Factors That May Complicate Inheritance Patterns

FACTORS THAT MAY COMPLICATE INHERITANCE PATTERNS The inheritance patterns described previously for postaxial polydactyly and albinism are quite straightforward. However, many autosomal diseases display more complex patterns. Some of these complexities are described next.

New Mutation If a child has been born with a genetic disease and there is no history of the disease in the family, it is possible that the disease is the product of a new mutation (this is especially likely if the disease in question is autosomal dominant). That is, the gene transmitted by one of the parents underwent a change in DNA resulting in a mutation from a normal to a diseasecausing allele. The genes at this locus in the parent's other germ cells would still be normal. In this case the recurrence risk for the parents' subsequent offspring would not be elevated above that of the general population. However, the offspring of the affected child may have a substantially elevated occurrence risk (e.g., it would be 50% for an autosomal dominant disease). A large proportion of the observed cases of many autosomal dominant diseases are the result of new mutations. For example, it is estimated that 7/8 of all cases of achondroplasia are caused by new mutations, while only 1/8 are transmitted by achondroplastic parents. This is primarily because the disease tends to limit the potential for reproduction. In order to provide accurate risk estimates, it is essential to know whether an observed case is due to an inherited disease gene or a new mutation. This can be done only if an adequate family history has been taken. ▪ New mutations are a frequent cause of the appearance of a genetic disease in an individual with no previous family history of the disorder. The recurrence risk for the individual's siblings is very low, but it may be substantially elevated for the individual's offspring.

Germline Mosaicism Occasionally, two or more offspring may present with an autosomal dominant or X-linked disease when there is no family history of the disease. Since mutation is a rare event, it is unlikely that this situation would be due to multiple mutations in the same family. The mechanism most likely to be responsible is termed germline mosaicism (mosaicism describes the presence of more than one genetically distinct cell line in the body). During the embryonic development of one of the parents, a mutation occurred that affected all or part of the germ line but few or none of the somatic cells of the embryo (Fig. 4-12). Thus, the parent carries the mutation in his or her germ line but does not actually express the disease because the mutation is absent in other cells of the body. As a result, the parent can transmit the mutation to multiple offspring. Although this phenomenon is relatively rare, it can have significant effects on recurrence risks when it does occur.

Figure 4.12 A mutation occurs in one cell of the developing embryo. All descendants of that cell have the same mutation, resulting in mosaicism. If the first mutated cell is part of the germline lineage, then germline mosaicism results. page 68 page 69

Germline mosaicism has been studied extensively in the lethal perinatal form of osteogenesis imperfecta (OI type II; see Chapter 2), which is caused by mutations in the type 1 procollagen genes. The fact that unaffected parents sometimes produced multiple offspring affected with

this disease led to the conclusion that type II OI was an autosomal recessive trait. This was disputed by studies in which the polymerase chain reaction (PCR) technique was used to amplify DNA from the sperm of a father of two children with type II OI. This DNA was compared with DNA extracted from his somatic cells (skin fibroblasts). Although procollagen mutations were not detected in the fibroblast DNA, they were found in approximately 1 of every 8 sperm cells. This was a direct demonstration of germline mosaicism in this individual. Although germline mosaicism has thus been demonstrated for type II OI, most non-inherited cases (approximately 95%) are thought to be caused by isolated new mutations, and a few cases of true autosomal recessive inheritance have also been documented. Other diseases in which germline mosaicism has been observed include achondroplasia, neurofibromatosis type 1, Duchenne muscular dystrophy, and hemophilia A. Germline mosaicism is relatively more common in the latter two diseases, both of which are discussed further in Chapter 5. It has been estimated that germline mosaicism accounts for up to 15% of Duchenne muscular dystrophy cases and 20% of hemophilia A cases in which there is no previous family history. ▪ Germline mosaicism occurs when all or part of a parent's germ line is affected by a disease mutation but the somatic cells are not. It elevates the recurrence risk for future offspring of the mosaic parent.

