMR-1-related disorders – including fragile X-associated tremor/ataxia syndrome (FXTAS) and FMR-1-related primary ovarian insufficiency (POI) – at the Division of Human Genetics, Johannesburg, for diagnostic, carrier and prenatal genetic testing.
Methods. The records of 2 690 patients who had genetic testing for FMR-1 between 1992 and 2012 were reviewed. Of these, 2 239 had diagnostic testing, 430 carrier or cascade testing and 17 prenatal testing for FXS. Four had FXTAS or POI testing. Polymerase chain reaction (PCR) and/or Southern blotting techniques were used to test the patients’ samples for FMR-1 and FMR-2 expansions.
Results. Of the 2 239 patients who had diagnostic FMR-1 testing, 128 (5.7%) had a full mutation, 12 (0.5%) had a premutation and 43 (1.9%) an intermediate allele. In 17 prenatal tests, eight fetuses tested positive for FXS. FMR-1 CGG repeat distribution analysis in 1 532 males negative for the FMR-1 expansion showed that 29 and 30 CGG repeats were the most common (61.1%), but the distribution was significantly different in the black and white populations.
Conclusion. The findings support the
presence of FXS, as the most common cause of ID, in
all local populations. The FMR-1 CGG repeat distribution
varied from that found in other studies. The number of
family members tested was relatively low suggesting
that many at-risk individuals are not being referred.
S Afr Med J 2013;103(12 Suppl
Intellectual disability (ID) has an estimated prevalence of 2 - 3% in developed countries, but the prevalence of ID and its epidemiology in developing countries is not well established.1 However, a study in a rural South African (SA) population reported the prevalence to be 3.6%.2 The most common cause of ID in SA is fetal alcohol syndrome, which occurs at the highest rate in the world and at epidemic proportions in some communities in the Western Cape Province.3 Fragile X syndrome (FXS) is the most common inherited cause of ID in all populations.
FXS is an X-linked neurodevelopmental disorder, with a variable phenotype in males and females, associated with intellectual, physical and behavioural features. ID is the most consistent feature, ranging from mild to severe, while the other clinical features include developmental delay, long narrow face, large prominent forehead and chin, protruding ears, macro-orchidism in post-pubertal males, hyperactivity, hand flapping, attention deficit and autism.4 FXS is estimated to have a prevalence of approximately 1/4 000 in males and 1/8 000 in females worldwide.5
Almost all cases of FXS are caused by the expansion of a CGG repeat in the 5' untranslated region of the fragile X mental retardation gene (FMR-1) (OMIM #309550) at the FRAXA fragile site Xq27.3. A small number of deletions and missense mutations have also been reported. The CGG repeat size varies from 5 to 44 repeats in unaffected individuals. Intermediate (IM) alleles (‘gray zone’) range from 45 to 54 CGG repeats. The clinical significance of carrying an allele in this size range is unclear. Premutation (PM) carriers have expansions of 55 to 200 repeats, which are not associated with ID. Affected individuals have >200 CGG repeats and are classified as having a full mutation (FM) which is associated with hypermethylation of an upstream gene promoter region (CpG island) and silencing of FMR-1 gene transcription which results in the absence of the gene product, fragile X mental retardation protein (FMRP).6
FMR-1-related disorders include FXS and two other clinically distinct conditions, fragile X-associated tremor/ataxia syndrome (FXTAS) (OMIM #300623) and FMR-1-related primary ovarian insufficiency (FXPOI) (OMIM #311360). Both males and females with a PM in FMR-1 have an increased risk of developing FXTAS, a progressive neurodegenerative disorder with a prevalence of 45% reported in males older than 50 years of age. Female PM carriers are less likely to develop FXTAS but have a 21% risk of developing primary ovarian insufficiency (POI) (menopause before the age of 40 years).6
The FMR-1 gene, identified in 1991, is 38 kb long and consists of 17 exons which are alternatively spliced. It codes for the cytoplasmic protein FMRP, a 631 amino acid RNA-binding protein that is expressed in almost every tissue, with highest levels observed in the brain and testes. FMRP has been shown to down-regulate the translation of specific target messenger RNAs and plays a vital role in synaptic plasticity.7
In 1981, the first SA cases of FXS were described in four families investigated by cytogenetic studies. Eleven affected males and one carrier female were found to have a fragile site on the X chromosome (fra(X)(q27)).8 Cytogenetic detection of FRAXA is not reliable as a diagnostic test for FXS as it cannot differentiate between this and other fragile sites, such as FRAXD, FRAXE and FRAXF, and it is costly. Goldman et al. 9 , 10 presented the first molecular evidence that FXS occurs in the SA black ID population. They found nine (6.1%) FXS cases in 148 patients with ID suggesting that the condition had been underdiagnosed in the past.
