Accurate clinical genetic testing for autoinflammatory diseases using the next-generation sequencing platform MiSeq

Autoinflammatory diseases occupy one of a group of primary immunodeficiency diseases that are generally thought to be caused by mutation of genes responsible for innate immunity, rather than by acquired immunity. Mutations related to autoinflammatory diseases occur in 12 genes. For example, low-level somatic mosaic NLRP3 mutations underlie chronic infantile neurologic, cutaneous, articular syndrome (CINCA), also known as neonatal-onset multisystem inflammatory disease (NOMID). In current clinical practice, clinical genetic testing plays an important role in providing patients with quick, definite diagnoses. To increase the availability of such testing, low-cost high-throughput gene-analysis systems are required, ones that not only have the sensitivity to detect even low-level somatic mosaic mutations, but also can operate simply in a clinical setting. To this end, we developed a simple method that employs two-step tailed PCR and an NGS system, MiSeq platform, to detect mutations in all coding exons of the 12 genes responsible for autoinflammatory diseases. Using this amplicon sequencing system, we amplified a total of 234 amplicons derived from the 12 genes with multiplex PCR. This was done simultaneously and in one test tube. Each sample was distinguished by an index sequence of second PCR primers following PCR amplification. With our procedure and tips for reducing PCR amplification bias, we were able to analyze 12 genes from 25 clinical samples in one MiSeq run. Moreover, with the certified primers designed by our short program—which detects and avoids common SNPs in gene-specific PCR primers—we used this system for routine genetic testing. Our optimized procedure uses a simple protocol, which can easily be followed by virtually any office medical staff. Because of the small PCR amplification bias, we can analyze simultaneously several clinical DNA samples with low cost and can obtain sufficient read numbers to detect a low level of somatic mosaic mutations.     

Absence of somatic mosaicism in 17 families with hemophilia B: an analysis with a sensitivity 10- to 1000-fold greater than that of sequencing gels

Most estimates of germ-line mosaicism have been derived from families in which there has been transmission of a mutated allele to two or more children by an unaffected individual. Previously, analyses for somatic mosaicism detected five such individuals by PCR-based sequencing and haplotype analysis at a sensitivity of approximately 1 mutant per 10 wild-type alleles. To determine whether mutations that occur later in embryogenesis also give rise to somatic mosaicism, we analyzed leukocyte DNA from 17 individuals in whom a mutation in the factor IX gene was known to have originated. Methods capable of detecting 1 mutant allele in 100–10 000 were utilized, and no further examples of somatic mosaicism were detected. If confirmed by future studies, the paucity of somatic mosaicism with mutant:wild-type allele frequencies ranging from 1:10 to 1:1000 (relative to the 11% of somatic mosaicism detected with mutant:wild-type allele frequencies of 1:1 to 1:10) may reflect a higher mutation rate and/or germ-line lineage allocation very early in embryogenesis. Received: 14 July 1995 / Revised: 1 April 1996     

Constitutively activated NLRP3 inflammasome causes inflammation and abnormal skeletal development in mice

The NLRP3 inflammasome complex is responsible for maturation of the pro-inflammatory cytokine, IL-1β. Mutations in NLRP3 are responsible for the cryopyrinopathies, a spectrum of conditions including neonatal-onset multisystem inflammatory disease (NOMID). While excessive production of IL-1β and systemic inflammation are common to all cryopyrinopathy disorders, skeletal abnormalities, prominently in the knees, and low bone mass are unique features of patients with NOMID. To gain insights into the mechanisms underlying skeletal abnormalities in NOMID, we generated knock-in mice globally expressing the D301N NLRP3 mutation (ortholog of D303N in human NLRP3). NOMID mice exhibit neutrophilia in blood and many tissues, including knee joints, and high levels of serum inflammatory mediators. They also exhibit growth retardation and severe postnatal osteopenia stemming at least in part from abnormally accelerated bone resorption, attended by increased osteoclastogenesis. Histologic analysis of knee joints revealed abnormal growth plates, with loss of chondrocytes and growth arrest in the central region of the epiphyses. Most strikingly, a tissue "spike" was observed in the mid-region of the growth plate in the long bones of all NOMID mice that may be the precursor to more severe deformations analogous to those observed in NOMID patients. These findings provide direct evidence linking a NOMID-associated NLRP3-activating mutation to abnormalities of postnatal skeletal growth and bone remodeling.     