Variable Expression A similar complication is variable expression. Here, the penetrance may be complete, but the severity of the disease can vary greatly. A well-studied example of variable expression in an autosomal dominant disease is neurofibromatosis type 1, or von Recklinghausen disease (after the German physician who described the disorder in 1882). Clinical Commentary 4-4 provides further discussion of this disorder. A parent with mild expression of the disease-so mild that he or she is not aware of it-can transmit the gene to a child, who may have severe expression. As with reduced penetrance, variable expression provides a mechanism for disease genes to survive at higher frequencies in populations. www It should be emphasized that penetrance and expression are distinct entities. Penetrance is an all-or-none phenomenon: one either has the disease phenotype or does not. Variable expression refers to the extent of expression of the disease phenotype. The causes of variable expression usually are not known. Environmental effects can sometimes be responsible: in the absence of a certain environmental factor, the gene is expressed with diminished severity or not at all. Another possible cause is the interaction of other genes, called modifier genes, with the disease gene. An example of a human modifier gene is a locus on chromosome 19 that appears to influence whether meconium ileus develops in individuals with cystic fibrosis (see Clinical Commentary 4-1). Finally, as the molecular basis of mutation becomes better understood, it is clear that some cases of variable expression are caused by different types of mutations (i.e., different alleles) at the same disease locus. This is termed allelic heterogeneity. In some cases, clinically distinct diseases may be the result of allelic heterogeneity, as in the β-globin mutations that can cause either sickle cell disease or various β-thalassemias. Osteogenesis imperfecta is one disease in which genetic studies have helped to explain variable expression. Mutations that affect amino acids near the carboxyl terminal of the procollagen molecule generally cause more severe consequences than do mutations affecting the molecule near its amino terminal. It is also well documented that affected members of the same family, having the same mutation, can nevertheless manifest large differences in disease severity. This may be a consequence of different genetic "backgrounds" (i.e., modifier genes) in related individuals. And nongenetic events, such as an accidental bone fracture, can influence the severity of the disorder. Once a fracture occurs, casting and immobilization lead to a loss of bone mass, which further predisposes the patient to future fractures. Thus, a

chance environmental event (e.g., trauma leading to a fracture in a baby during delivery) can cause a significant increase in severity of expression. Osteogenesis imperfecta thus provides examples of each factor thought to influence variable expression: environmental events, modifier genes, and allelic heterogeneity. page 71 page 72

CLINICAL COMMENTARY 4.3 CLINICAL COMMENTARY 4.3 Huntington Disease Huntington disease (HD) affects approximately 1 in 20,000 persons of European descent. It is substantially less common among Japanese and Africans. The disorder usually presents between the ages of 30 and 50 years, although it has been observed as early as 2 years of age and as late as 80 years of age. HD is characterized by a progressive loss of motor control and by psychiatric problems, including dementia and affective disorder. There is a substantial loss of neurons in the brain, detectable by imaging techniques such as magnetic resonance imaging (MRI). Decreased glucose uptake in the brain, an early sign of the disorder, can be demonstrated by positron-emission tomography (PET). Although many parts of the brain are affected, the area most noticeably damaged is the corpus striatum. In some patients the disease leads to a loss of 25% or more of total brain weight (Fig. 4-16). The clinical course of HD is protracted. Typically, the interval from initial diagnosis to death is approximately 15 years. As in many neurological disorders, patients with HD experience difficulties in swallowing; aspiration pneumonia is the most common cause of death. Cardiorespiratory failure and subdural hematoma (due to head trauma) are other frequent causes of death. The suicide rate among HD patients is several times higher than that in the general population. Treatment includes the use of drugs such as benzodiazepines to help control the choreic movements. Affective disturbances, which are seen in nearly half of the patients, are sometimes controlled with antipsychotic drugs and tricyclic antidepressants. Although these drugs help to control some of the symptoms of HD, there is currently no way to alter the outcome of the disease. HD is notable in that affected homozygotes appear to display exactly the same clinical course as heterozygotes (in contrast to most dominant disorders, in which homozygotes are more severely affected). This attribute, and the fact that mouse models in which one copy of the gene is inactivated are perfectly normal, support the hypothesis that the mutation causes a gain of function (see Chapter 3). In more than 95% of cases, the mutation is inherited from an affected parent. HD has the distinction of being the first genetic disease mapped to a specific chromosome using an RFLP marker. James Gusella and colleagues mapped the disease gene to a region on the distal short arm of chromosome 4 in 1983. After 10 years of work by a large number of investigators, the disease gene was cloned. DNA sequence analysis showed that the mutation is a CAG expanded repeat (see Chapter 3) located within the coding portion of the gene. The normal repeat number ranges from 10 to 26. Individuals with 27 to 35 repeats are unaffected but are more likely to transmit a still larger number of repeats to their offspring. The inheritance of 36 or more copies of the repeat can produce disease, although incomplete penetrance of the disease phenotype is seen in those who have 36 to 41 repeats. As in many disorders caused by trinucleotide repeat expansion, a larger number of repeats is correlated with earlier age of onset of the disorder. Also, there is a tendency for greater repeat expansion when the father, rather than the mother, transmits the disease gene. This helps to explain the difference in ages of onset for maternally and paternally transmitted disease seen in Fig. 4-15. In particular, 80% of cases with onset before 20 years of age (termed "juvenile Huntington disease") are paternally transmitted, and these cases are accompanied by especially large repeat expansions. It remains to be seen why the degree of repeat instability in the HD