DNA testing for FXS
has been available in SA since 1994 at the Division of
Human Genetics (DHG), National Health Laboratory
Service (NHLS) and the University of the
Witwatersrand, Johannesburg. Prior to this, the
diagnostic test was based exclusively on cytogenetic
detection of the FRAXA fragile site. Currently,
testing is performed by duplex polymerase chain
reaction (PCR) analysis to size the normal and lower
PM FMR-1 alleles, followed by
Southern blotting to confirm expansions. The PCR assay
is reliable, relatively cheap, and an efficient test
as a primary screen for FXS. Mosaic males with a
normal allele and an FM will yield a false-negative
To review and
analyse the findings from DNA testing for diagnostic,
carrier and prenatal purposes for FXS and other FMR-1-related disorders,
conducted at the DHG over a 20-year period. The study
provided a follow-up to a previous study by Goldman11 in which FXS testing
was evaluated by the DHG. The presence of FRAXE
mutations and the distribution of FMR-1 and FMR-2 repeat sizes in the ID
cohort negative for FXS were also determined as part
of the present study.
Patients tested for FXS and FMR-1-related disorders
A total of 2 690 individuals from 2 243 families underwent genetic testing for FXS and other FMR-1-associated disorders through the DHG during the 20-year period (January 1992 - July 2012). Of these, 2 239 unrelated individuals (887 white, 959 black, 211 mixed ancestry and 182 Indian) had a suspected diagnosis of FXS or ID of unknown cause (1 961 males and 278 females), 430 were extended family members referred for FXS carrier or diagnostic testing (mothers of affected individuals, siblings and other at-risk second and third-degree relatives), four requested POI or FXTAS testing, and 17 were prenatal samples (obtained by chorionic villus sampling (CVS) or amniocentesis) from at-risk pregnancies.
The majority of referrals
were from the Assessment and Learning Clinics at the
Johannesburg (now the Charlotte Maxeke) Academic
Hospital, Coronation (now Rahima Moosa) Hospital,
Alexandra Clinic and institutions for people with ID.
Further referrals were from local general practitioners,
paediatricians and medical geneticists at the Genetic
Counselling Clinics (at Chris Hani Baragwanath Academic
Hospital, Johannesburg Hospital, Donald Gordon Medical
Centre and Coronation Hospital) and genetic nurses.
However, a small number of cases were referred from
other genetic centres around SA and from doctors in
private practice. Blood samples from the patients were
sent in with a referral form stating the possible
diagnosis and, in some cases, the clinical features
observed. Information collected from patient files
included (where available) each subject’s gender,
ethnicity, their test results and pregnancy information
Genomic DNA was
extracted from whole blood using a salting out method
or a commercial DNA extraction kit (High Pure PCR
Template Preparation Kit, Roche Diagnostics) for blood
samples <1 ml. DNA was extracted
from CVS and amniocyte material using a
phenol-chloroform extraction method.
PCR amplification was performed using published primers12
with detection of products on the ABI PRISM 377 or ABI
3130xl Genetic Analyzer (Applied Biosystems). Testing
for AGG interruptions was not performed.
Southern blot analysis
analysis was performed to establish the methylation
status of the promoter region, to detect FMs that
failed to amplify on PCR analysis and to differentiate
females who were homozygous for a normal-sized FMR-1 allele from female
PM/FM heterozygotes. Analysis of the FMR-1 expansion was carried
out by digestion of genomic DNA (5 - 15
μg) with both EcoRI (Roche) and a
methylation-sensitive enzyme, EclXI (Roche).
Hybridisation was performed using the StB12.3 probe
and labelled with the radionuclide isotope α-32PdCTP using the
Megaprime DNA Labelling Systems kit (Amersham).