Parental Somatic Mosaicism Is Underrecognized and Influences Recurrence Risk of Genomic Disorders

New human mutations are thought to originate in germ cells, thus making a recurrence of the same mutation in a sibling exceedingly rare. However, increasing sensitivity of genomic technologies has anecdotally revealed mosaicism for mutations in somatic tissues of apparently healthy parents. Such somatically mosaic parents might also have germline mosaicism that can potentially cause unexpected intergenerational recurrences. Here, we show that somatic mosaicism for transmitted mutations among parents of children with simplex genetic disease is more common than currently appreciated. Using the sensitivity of individual-specific breakpoint PCR, we prospectively screened 100 families with children affected by genomic disorders due to rare deletion copy-number variants (CNVs) determined to be de novo by clinical analysis of parental DNA. Surprisingly, we identified four cases of low-level somatic mosaicism for the transmitted CNV in DNA isolated from parental blood. Integrated probabilistic modeling of gametogenesis developed in response to our observations predicts that mutations in parental blood increase recurrence risk substantially more than parental mutations confined to the germline. Moreover, despite the fact that maternally transmitted mutations are the minority of alleles, our model suggests that sexual dimorphisms in gametogenesis result in a greater proportion of somatically mosaic transmitting mothers who are thus at increased risk of recurrence. Therefore, somatic mosaicism together with sexual differences in gametogenesis might explain a considerable fraction of unexpected recurrences of X-linked recessive disease. Overall, our results underscore an important role for somatic mosaicism and mitotic replicative mutational mechanisms in transmission genetics.     

Temperature modulation of DHPLC analysis for detection of coexisting constitutional and mosaic sequence variants in TSC2

Somatic mosaicism is a frequent phenomenon in Mendelian disorders that exhibit a high proportion of new mutations. However, mutant alleles present at low frequency may escape detection. We have previously shown that denaturing high-performance liquid chromatography (DHPLC) at the recommended melt temperature can detect TSC1 and TSC2 mutations in tuberous sclerosis patients with low-level somatic mosaicism, even when direct sequencing cannot identify the causative lesion. Here, we report the use of temperature modulation in DHPLC analysis to facilitate the robust detection of a mosaic mutation, N1643K, in the presence of a coexisting constitutional polymorphism.     

Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles

The exploration of copy-number variation (CNV), notably of somatic cells, is an understudied aspect of genome biology. Any differences in the genetic makeup between twins derived from the same zygote represent an irrefutable example of somatic mosaicism. We studied 19 pairs of monozygotic twins with either concordant or discordant phenotype by using two platforms for genome-wide CNV analyses and showed that CNVs exist within pairs in both groups. These findings have an impact on our views of genotypic and phenotypic diversity in monozygotic twins and suggest that CNV analysis in phenotypically discordant monozygotic twins may provide a powerful tool for identifying disease-predisposition loci. Our results also imply that caution should be exercised when interpreting disease causality of de novo CNVs found in patients based on analysis of a single tissue in routine disease-related DNA diagnostics.     

Somatic mosaicism in healthy human tissues

From the fertilization of an egg until the death of an individual, somatic cells can accumulate genetic changes, such that cells from different tissues or even within the same tissue differ genetically. The presence of multiple cell clones with distinct genotypes in the same individual is referred to as 'somatic mosaicism'. Many endogenous factors such as mobile elements, DNA polymerase slippage, DNA double-strand break, inefficient DNA repair, unbalanced chromosomal segregation and some exogenous factors such as nicotine and UV exposure can contribute to the generation of somatic mutations, thereby leading to somatic mosaicism. Such changes can potentially affect the epigenetic patterns and levels of gene expression, and ultimately the phenotypes of cells. Although recent studies suggest that somatic mosaicism is widespread during normal development and aging, its implications for heightened disease risks are incompletely understood. Here, I discuss the origins, prevalence and implications of somatic mosaicism in healthy human tissues.     