gene is greater in paternal transmission than in maternal transmission. Cloning of the HD gene led quickly to the identification of the gene product, huntingtin. This protein is involved in the transport of vesicles in cellular secretory pathways. In addition, there is evidence that huntingtin is necessary for the normal production of brainderived neurotrophic factor. The CAG repeat expansion produces a lengthened series of glutamine residues near huntingtin's amino terminal. Although the precise role of the expanded glutamine tract in disease causation is unclear, recent studies show that it leads to a buildup of toxic protein aggregates within and near neuronal nuclei. This buildup is associated with the early neuronal death that is characteristic of HD.

Figure 4.16 Cross section of the brain of an adult with Huntington disease, illustrating marked striatal atrophy. (Courtesy Dr. Jeanette Townsend, University of Utah Health Sciences Center.) page 72 page 73

CLINICAL COMMENTARY 4.4 Neurofibromatosis: A Disease with Highly Variable Expression Neurofibromatosis type 1 (NF1) (Fig. 4-17) is one of the most common autosomal dominant disorders, affecting approximately 1 in 3,000 individuals. It offers a good example of variable expression in a genetic disease. Some patients have only a few café-au-lait spots (from the French for "coffee with milk," describing the color of the hyperpigmented skin patches), Lisch nodules (benign growths on the iris), and perhaps neurofibromas (nonmalignant peripheral nerve tumors). These individuals are often unaware that they have the condition. Other patients have a much more severe expression of the disorder, including hundreds to thousands of neurofibromas, optic gliomas (benign tumors of the optic nerve), learning disabilities, hypertension, scoliosis (lateral curvature of the spine), and malignancies (e.g., malignant peripheral nerve sheath tumors, which can arise from plexiform neurofibromas). Fortunately, about two thirds of patients have only a mild cutaneous involvement. Fewer than 10% develop malignancies as a result of the disorder. Expression can vary significantly within the same family: a mildly affected parent can produce a severely affected offspring.

Figure 4.17 Neurofibromatosis type 1 (NF1). A, A young adult with multiple dermal neurofibromas of the trunk. A caféau-lait spot can be seen in the right upper abdomen. B, In a second patient with NF1, a large plexiform neurofibroma hangs from the lower right back, causing considerable inconvenience and discomfort. (The term plexiform is from the Latin plexus = "braid," describing the complex, tangled structure of these tumors). The tumor was surgically removed. Approximately 25% of NF1 patients develop plexiform neurofibromas. (B, Courtesy Dr. David Viskochil, University of Utah Health Sciences Center.)

A standard set of diagnostic criteria for NF1 has been developed. Two or more of the following must be present:

1. Six or more café-au-lait spots greater than 5 mm in diameter in prepubertal subjects and greater than 15 mm in postpubertal subjects 2. Freckling in the armpits or groin area 3. Two or more neurofibromas of any type or one plexiform neurofibroma (i.e., an extensive growth that occurs along a large nerve sheath) 4. Two or more Lisch nodules 5. Optic glioma 6. Distinctive bone lesions, particularly an abnormally formed sphenoid bone or tibial pseudarthrosis* 7. A first-degree relative diagnosed with neurofibromatosis using the previous six criteria page 73 page 74

Although NF1 has highly variable expression, the penetrance of this gene is virtually 100%. It has one of the highest known mutation rates, about 1 in 10,000 per generation. Approximately 50% of NF1 cases are the result of new mutations. In 1987 the gene was mapped to chromosome 17q by researchers in Salt Lake City, and it was isolated and cloned 3 years later.