Of the 2 239 unrelated individuals with ID/?FXS, 128 (5.7%) tested positive for the FMR-1 FM (117; 5.2% male and 11; 0.5% female probands). Of these, the FM was observed in 4.8% (43/887) of white patients, 5.2% (50/959) of black, 8.1% (17/211) of mixed ancestry and 9.9% (18/182) of Indian patients. Chi-squared analysis showed no significant difference (p=0.72) between the observed frequencies of FMs in the black and white populations (Table 1).
PMs accounted for
0.54% (12/2 239) of the total
diagnostic cohort, ranging from 0% to 1% in the
different ethnic groups tested; while IM alleles were
observed in 1.8% (41/2 239) of the cohort,
ranging from 1.1% to 2.7% in the different ethnic
Of the 430 extended family members tested in the 2 243 families, an FM was
found in 51 (11.8%; 22 males; 29 females) subjects, a PM
and/or IM in 70 (16.3%; 59 females; 11 males) and 309
(71.9%) individuals tested negative.
FXTAS and POI referrals
A PM was found in 1/3 females tested, and one male
referred for FXTAS tested negative.
Prenatal diagnostic testing
A total of 17
prenatal tests (12 white, four Indian, one black and
none of mixed ancestry) were performed on CVS material
or amniotic cells. Eight of the 17 (47%) fetuses had
an FMR-1 FM (five male and three
female), two a PM (one male and one female) and seven
(41%) tested negative.
Female siblings with FMR-1 expansions of both alleles
A large family
positive for the FMR-1 expansion (n=24) was investigated
and two female siblings were identified as having an FMR-1 expansion on both X
chromosomes. Both siblings were found to have one IM
allele of 51 CGG repeats and a second PM allele of 57
repeats in one sibling and 69 repeats in the other.
Both siblings inherited the 51 repeat unchanged from
their father while the 57 repeat from their mother was
inherited stably in one, but was shown to have
expanded to 69 repeats in the other, and subsequently
it expanded to an FM in her grandson.
FRAXE mental retardation syndrome
the FMR-2 GCC repeat was used as
an internal control primer in the PCR assay. An
expansion of >200 GCC repeats in the FMR-2
gene at Xq28 has been
shown to cause FRAXE syndrome (OMIM #309548), a rare
condition associated with a mild form of ID.14 Of the 1 961 males with ID tested
on the duplex PCR assay, FMR-2 results were obtained
for 1 558 subjects (who tested
negative for the FMR-1 expansion). None were
found to have an FMR-2 FM. However, an IM
allele (26 - 59 GCC repeats) was observed in four
Distribution of FMR-1 CGG repeat sizes
The distribution of
the CGG repeat size in 1 532 FMR-1 expansion-negative ID
subjects is shown in Fig. 1. The smallest FMR-1 CGG repeat size
detected was 9. The most frequent repeat size observed
was 29 CGG repeats (32.7%; 501/1 532) followed by 30 CGG
repeats (28.4%; 435/1 532). CGG repeat sizes
20, 23, 24, 29, 30, 31 and 32 together accounted for
the majority (86%; 1 317/1 532) of alleles in the
study population. The most common allele observed in
the black and mixed-ancestry population was 29 CGG
repeats (the frequency was 41%; 284/689 and 53/131,
respectively), while 30 CGG repeats was the most
common number in the white (34%; 207/606) and Indian
populations (41%; 43/106). Chi-squared analysis showed a
significant difference (p<10-5) in the frequency of
the 29 and 30 CGG repeat sizes in the black and white
Fig. 1. Distribution of FMR-1 CGG repeat sizes in subjects negative for the FMR-1 expansion (N=1 532).
Distribution of FMR-2 GCC repeat sizes
The distribution of
the GCC repeat size in 1 558 FMR-1-expansion-negative
subjects is shown in Fig. 2. FMR-2 allele sizes ranged
from 5 to 40 GCC repeats, with 15 being the most
common repeat size, accounting for 41% (640/1 558) of alleles
identified. The distribution was evident in each of
the four ethnic groups, and observed at a frequency of
33.8% (207/613) in whites, 48.7% (343/704) in blacks,
37.6% (50/133) in mixed ancestry and 37% (40/108) in
analysis showed a
significant difference (p<10-5) between the
distributions of FMR-2 alleles in the black and
Fig 2. Distribution of FMR-2 G CC repeat sizes in subjects negative for the FMR-1 expansion (N=1 558).
