What do the findings in epigenetics imply about the concept of heritability?

Some Misinterpretations

The magnitude of the heritability does not tell us a lot of things. For example, as it applies to individuals within populations, it cannot be used to predict genetic differences between races or other populations from phenotypic differences, whether or not they share the same environment.

The prediction formula R=h2S applies only (other than in very special circumstances) if selection is practiced on the trait on which response is measured. If selection is practiced on some trait or combination of traits other than the one of interest, the regression of response on selection differential is not therefore an unbiased estimate of heritability, but depends inter alia on the genetic and phenotypic correlations among the traits. This is a serious problem in inferences about selection in nature, where the actual selection applied is not known. Methods exist to overcome this problem, but require that records be available on all traits on which selection is practiced or to which fitness is related.

As the heritability is a summary parameter over loci, it does not tell us about either the numbers of genes that affect a quantitative trait or the magnitude of their effects. It is not therefore a constant as a population changes. But heritability is nevertheless a useful concept when properly used.

16.3.3 Heritability

Heritability is depending on the genetic variability related to the trait under study, it is then depending on the population under study. Its measurement is not indispensable to the interpretation of natural metric variation, but it can provide valuable information about the adaptiveness of metric traits. In insects, morphological traits commonly have the highest heritability values compared to other trait categories such as life history, probably because the former are less concerned with fitness.

Geometric techniques allow separate estimations of size and shape heritabilities. Size in insects may show consistent heritability values (Daly, 1992; Lehmann et al., 2006), so that they can be experimentally selected to constitute subpopulations genetically distinct for size (Anderson, 1973; Partridge et al., 1994). Various studies examining cross-environment heritability of wing shape in Diptera produced high and stable heritability, reaching 60% or more (Roff and Mousseau, 1987; Bitner-Mathé and Klaczko, 1999; Gilchrist and Partridge, 2001; Hoffman and Shirriffs, 2002). The consistent values of shape heritability suggest that a large fraction of morphometric divergence seen between natural populations of insects (Camara et al., 2006.; Henry et al., 2010; Morales et al., 2010) may be due to additive effects of genes.

In Ae. aegypti, shape appears to be more heritable than size. When comparing size and shape cross-environment heritability on the same populations in Ae. aegypti, much higher values for shape (Figure 16.2) than for size were found, providing indirect evidence for different genetic sources of variation (Morales et al., unpublished data).

Twin Studies.

The observation that a trait aggregates in families does not, in itself, establish that genes influence the phenotype. Traits and disorders may run in families for nongenetic reasons. For example, shared environmental experiences may produce the disorder in multiple family members. Twin and adoption studies can be used to separate the contribution of genetic and environmental causes of familial aggregation.

Twin studies compare concordance rates between monozygotic (MZ) twins (who are genetically identical) and dizygotic (DZ) twins (who share on average 50% of their alleles). A twin pair is concordant if both co-twins have the phenotype. If we can assume that environmental influences on MZ twins are not different from environmental influences on DZ twins (the “equal environments assumption”), then significantly higher concordance rates in MZ twins reflect the action of genes. Nevertheless, an MZ concordance rate that is less than 100% means that environmental factors influence the phenotype. Twin studies can provide an estimate of the heritability of the disorder, which refers to the proportion of the phenotypic differences among individuals in a population that can be attributed to genetic factors. The total variance in phenotypes (VP) in a population can be decomposed to a genetic component (VG) and an environmental component (VE): that is, VP = VG + VE. Thus, the heritability is the proportion of the total variance represented by the genetic variance: (VG/VP).

For quantitative traits, heritability can be estimated as 2(rMZ − rDZ) where rMZ refers to the co-twin phenotypic correlation for MZ twins and rDZ refers to the correlation for DZ twins. For categorical traits (such as diagnosis), the concordance rate can be substituted for these correlations to obtain a rough estimate of heritability. Several caveats are important to note regarding the interpretation of heritability estimates.