Monozygotic twins discordant for constitutive BRCA1 promoter methylation, childhood cancer and secondary cancer

We describe monozygotic twins discordant for childhood leukemia and secondary thyroid carcinoma. We used bisulfite pyrosequencing to compare the constitutive promoter methylation of BRCA1 and several other tumor suppressor genes in primary fibroblasts. The affected twin displayed an increased BRCA1 methylation (12%), compared with her sister (3%). Subsequent bisulfite plasmid sequencing demonstrated that 13% (6 of 47) BRCA1 alleles were fully methylated in the affected twin, whereas her sister displayed only single CpG errors without functional implications. This between-twin methylation difference was also found in irradiated fibroblasts and untreated saliva cells. The BRCA1 epimutation may have originated by an early somatic event in the affected twin: approximately 25% of her body cells derived from different embryonic cell lineages carry one epigenetically inactivated BRCA1 allele. This epimutation was associated with reduced basal protein levels and a higher induction of BRCA1 after DNA damage. In addition, we performed a genome-wide microarray analysis of both sisters and found several copy number variations, i.e., heterozygous deletion and reduced expression of the RSPO3 gene in the affected twin. This monozygotic twin pair represents an impressive example of epigenetic somatic mosaicism, suggesting a role for constitutive epimutations, maybe along with de novo genetic alterations in recurrent tumor development.     

Intracolonial genetic variation in the scleractinian coral Seriatopora hystrix

In recent years, increasing numbers of studies revealed intraorganismal genetic variation, primarily in modular organisms like plants or colonial marine invertebrates. Two underlying mechanisms are distinguished: Mosaicism is caused by somatic mutation, whereas chimerism originates from allogeneic fusion. We investigated the occurrence of intracolonial genetic variation at microsatellite loci in five natural populations of the scleractinian coral Seriatopora hystrix on the Great Barrier Reef. This coral is a widely distributed, brooding species that is at present a target of intensive population genetic research on reproduction and dispersal patterns. From each of 155 S. hystrix colonies, either two or three samples were genotyped at five or six loci. Twenty-seven (~17%) genetically heterogeneous colonies were found. Statistical analyses indicated the occurrence of both mosaicism and chimerism. In most cases, intracolonial variation was found only at a single allele. Our analyses suggest that somatic mutations present a major source of genetic heterogeneity within a single colony. Moreover, we observed large, apparently stable chimeric colonies that harbored clearly distinct genotypes and contrast these findings with the patterns typically observed in laboratory-based experiments. We discuss the error that mosaicism and chimerism introduce into population genetic analyses.     

Two-stage designs in case-control association analysis

DNA pooling is a cost-effective approach for collecting information on marker allele frequency in genetic studies. It is often suggested as a screening tool to identify a subset of candidate markers from a very large number of markers to be followed up by more accurate and informative individual genotyping. In this article, we investigate several statistical properties and design issues related to this two-stage design, including the selection of the candidate markers for second-stage analysis, statistical power of this design, and the probability that truly disease-associated markers are ranked among the top after second-stage analysis. We have derived analytical results on the proportion of markers to be selected for second-stage analysis. For example, to detect disease-associated markers with an allele frequency difference of 0.05 between the cases and controls through an initial sample of 1000 cases and 1000 controls, our results suggest that when the measurement errors are small (0.005), approximately 3% of the markers should be selected. For the statistical power to identify disease-associated markers, we find that the measurement errors associated with DNA pooling have little effect on its power. This is in contrast to the one-stage pooling scheme where measurement errors may have large effect on statistical power. As for the probability that the disease-associated markers are ranked among the top in the second stage, we show that there is a high probability that at least one disease-associated marker is ranked among the top when the allele frequency differences between the cases and controls are not <0.05 for reasonably large sample sizes, even though the errors associated with DNA pooling in the first stage are not small. Therefore, the two-stage design with DNA pooling as a screening tool offers an efficient strategy in genomewide association studies, even when the measurement errors associated with DNA pooling are nonnegligible. For any disease model, we find that all the statistical results essentially depend on the population allele frequency and the allele frequency differences between the cases and controls at the disease-associated markers. The general conclusions hold whether the second stage uses an entirely independent sample or includes both the samples used in the first stage and an independent set of samples.     