The present study reviewed the molecular genetic testing for FXS and other FMR-1-related disorders conducted over a 20-year period at the DHG, NHLS, Johannesburg.
In 2000, the PCR
assay was introduced as an initial screen for the
routine diagnosis of FXS. This has improved and
refined the diagnostic testing by accurate
determination of the FMR-1 CGG repeat size, which
was not possible with Southern blotting. It has also
reduced the number of Southern blots required and
decreased the turnaround time for patients with CGG
repeat sizes up to 120 repeats.
Full mutations in the total cohort
An FMR-1 FM was identified in 5.7% (128/2 239) of the probands tested, comparable with findings from ID cohorts worldwide which show FXS estimates ranging from 0.5% to 5%, depending on how cohorts were determined.15 Our study provides further evidence to support that of Goldman et al.9 that FXS, previously thought to be rare in the SA black population, is underdiagnosed. FMs were observed in 5.2% (50/959) of all black subjects tested, which is slightly lower than that found in Goldman et al.’s small sample (nine affected in 148 patients; 6.1%), but may be considered to be a more accurate reflection due to the larger sample involved. This result lends further support to the findings of Goldman et al.9 that FXS is a common cause of ID in all SA ethnic groups.
In the Indian
population, FMs were found in 9.9% (18/182) of
referrals for ID, which is higher than that observed
in the other three ethnic groups. The majority of
these Indian probands were referred from
KwaZulu-Natal, and a selection bias is possible, as
testing is expensive and referrals may be limited to
patients with a high clinical suspicion of FXS.
Interestingly, studies in the Indian population on the
Indian subcontinent report a frequency of
approximately 7.48% for FXS due to FMs using molecular
analysis,16 which is lower than the
frequency observed in the present study but higher
than worldwide figures. The frequency of FMs in the
mixed-ancestry population was 8.1%, which is also
higher than the frequency observed in the white and
black ethnic groups, but the sample size was small and
the differences were not statistically significant.
Premutations and intermediate alleles in the ID cohort
In the total cohort, a PM was detected in 0.54% (12/2 239) while an IM allele was detected in 1.83% (41/2 239). Both PMs and IM alleles were observed at higher frequencies within the white ID population (1% and 2.7%, respectively) compared with the other populations. PMs are estimated to occur in the general population at a prevalence of ~1/113 - 1/259 among females (~0.5%) and ~1/260 - 1/800 (~0.2%) among males; i.e. much more common than in FXS (1/4 000 males).4 Further, a significant excess of IM alleles in boys with special education needs compared with normal controls has been reported. There is controversy around whether or not males and females with a PM have subtle behavioural or learning difficulties. However, a number of studies have suggested that there is no clinical significance to the presence of PMs or IMs in individuals with ID.17
of extended family members detected an additional 121
subjects with an FMR-1 FM, including mothers
of affected individuals, siblings and other at-risk
second- and third-degree relatives. As the FMR-1 mutation is rarely a
new mutation, cascade screening from a proband is very
worthwhile as it often detects other affected
individuals in a family. The number of family members
tested by cascade screening in the present study was
low, suggesting that many at-risk family members are
not being tested and/or offered appropriate genetic
Female siblings with FMR-1 expansions of both alleles
The family in which two females were found to have
expansions of both alleles is the first such case
described in SA and the fifth such family in the world.18
Clinical data were not available on these females, but
the two sisters would not be expected to present with
ID. Mostly, studies have shown that females with two PMs
do not have ID nor physical features typical of FXS,
while those with one FM have characteristics typical of
females with FXS.18 They would be
predicted to be at risk of POI and FXTAS, but the exact
risk is unclear.
The present study
confirms the rarity of FRAXE syndrome in SA,
consistent with reports from other studies.19 FRAXE syndrome has an
estimated population prevalence of 1/23 500, based on two large
population-based studies undertaken on special
education needs individuals. A study on a cohort of
white and African American subjects showed no FMR-2 expansion, while
another report found one FRAXE syndrome-positive
subject among 3 731 boys with ID.19 FRAXE screening is
therefore not indicated and testing has not been
implemented routinely in the DHG. Amplification of the
FRAXE allele is, however, useful in the duplex PCR
FRAX assay as a control for PCR failure, especially in
males, and as an indicator of an FMR-1 expansion. Its value
has also been demonstrated as a linked marker in
tracking the FMR-1 expansion in families,
especially in prenatal testing, as it may reduce the
time taken to obtain an initial result.