It is a large gene, spanning approximately 350 kb of DNA. Its large size, which presents a sizable "target" for mutation, may help to account for the high mutation rate. The gene encodes a 13-kb mRNA transcript, and the gene product, termed neurofibromin, acts as a tumor suppressor (see Chapter 11). A mutation in the NF1 gene that occurs during embryonic development will affect only some cells of the individual, resulting in somatic mosaicism. In this case, the disease features may be confined to only one part of the body (segmental neurofibromatosis). Neurofibromatosis type 2 (NF2) is much rarer than NF1 and involves caféau-lait spots and bilateral acoustic neuromas (tumors affecting the eighth cranial nerve). It does not, however, involve true neurofibromas. The term "neurofibromatosis type 2" is thus a misnomer. The NF2 gene, which was mapped to chromosome 22, encodes a tumor suppressor protein called merlin or schwannomin. Mild cases of neurofibromatosis may involve very little clinical management. However, surgery may be required if malignancies develop or if benign tumors interfere with normal function. Scoliosis, tibial pseudarthrosis, and/or tibial bowing, seen in fewer than 5% of cases, may require orthopedic management. Hypertension may develop and is often secondary to a pheochromocytoma or a stenosis (narrowing) of the renal artery. The most common clinical problems in children are learning disabilities (seen in about 50% of individuals with NF1), short stature, and optic gliomas (which can lead to vision loss). Close follow-up can help to detect these problems and minimize their effects. ▪ Variable expression of a genetic disease may be caused by environmental effects, modifier genes, or allelic heterogeneity.

Reduced Penetrance

Figure 4.13 Pedigree illustrating the inheritance pattern of retinoblastoma, a disorder with reduced penetrance. The unaffected obligate carrier is lightly shaded, and affected individuals are heavily shaded.

Another important characteristic of many genetic diseases is reduced penetrance: an individual who has the genotype for a disease may not exhibit the disease phenotype at all, even though he or she can transmit the disease gene to the next generation. Retinoblastoma, a malignant eye tumor (Clinical Commentary 4-2), is a good example of an autosomal dominant disorder in which reduced penetrance is seen. The transmission pattern of this disorder is illustrated in Fig. 4-13. Family studies have shown that about 10% of the obligate

carriers of the retinoblastoma susceptibility gene (i.e., those who have an affected parent and affected children and therefore must themselves carry the gene) do not have the disease. The penetrance of the gene is then said to be 90%. Penetrance rates are usually estimated by examining a large number of families and determining what proportion of the obligate carriers (or obligate homozygotes, in the case of recessive disorders) develop the disease phenotype. www ▪ Reduced penetrance describes the situation in which individuals who have a disease-causing genotype do not develop the disease phenotype.

Anticipation and Repeat Expansion Since the early part of the 20th century, it has been observed that some genetic diseases seem to display an earlier age of onset and/or more severe expression in the more recent generations of a pedigree. This pattern is termed anticipation, and it has been the subject of considerable controversy and speculation. Many researchers believed that it was an artifact of better observation and clinical diagnosis in more recent times: a disorder that previously may have remained undiagnosed until age 60 years might now be diagnosed at age 40 simply because of better diagnostic tools. Others, however, believed that anticipation could be a real biological phenomenon, although evidence for the actual mechanism remained elusive.

Figure 4.21 A three-generation family affected with myotonic dystrophy. The degree of severity increases in each generation. The grandmother (right) is only slightly affected, but the mother (left) has a characteristic narrow face and somewhat limited facial expression. The baby is more severely affected and has the facial features of children with neonatal-onset myotonic dystrophy, including an open, triangle-shaped mouth. The infant has more than 1,000 copies of the trinucleotide repeat, whereas the mother and grandmother each have approximately 100 repeats.