•

Heritability refers to the strength of genetic influences in a population, not a particular individual, and heritability estimates may differ depending on the population studied. A heritability of 60% says nothing about the contribution of genes to an individual's risk of a phenotype. Also, since the heritability is defined as the ratio VG/(VG + VE), as the environmental variance increases, the genetic component (heritability) decreases. Put another way, if the genetic homogeneity of a population increases, the apparent heritability of a trait will decrease.

Heritability refers to the additive sum of all genetic influences on a trait in a population. Thus, a heritability of .80 (80%) suggests that genes contribute more to trait variance in the population than does a heritability of 40%. However, heritability provides no information regarding how many genes are involved, how strong the effect of any given gene is, or how easy it will be to identify contributing genes. For example, the heritability of breast cancer has been estimated at 27%,25 substantially lower than the estimated heritability of bipolar disorder (80%), yet dominantly inherited single major genes (e.g., BRCA1) have been identified for breast cancer, whereas no single major genes have been found for schizophrenia. This reflects the fact that, while the penetrance of these breast cancer genes is very high in families that carry them, their contribution to the overall population variance is smaller than the contribution of potentially many genes of small effect that contribute to common forms of schizophrenia. The number and effect of genetic influences on a trait is sometimes referred to as the “genetic architecture.” Heritability does not resolve the genetic architecture of a trait, although a heritability greater than 0 does imply that genes play a role.

The magnitude of heritability is not a strong predictor of the potential impact of environmental interventions. A classic illustration is the case of phenylketonuria (PKU), a recessively inherited disorder due to a mutation in the gene encoding phenylalanine hydroxylase that results in a toxic accumulation of phenylalanine. Untreated, PKU can result in progressive brain damage with seizures and mental retardation. However, these devastating outcomes can be minimized by entirely environmental interventions: avoidance of dietary phenylalanine and supplementation with tyrosine.

Heritability Estimation

Heritability can be broadly defined as the proportion of phenotypic variability that is attributable to genetic factors; higher estimates suggest that genetic variability has a large influence on the variability of a given trait in the population. Heritability analysis has been used for decades to estimate whether a given phenotype is influenced by genetic factors and how strong that influence is relative to nongenetic risk factors. Numerous techniques exist for estimating heritability; these range from use of phenotype information from twins16 or family pedigree data17,18 to more recently developed statistical techniques for estimating heritability based on genome-wide genotyped data on unrelated individuals.19

Estimating the heritability of a given trait with use of twin or family data does not require specific measurement of genetic variants. Rather, these methods take advantage of the known shared genetic variance among related individuals. The general principle behind heritability analysis is that people who are more genetically related to each other should be more similar to each other for the phenotype of interest. For binary traits like sleep disorders, one can measure the recurrence risk to relatives. That is, given that a family member has been diagnosed with a disorder, what is the risk in the family members of having the same disorder? This recurrence risk in families can be compared with the risk of disease in the general population to give an estimate of heritability. For heritable traits, the relative recurrence risk ratio should decrease as the family relationships examined become less similar genetically; for example, the recurrence risk in siblings of affected individuals should be greater than for first cousins of the affected.

Results from family studies are complicated by the fact that family members often share a similar environment. It can be difficult to parse out whether the higher risk in certain families compared with the general population is due to shared genetic risk factors, shared environmental risk factors, or a combination of the two. Twin studies help to separate out shared genetic variance from other sources of variance because twin pairs are assumed to share many common environmental factors—they are born at the same time, shared the same intrauterine environment, and often attend the same school. With this source of variance controlled for, the similarities and differences between twins can be separated into genetic and environmental sources. Estimates of heritability are derived by comparisons between monozygotic and dizygotic twin pairs. Increased similarity in phenotype between monozygotic pairs (who are genetically identical) compared with dizygotic twin pairs (who share half of their genetic variants with each other) provides evidence for heritability. As presented in a recent publication establishing the heritability of performance deficit accumulation during sleep deprivation,20 there are several complementary methods that may be used for evaluating heritability in twin samples. We briefly describe these methods.