A systematic study of mosaicism in heritable thoracic aortic aneurysm and dissection

Mosaicisms are often overlooked in routine molecular diagnosis. Although not common, they are of great significance for accurate diagnosis and genetic counseling. In this study, we systematically evaluated the frequency of mosaicisms in both asymptomatic parents and affected patients with thoracic aortic aneurysm and dissection (TAAD). Next-generation sequencing (NGS) data from 1085 patients was reanalyzed with a more lenient allele frequency to detect potential mosaic variants. In addition, parental mosaicisms were investigated in 80 TAAD families. Finally, a total of six mosaic variants were detected in our cohort. Three of them were identified in symptomatic patients and three were in asymptomatic parents. Notably, a low-level mosaic variant in TGFB2 was detected combined with a causative FBN1 variant in patient AD2001, which might partially explain the clinical heterogeneity in his family. Our study hinted that it is necessary and feasible to implement mosaicism analysis in routine molecular diagnosis.     

Post-zygotic Point Mutations Are an Underrecognized Source of De Novo Genomic Variation

De novo mutations are recognized both as an important source of genetic variation and as a prominent cause of sporadic disease in humans. Mutations identified as de novo are generally assumed to have occurred during gametogenesis and, consequently, to be present as germline events in an individual. Because Sanger sequencing does not provide the sensitivity to reliably distinguish somatic from germline mutations, the proportion of de novo mutations that occur somatically rather than in the germline remains largely unknown. To determine the contribution of post-zygotic events to de novo mutations, we analyzed a set of 107 de novo mutations in 50 parent-offspring trios. Using four different sequencing techniques, we found that 7 (6.5%) of these presumed germline de novo mutations were in fact present as mosaic mutations in the blood of the offspring and were therefore likely to have occurred post-zygotically. Furthermore, genome-wide analysis of “de novo” variants in the proband led to the identification of 4/4,081 variants that were also detectable in the blood of one of the parents, implying parental mosaicism as the origin of these variants. Thus, our results show that an important fraction of de novo mutations presumed to be germline in fact occurred either post-zygotically in the offspring or were inherited as a consequence of low-level mosaicism in one of the parents.     

X-linked Menkes disease: first documented report of germ-line mosaicism

This work investigated a three-generation Menkes disease family, where germ-line mosaicism was suspected in the maternal grandmother of the index patient. She had given birth to 2 boys who died of suspected Menkes disease on the basis of clinical and photographic evidence. Biochemical analysis of the index patient confirmed the diagnosis of Menkes disease, and DNA analysis established a partial gene deletion (EX11_EX23del), involving exons 11-23 and the 3'-untranslated region (UTR) of ATP7A. A junction fragment was detectable by Southern blot analysis, which enabled carrier analysis. The mother was demonstrated to be a carrier, whereas analysis of lymphoblasts and skin fibroblasts from the maternal grandmother gave no indication of a partial gene deletion. No materials were available from the possibly affected maternal uncles. Further genetic analyses, including biochemical testing of the grandmother and haplotype analysis using four intragenic markers on DNA from selected members of the family, corroborated this finding. The combined results from DNA analyses showed that the grandmother had transmitted three different ATP7A haplotypes to her offspring: (1) the at-risk allele (CA(B))-1 and the deletion; (2) the at-risk allele (CA(B))-1 without deletion; and (3) the second allele (CAB)-2 without deletion. In conclusion, our study demonstrated segregation of Menkes disease within the family investigated that can best be explained by extensive germ-line mosaicism in the maternal grandmother. The finding of germ-line mosaicism has obvious implications for genetic counseling of Menkes disease families.     

Selection and mutation in the “new” genetics: an emerging hypothesis

It has been anticipated that new, much more sensitive, next generation sequencing (NGS) techniques, using massively parallel sequencing, will likely provide radical insights into the genetics of multifactorial diseases. While NGS has been used initially to analyze individual human genomes, and has revealed considerable differences between healthy individuals, we have used NGS to examine genetic variation within individuals, by sequencing tissues “in depth”, i.e., oversequencing many thousands of times. Initial studies have revealed intra-tissue genetic heterogeneity, in the form of multiple variants of a single gene that exist as distinct “majority and “minority” variants. This highly specialized form of somatic mosaicism has been found within both cancer and normal tissues. If such genetic variation within individual tissues is widespread, it will need to be considered as a significant factor in the ontogeny of many multifactorial diseases, including cancer. The discovery of majority and minority gene variants and the resulting somatic cell heterogeneity in both normal and diseased tissues suggests that selection, as opposed to mutation, might be the critical event in disease ontogeny. We, therefore, are proposing a hypothesis to explain multifactorial disease ontogeny in which pre-existing multiple somatic gene variants, which may arise at a very early stage of tissue development, are eventually selected due to changes in tissue microenvironments.     