Distribution of FMR-1 alleles
investigated the distribution of the CGG repeat in 1 532 FMR-1-expansion-negative ID
subjects. The CGG repeat sizes of 29 and 30 were found
to be most common in all ethnic groups, together
accounting for 61.1% of the alleles in the cohort and
comparable with populations of western European
ancestry and other populations in Asia, Indonesia,
Brazil, Chile, Cameroon, Ghana and India, and in
African Americans. 20 In the present study the
30 CGG repeat (found in 34% of cases) was the most
common allele in the white population, which
corroborates well with the 39% frequency reported by
Goldman.11 The most common allele
in the black population in this study was shown to be
29 CGG repeats (in 41% of the total black cohort).
This finding differs from that of Goldman,11 who showed that 28 CGG
repeats (32%) was a common finding in SA black
individuals. Our study is more likely to reflect the
correct size as newer techniques have allowed for more
accurate sizing. However, a study in a Cameroonian
population also showed the 29 CGG repeat to be most
common, whereas studies among African Americans and
Ghanaians suggested that the 30 CGG repeat was the
most common allele. Further studies are needed to
distinguish true distribution differences from
technical ones. Very few studies have been performed
on African populations and the present study adds
significantly to the African population data.
Distribution of FMR-2 alleles
This is the first study conducted on SA males with ID which provides information on the range of the FMR-2 GCC repeat sizes. FMR-2 allele distribution in most ID populations showed that the 15 GCC repeat was the most frequent allele, similar to that observed in our study. Four males in this sample were found to have an FMR-2 repeat size in the IM range. Studies have disputed the association of IMs with ID and suggested that more studies with larger numbers are needed to resolve this matter.19
This study strongly supports the fact that FXS is the most common cause of ID in all the local populations. However, FRAXE syndrome was not shown to contribute significantly to ID in the SA population and has therefore not been incorporated into the routine diagnostic testing. Earlier studies undertaken in our laboratory on X-linked mental retardation genes also suggested that mutations in ARX, XNP and HOPA like most others, are rare, and not significant contributors to ID locally.21 , 22
The presence of an
FMR-1 expansion in a family
member has significant implications, as there are
likely to be other affected and at-risk relatives. A
confirmed diagnosis leads to improved support for
newly-diagnosed patients and potentially allows
identification of all the carriers in the family.
Prenatal and pre-implantation genetic diagnosis can be
offered to at-risk individuals and this service allows
couples to make informed choices about their future.
Attempts should be made to expand offers of cascade
testing once a proband is diagnosed. Comprehensive
genetic counselling services need to become more
widely available so that they can assist affected
families with these issues.
authors thank colleagues at the DHG Molecular Diagnostic
Laboratory, who have performed the routine testing over
the years. Funding for this study was provided from the
NHLS Research Trust, the SA Medical Research Council and
the HE Griffin Charitable Trust, University of the
1. Roeleveld N, Zielhuis GA, Gabreëls F. The prevalence of mental retardation: A critical review of recent literature. Dev Med Child Neurol 1997;39(2):125-132. [http://dx.doi.org/10.1111/j.1469-8749.1997.tb07395.x]
2. Kromberg J, Zwane E, Manga P, Venter A, Rosen E, Christianson A. Intellectual disability in the context of a South African population. JPPID 2008;5(2):89-95. [http://dx.doi.org/10.1111/j.1741-1130.2008.00153.x]
3. May PA, Gossage JP, Marais A-S, et al. The epidemiology of fetal alcohol syndrome and partial FAS in a South African community. Drug Alcohol Depend 2007;88(2-3):259-271. [http://dx.doi.org/10.1016/j.drugalcdep.2006.11.007]
4. Pirozzi F, Tabolacci E, Neri G. The FRAXopathies: Definition, overview, and update. Am J Med Genet 2011;155A(8):1803-1816. [http://dx.doi.org/10.1002/ajmg.a.34113]
5. Coffee B, Keith K, Albizua I, et al. Incidence of fragile X syndrome by newborn screening for methylated FMR1 DNA. Am J Hum Genet 2009;85(4):503-514. [http://dx.doi.org/10.1016/j.ajhg.2009.09.007]
6. Saul R, Tarleton J. FMR-1 related disorders. In: Pagon RA, Adam MP, Bird TD, et al, eds. GeneReviews. Seattle: University of Washington, 1993-2013. http://www.ncbi.nlm.nih.gov/books/NBK1384/ (accessed 9 May 2013).