As discussed with respect to performance deficits during sleep deprivation,20 three methods for estimating heritability in twins are (1) classical heritability estimation, (2) analysis of variance (ANOVA) approach, and (3) likelihood-based estimation of variance components. Each of these methods can be useful in comparing the heritability to existing literature as well as in evaluating different assumptions. Classical heritability is derived using the differences in the intraclass correlation coefficient (ICC) statistics between monozygotic (ICCMZ) and dizygotic (ICCDZ) twin pairs.21 Using these values, heritability (denoted h2) is estimated as h2 = 2 • (ICCMZ − ICCDZ). In addition to estimating heritability, the classical approach can also provide an estimate of the shared common environmental variance, which is estimated as 2 • ICCDZ − ICCMZ. Next, the ANOVA approach uses combinations of the monozygotic and dizygotic within-twin and among-twin pair mean squares estimates in combination with specific assumptions about the variability (e.g., that total variability is equal in monozygotic and dizygotic twins).22,23 Finally, the maximum likelihood variance components approach uses model-specific covariance matrices23-25 and, importantly, allows the examination of specific patterns of genetic transmission and calculations of standard errors and P values associated with heritability estimates. These models of genetic transmission include components related to additive genetic effects (A), dominant genetic effects (D), common environmental effects (C), and unique individual effects (E).25 For example, the ACE model assumes additive genetic effects, shared environments, and unique individual components of variability. By comparison of different models, specific questions about the mode of genetic inheritance can be assessed. Overall, we see that each method of heritability estimation has unique advantages. Whereas the classical method provides a more simplistic approach to the computation, the advantage of the ANOVA model is the ability to assess specific assumptions about the validity of the twin model. Although it is potentially more complex, the maximum likelihood variance components approach can provide information on specific genetic transmission models.

As discussed later in the chapter, more recently established techniques allow estimation of heritability in unrelated individuals by simultaneously examining the association between a given trait and all genotyped genetic polymorphisms.19,26-28 These techniques have recently been used and extended to more accurately capture the amount of variability we can expect to explain through genome-wide association analyses.

Establishing that a given trait is heritable strongly implies that underlying genetic factors play a role in determining the phenotype. Many sleep-related disorders and intermediate phenotypes have been shown to be heritable in the last few decades, including sleep duration,29-31 chronotype,32-35 response to sleep loss,20 restless legs syndrome (RLS),36-38 insomnia,29,39-41 parasomnia,42 obstructive sleep apnea (OSA),43-49 and key intermediate traits for OSA (such as craniofacial structures,50 upper airway soft tissue volumes,51 and ventilatory responses to hypoxia and hypercapnia52). Among behavioral traits, one of the most heritable is the spectral characteristics of the electroencephalogram (EEG) during sleep.53

Heritability is only an estimate for the specific population included in a study. There is not one true heritability for a given disorder or trait. Instead, the heritability can vary over time as environments change, and it can be different in specific ethnic groups or in particular age groups (see Visscher et al19 for a review of heritability concepts). For instance, the heritability of sleep duration in adolescents is likely to be different from that in older adults. The relative importance of genes and environment in population variation may vary over the life span. Hence, estimates of heritability can vary substantially across studies.

Despite observing heritability estimates of more than 50% for some of these traits, the genetic variants discovered to date typically explain on the order of less than 5% of the known overall variability in any given phenotype. Finding the causes of this “missing heritability” is an ongoing area of research, and methods for determining heritability of a given phenotype continue to develop (for reviews, see1,2,54-56). Explanations for missing heritability are numerous, including a large number of common variants with small effects, multiple rare variants with large effects, insufficient tagging of causal variants in current genotyping platforms, gene-gene and gene-environment interactive effects, and other types of genetic variations (such as copy number variants and epigenetic modification). Ultimately, explaining the missing heritability is likely to require very large sample sizes and both rigorous and novel analytic techniques.