Recently, molecular genetics has provided good evidence that anticipation does in fact have a biological basis. This evidence has come, in part, from studies of myotonic dystrophy, an autosomal dominant disease that involves progressive muscle deterioration (Fig. 4-21). Seen in approximately 1 in 8,000 individuals, myotonic dystrophy is the most common muscular dystrophy that affects adults. In addition to affecting skeletal muscles, this disorder produces cardiac arrhythmias (abnormal heart rhythms), testicular atrophy, and cataracts. The diseasecausing gene, which has been mapped to chromosome 19 and subsequently cloned, encodes a protein kinase. www page 80 page 81

Analysis of the gene has produced some interesting results. The disease mutation is an expanded CTG trinucleotide repeat (see Chapter 3) that lies in the 3' untranslated portion of the gene (i.e., a region transcribed into mRNA but not translated into protein). The number of these repeats is strongly correlated with severity of the disease. Unaffected individuals typically have 5 to 30 copies of the repeat. Those with 50 to 100 copies may be mildly affected or have no symptoms. Those with full-blown myotonic dystrophy have anywhere from 100 to several thousand copies of the repeat sequence. The number of repeats often increases with succeeding generations: a mildly affected parent with 80 repeats may produce a severely affected offspring who has more than 1,000 repeats (Fig. 4-22). Many families have now been documented in which the number of repeats increases through successive generations, accompanied by increasing severity of the disorder. There is thus strong evidence that expansion of this trinucleotide repeat is the cause of anticipation in myotonic dystrophy. Figure 4.22 A, Myotonic dystrophy pedigree illustrating anticipation. In this case, the age of onset for family members affected with an autosomal dominant disease is lower in more recent generations. B, An autoradiogram from a Southern blot analysis of the myotonic dystrophy gene in three individuals. Individual A is homozygous for a 4- to 5-repeat allele and is normal. Individual B has one normal allele and one disease allele of 175 repeats; this individual has myotonic dystrophy. Individual C is also affected with myotonic dystrophy and has one normal allele and a disease-causing allele of approximately 900 repeats. (B, Courtesy Dr. Kenneth Ward and Dr. Elaine Lyon, University of Utah Health Sciences Center.)

How does a mutation in the 3' untranslated portion of the gene produce the many disease features of myotonic dystrophy? Mouse models of this disease indicate that the expanded repeat decreases production of the protein product (a protein kinase), which results in cardiac conduction defects that produce arrhythmias. In addition, the mutation alters the mRNA transcript such that it remains in the nucleus and interacts with RNA-binding proteins to block their normal activity. This produces myotonic myopathy. Finally, the mutation may interfere (by altering chromatin structure) with a downstream transcription-factor gene, SIX5, resulting in cataract formation. Thus, analysis of the disease-causing mutation and its effect on nearby genes helps to explain the pleiotropy observed in myotonic dystrophy. page 81 page 82

Table 4-3. Disease Associated with Repeat Expansions Disease Description Repeat Normal Parent in Location sequence range, whom of abnormal expansion expansion range usually occurs Category 1 Huntington disease

Loss of motor CAG control, dementia, affective disorder

6-34; 36-100 More often Exon or more through father

Spinal and bulbar muscular atrophy

Adult-onset motor-neuron disease associated with androgen insensitivity

CAG

11-34; 40-62 More often Exon through father

Spinocerebellar ataxia type 1

Progressive ataxia, dysarthria, dysmetria

CAG

6-39; 41-81 More often Exon through father

Spinocerebellar ataxia type 2

Progressive CAG ataxia, dysarthria

15-29; 35-59 -

Spinocerebellar ataxia type 3 (Machado-Joseph disease)

Dystonia, distal muscular atrophy, ataxia, external

13-36; 68-79 More often Exon through father

CAG

Exon

ophthalmoplegia Spinocerebellar ataxia type 6

Progressive ataxia, dysarthria, nystagmus

CAG

Spinocerebellar ataxia type 7

Progressive CAG ataxia, dysarthria, retinal degeneration

4-16; 21-27 -

Exon

7-35; 38-200 More often through father

Spinocerebellar ataxia type 17 Progressive CAG ataxia, dementia, bradykinesia, dysmetria

29-42; 47-55 -

Dentatorubral-pallidoluysian atrophy/Haw River syndrome

7-25; 49-88 More often Exon through father

Cerebellar CAG atrophy, ataxia, myoclonic epilepsy, choreoathetosis, dementia

Exon

Category 2 Pseudoachondroplasia/multiple Short stature, epiphyseall dysplasia joint laxity, degenerative joint disease

GAC

5; 6-7

-

Exon

Oculopharyngeal muscular dystrophy

Proximal limb weakness, dysphagia, ptosis

GCG

6; 7-13

-

Exon

Cleidocranial dysplasia

Short stature, GCG, GCT, open skull GCA sutures with bulging calvaria, clavicular hypoplasia, shortened fingers, dental anomalies