7. Santoro MR, Bray SM, Warren ST. Molecular mechanisms of fragile X syndrome: A twenty-year perspective. Annu Rev Pathol 2012;7(1):219-245. [http://dx.doi.org/10.1146/annurev-pathol-011811-132457]
8. Venter PA, Gericke GS, Dawson B, Op’t Hof J. A marker X chromosome associated with nonspecific male mental retardation. The first South African cases. S Afr Med J 1981;60(21):807-811.
9. Goldman A, Krause A, Jenkins T. Fragile X syndrome occurs in the South African black population. S Afr Med J 1997;87(4):418-420.
10. Goldman A, Jenkins T, Krause A. Molecular evidence that fragile X syndrome occurs in the South African black population. J Med Genet 1998;35(10):878. [http://dx.doi.org/10.1136/jmg.35.10.878]
11. Goldman, A. Trinucleotide Repeat Disorders in Southern African Populations: A Molecular Study. PhD thesis. Johannesburg: University of the Witwatersrand, 1997.
12. Fu YH, Kuhl DP, Pizzuti A, et al. Variation of the CGG repeat at the fragile X site results in genetic instability: Resolution of the Sherman paradox. Cell 1991;67(6):1047-1058. [http://dx.doi.org/10.1016/0092-8674(91)90283-5]
13. Wang Q, Green E, Bobrow M, Mathew CG. A rapid, non-radioactive screening test for fragile X mutations at the FRAXA and FRAXE loci. J Med Genet 1995;32(3):170-173. [http://dx.doi.org/10.1136/jmg.32.3.170]
14. Gecz J. The FMR2 gene, FRAXE and non-specific X-linked mental retardation: Clinical and molecular aspects. Ann Hum Genet 2000;64(2):95-106. [http://dx.doi.org/10.1046/j.1469-1809.2000.6420095.x]
15. Biancalana V, Beldjord C, Taillandier A, et al. Five years of molecular diagnosis of fragile X syndrome (1997-2001): A collaborative study reporting 95% of the activity in France. Am J Med Genet 2004;129A(3):218-224. [http://dx.doi.org/10.1002/ajmg.a.30237]
16. Kabra M, Sharma D, Singh D, et al. Fragile X screening for FRAXA and FRAXE mutations using PCR based studies: Results of a five year study. Ind J Hum Genet 2006;12(1):17. [http://dx.doi.org/10.4103/0971-6866.25297]
17. Sherman SL, Marsteller F, Abramowitz AJ, Scott E, Leslie M, Bregman J. Cognitive and behavioral performance among FMR1 high-repeat allele carriers surveyed from special education classes. Am J Med Genet 2002;114(4):458-465. [http://dx.doi.org/10.1002/ajmg.10303]
18. Linden MG, Tassone F, Gane LW, Hills JL, Hagerman RJ, Taylor AK. Compound heterozygous female with fragile X syndrome. Am J Med Genet 1999;83(4):318-321. [http://dx.doi.org/10.1002/(SICI)1096-8628(19990402)83:4<318::AID-AJMG16>3.3.CO;2-P]
19. Crawford DC, Meadows KL, Newman JL, et al. Prevalence and phenotype consequence of FRAXA and FRAXE alleles in a large, ethnically diverse, special education-needs population. Am J Hum Genet 1999;64(2):495-507. [http://dx.doi.org/10.1086/302260]
20. Peprah E. Fragile X syndrome: The FMR1 CGG repeat distribution among world populations. Ann Hutam Genet 2012;76(2):178-191. [http://dx.doi.org/10.1111/j.1469-1809.2011.00694.x]
21. Essop FB. Molecular Aspects of X-Linked Mental Retardation Loci. MSc (Med) dissertation. Johannesburg: University of the Witwatersrand, 2010.
22. Friez MJ, Essop FB, Krause A, et al. Evidence that a dodecamer duplication in the gene HOPA in Xq13 is not associated with mental retardation. Hum Genet 2000;106(1):36-39. [http://dx.doi.org/10.1007/s004390051006]
12 August 2013.
Full text views: 3085