Equal proportions of affected cells in muscle and blood of a mosaic carrier of facioscapulohumeral muscular dystrophy

Autosomal dominant facioscapulohumeral muscular dystrophy (FSHD) is associated with contractions of D4Z4 repeat on 4q35. It displays a remarkable inter- and intra-familial clinical variability ranging from severe phenotype to asymptomatic carriers. Mosaicism for the contracted FSHD-sized allele is a recurrent finding, but only DNA from lymphocytes had been studied. It is currently not known if mosaicism is unequally distributed between different tissues and if muscle is relatively spared for the presence of the disease allele in mosaic asymptomatic carriers of a disease allele. Here we compare DNA extracted from peripheral blood lymphocytes (PBL), fibroblasts and muscle from a mosaic asymptomatic female carrier and mother of a FSHD patient. PFGE analysis showed a complex allelic segregation: two independent mitotic rearrangement episodes occurred, resulting in mosaicism for a contracted D4Z4 repeat on 4q35 in the mother and mosaicism for an expanded D4Z4 repeat on 10q26 in the affected daughter. The results show that the proportion of mosaicism in PBL and muscle were comparable, while in fibroblasts there was some variation in the mosaicism, which might be caused by culturing artefacts. This finding supports the hypothesis that a mitotic contraction of D4Z4 is an early embryonic event and indicates that the degree of mosaicism in PBL is representative for that of muscle.     

A polymorphism in the MSH3 mismatch repair gene is associated with the levels of somatic instability of the expanded CTG repeat in the blood DNA of myotonic dystrophy type 1 patients

Somatic mosaicism of the expanded CTG repeat in myotonic dystrophy type 1 is age-dependent, tissue-specific and expansion-biased, contributing toward the tissue-specificity and progressive nature of the symptoms. Previously, using regression modelling of repeat instability we showed that variation in the rate of somatic expansion in blood DNA contributes toward variation in age of onset, directly implicating somatic expansion in the disease pathway. Here, we confirm these results using a larger more genetically homogenous Costa Rican DM1 cohort (p < 0.001). Interestingly, we also provide evidence that supports subtle sex-dependent differences in repeat length-dependent age at onset and somatic mutational dynamics. Previously, we demonstrated that variation in the rate of somatic expansion was a heritable quantitative trait. Given the important role that DNA mismatch repair genes play in mediating expansions in mouse models, we tested for modifier gene effects with 13 DNA mismatch gene polymorphisms (one each in MSH2, PMS2, MSH6 and MLH1; and nine in MSH3). After correcting for allele length and age effects, we identified three polymorphisms in MSH3 that were associated with variation in somatic instability: Rs26279 (p = 0.003); Rs1677658 (p = 0.009); and Rs10168 (p = 0.031). However, only the association with Rs26279 remained significant after multiple testing correction. Although we revealed a statistically significant association between Rs26279 and somatic instability, we did not detect an association with the age at onset. Individuals with the A/A genotype for Rs26279 tended to show a greater propensity to expand the CTG repeat than other genotypes. Interestingly, this SNP results in an amino acid change in the critical ATPase domain of MSH3 and is potentially functionally dimorphic. These data suggest that MSH3 is a key player in generating somatic variation in DM1 patients and further highlight MSH3 as a potential therapeutic target.     

Frequency of somatic and germ-line mosaicism in retinoblastoma: implications for genetic counseling.

Although mosaicism can have important implications for genetic counseling of families with hereditary disorders, information regarding the incidence of mosaicism is available for only a few genetic diseases. Here we describe an evaluation of 156 families with retinoblastoma; the initial oncogenic mutation in the retinoblastoma gene had been identified in these families. In 15 ( approximately 10%) families, we were able to document mosaicism for the initial mutation in the retinoblastoma gene, either in the proband or in one of the proband's parents. The true incidence of mosaicism in this group of 156 families is probably higher than our findings indicate; in some additional families beyond the 15 we identified, mosaicism was likely but could not be proven, because somatic or germ-line DNA from key family members was unavailable. Germ-line DNA from two mosaic fathers was analyzed: in one of these, the mutation was detected in both sperm and leukocyte DNA; in the other, the mutation was detected only in sperm DNA. Our data suggest that mosaicism is more common than is generally appreciated, especially in disorders such as retinoblastoma, in which a high proportion of cases represent new mutations. The possibility of mosaicism should always be considered during the genetic counseling of newly identified families with retinoblastoma. As demonstrated here, genetic tests of germ-line DNA can provide valuable information that is not available through analysis of somatic (leukocyte) DNA.     