17; 27 (expansion observed in one family)

Exon

Synpolydactyly

Polydactyly and GCG, GCT, 15; 22-25 syndactyly GCA

-

Exon

Category 3 Myotonic dystrophy (DM1; chromosome 19)

Muscle loss, CTG cardiac arrhythmia, cataracts, frontal balding

5-37; 100 to Either parent, 3' several but expansion untranslated thousand to congenital region form through mother

Myotonic dystrophy (DM2; chromosome 3)

Muscle loss, CCTG cardiac arrhythmia, cataracts, frontal balding

<75; 7511,000

Friedreich ataxia

Progressive limb GAA ataxia, dysarthria, hypertrophic cardiomyopathy, pyramidal weakness in legs

7-22; 200- Disorder is Intron 900 or more autosomal recessive, so disease alleles are inherited from both parents

Fragile X syndrome (FRAXA)

Mental CGG retardation, large ears and jaws, macroorchidism in males

6-52; 2002000 or more

Exclusively through mother

5' untranslated region

Fragile site (FRAXE)

Mild mental retardation

GCC

6-35; >200

More often through mother

5' untranslated region

Spinocerebellar ataxia type 8

Adult-onset ataxia, dysarthria,

CTG

16-37; 107- More often 127 through mother

3' untranslated region

-

3' untranslated region

nystagmus Spinocerebellar ataxia type 10 Ataxia and seizures

ATTCT

12-16; 800- More often Intron 4500 through father

Spinocerebellar ataxia type 12 Ataxia, eye movement disorders; variable age at onset

CAG

7-28; 66-78 -

Progressive myoclonic epilepsy Juvenile-onset type 1 convulsions, myoclonus, dementia

12-bp repeat 2-3; 30-75 motif

5' untranslated region

Autosomal 5' recessive untranslated inheritance so region transmitted by both parents page 82 page 83

Recently, a locus on chromosome 3 was discovered in which a 4-bp (CCTG) expanded repeat can also cause myotonic dystrophy. Again, the repeat is located in the 3' untranslated region of the gene. The phenotype associated with the chromosome 3 mutation is highly similar to that of the chromosome 19 mutation, although it sometimes is less severe. Myotonic dystrophy thus illustrates several important genetic principles: anticipation, pleiotropy, and locus heterogeneity. As discussed in Clinical Commentary 4-3, trinucleotide repeat expansion is also associated with anticipation in Huntington disease. It has also been observed in the fragile X syndrome, a leading genetic cause of mental retardation to be discussed in Chapter 5. Repeat expansions have now been identified as a cause of more than 20 genetic diseases (Table 4-3), and anticipation is observed in most of these diseases. As more repeat expansion diseases have been identified, some general patterns have begun to emerge. These diseases can be grouped into three categories, as indicated in Table 4-3. The first category consists of neurological diseases, such as Huntington disease and most of the spinocerebellar ataxias, that are caused by a CAG repeat expansion in a protein-coding portion of the gene. The repeats generally expand in number from a normal range of 10 to 35 to a disease-causing range of approximately 50 to 100. Expansions tend to be larger when transmitted through the father than through the mother, and the mutations have a gain-offunction effect. The second group consists of phenotypically more diverse diseases in which the expansions are again small in magnitude and are found in exons. The repeat sequence is heterogeneous, however, and anticipation is not a typical feature. The third category includes fragile X syndrome, myotonic dystrophy, two of the spinocerebellar ataxias, juvenile myoclonic epilepsy, and Friedreich ataxia. The repeat expansions are typically much larger than in the first two categories: the normal range is generally 5 to 50 trinucleotides, but the diseasecausing range can vary from 200 to several thousand trinucleotides. The repeats are located outside the protein-coding regions of the gene in all of these disorders, and the mutations usually have a loss-of-function effect. Repeat expansions are often larger when they are transmitted through the mother. Anticipation is seen in most of the diseases in categories 1 and 3. ▪ Anticipation refers to progressively earlier or more severe expression of a disease in more recent times. Expansion of DNA repeats has been shown to cause anticipation in some genetic diseases. These diseases can be divided into three major categories, depending on the size of the expansion, the location of the repeat, the phenotypic consequences of the expansion, the effect of the mutation, and the parent in whom large expansions typically occur.