Germ-line mosaicism in tuberous sclerosis: how common?

Two-thirds of cases of tuberous sclerosis complex (TSC) are sporadic and usually are attributed to new mutations, but unaffected parents sometimes have more than one affected child. We sought to determine how many of these cases represent germ-line mosaicism, as has been reported for other genetic diseases. In our sample of 120 families with TSC, 7 families had two affected children and clinically unaffected parents. These families were tested for mutations in the TSC1 and TSC2 genes, by Southern blotting and by single-strand conformational analysis. Unique variants were detected in six families. Each variant was present and identical in both affected children of a family but was absent in both parents and the unaffected siblings. Sequencing of the variants yielded two frameshift mutations, one missense mutation, and two nonsense mutations in TSC2 and one nonsense mutation in TSC1. To determine which parent contributed the affected gametes, the families were analyzed for linkage to TSC1 and TSC2, by construction of haplotypes with markers flanking the two genes. Linkage analysis and loss-of-heterozygosity studies indicated maternal origin in three families, paternal origin in one family, and either being possible in two families. To evaluate the possibility of low-level somatic mosaicism for TSC, DNA from lymphocytes of members of the six families were tested by allele-specific PCR. In all the families, the mutant allele was detected only in the known affected individuals. We conclude that germ-line mosaicism was present in five families with mutations in the TSC2 gene and in one family with the causative mutation in the TSC1 gene. The results have implications for genetic counseling of families with seemingly sporadic TSC.     

Somatic mosaicism for copy-neutral loss of heterozygosity and DNA copy number variations in the human genome

Background

Somatic mosaicism denotes the presence of genetically distinct populations of somatic cells in one individual who has developed from a single fertilised oocyte. Mosaicism may result from a mutation that occurs during postzygotic development and is propagated to only a subset of the adult cells. Our aim was to investigate both somatic mosaicism for copy-neutral loss of heterozygosity (cn-LOH) events and DNA copy number variations (CNVs) in fully differentiated tissues.

Results

We studied panels of tissue samples (11–12 tissues per individual) from four autopsy subjects using high-resolution Illumina HumanOmniExpress-12 BeadChips to reveal the presence of possible intra-individual tissue-specific cn-LOH and CNV patterns.We detected five mosaic cn-LOH regions >5 Mb in some tissue samples in three out of four individuals. We also detected three CNVs that affected only a portion of the tissues studied in one out of four individuals. These three somatic CNVs range from 123 to 796 kb and are also found in the general population. An attempt was made to explain the succession of genomic events that led to the observed somatic genetic mosaicism under the assumption that the specific mosaic patterns of CNV and cn-LOH changes reflect their formation during the postzygotic embryonic development of germinal layers and organ systems.

Conclusions

Our results give further support to the idea that somatic mosaicism for CNVs, and also cn-LOHs, is a common phenomenon in phenotypically normal humans. Thus, the examination of only a single tissue might not provide enough information to diagnose potentially deleterious CNVs within an individual. During routine CNV and cn-LOH analysis, DNA derived from a buccal swab can be used in addition to blood DNA to get information about the CNV/cn-LOH content in tissues of both mesodermal and ectodermal origin. Currently, the real frequency and possible phenotypic consequences of both CNVs and cn-LOHs that display somatic mosaicism remain largely unknown. To answer these questions, future studies should involve larger cohorts of individuals and a range of tissues.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-1916-3) contains supplementary material, which is available to authorized users.     

Mechanisms and consequences of somatic mosaicism in humans

Somatic mosaicism -- the presence of genetically distinct populations of somatic cells in a given organism -- is frequently masked, but it can also result in major phenotypic changes and reveal the expression of otherwise lethal genetic mutations. Mosaicism can be caused by DNA mutations, epigenetic alterations of DNA, chromosomal abnormalities and the spontaneous reversion of inherited mutations. In this review, we discuss the human disorders that result from somatic mosaicism, as well as the molecular genetic mechanisms by which they arise. Specifically, we emphasize the role of selection in the phenotypic manifestations of mosaicism.