Behavioral Physical Functioning

Heritability estimates for physical strength in adulthood range from 30% to 60% for both measures of upper body strength such as hand grip and measures of lower body strength such as knee extension (Finkel et al., 2003; Frederiksen et al., 2002; Tiainen et al., 2004, 2005). In addition, heritability of these measures of physical strength appears to be stable across adulthood (McGue & Christensen, 2013). Beyond simple strength, physical function measures tasks that generally have more ecological validity, such as Activities of Daily Living (ADL) or behaviors such as walking, balance, and chair stands. Unlike the results for physical strength, however, heritability estimates for physical functioning tend to be mixed, with evidence for significant age and gender effects (Christensen, Frederiksen, Vaupel, & McGue, 2003; Christensen, Gaist, Vaupel, & McGue, 2002; Finkel, Pedersen, & Harris, 2000; Finkel et al., 2003). As variability in functional measures increases with age, heritability estimates also increase, but heritability tends to be higher for women than for men (Christensen et al., 2003; Finkel, Ernsth-Bravell, & Pedersen, 2013). This gender difference likely reflects both different susceptibility to chronic disabling conditions and differential life experiences.

Heritability and Cranial Nonmetric Traits

Heritability (h2) is the additive genetic variance divided by the phenotypic variance,

(5.1)h2=σG2σP2,

which essentially quantifies the genetic contribution to the expression of the trait. This is narrow sense heritability, only addressing the expression of a phenotype as determined by the genes (Falconer, 1989). Studies specifically addressing issues surrounding the heritability of these traits (writ large) are numerous (Berry, 1968, 1975; Berry et al., 1967; Cheverud and Buikstra, 1981a,b; Cheverud and Buikstra, 1982), although suitable samples have permitted only a few to address heritability in human samples (Carson, 2006; Lane, 1977; Sjøvold, 1984).

Sjøvold (1984) analyzed a pedigreed skeletal sample from Hallstatt, Austria, and found h2 values ranging from 0.0 to 0.954. More recently, Carson (2006) reexamined the same population as Sjøvold (1984) and calculated heritabilities using a maximum-likelihood variance components analysis on traits scored both dichotomously and on a multilevel system. Using this approach, she found uniformly low h2 values for the 36 traits considered. Of note, Carson (2006) also found that h2 values were higher when the dichotomous scoring system was used. Clearly, the findings of Sjøvold (1984) and Carson (2006) are concerning for researchers attempting to address biological distance based on cranial nonmetric trait data, since many analytical methods assume high heritability. Following the influential research of Cheverud (1988), many biological distance studies assume the genotypic (G) and phenotypic (P) distance matrices are similar and proportional. However, Martínez-Abadías et al. (2009) analysis of the heritability of human cranial dimensions found correlated G and P matrices reflecting genetic patterns, but the matrices were neither identical nor necessarily proportional.

A Heritability of Salt Sensitivity

Heritability analyses have consistently identified a significant contribution of genetic factors to salt sensitivity, documenting the BP response to sodium as a moderately heritable trait. Among the 1906 Han Chinese participants of the GenSalt dietary feeding study, heritabilities of salt sensitivity, defined as the percent changes in SBP, DBP, and MAP when switching from a low-sodium to high-sodium intervention, were 22%, 33%, and 33%, respectively.63 In a study of the BP response to a 12-week sodium-restricted diet in Caucasian families, heritable factors accounted for 64% of the SBP response.64 Finally, using an intravenous sodium-loading and furosemide volume-depletion protocol, Svetkey et al. examined 20 African-American families and calculated heritability estimates of 26–84% for MAP and 26–74% for SBP responses to the salt-sensitivity maneuver.65

Heritability

Heritability can be estimated by measuring individuals in families that are related to each other to some degree. Because relatives are expected to share a certain proportion of their genes, we can use this information to partition resemblance among relatives into genetic and environmental causal factors. The precision of heritability estimates is influenced by the type of relatives under consideration.

Formal crossing designs, such as combinations of full- and half-siblings or parents and their offspring, allows additive genetic variation to be partitioned from other effects (see Lynch and Walsh, 1998 for a discussion of design options). Natural family structure in wild populations that has been observed over several generations can also be used to build a pedigreed and estimate heritability (Kruuk, 2004; Shaw, 1987). The use of either experimental or natural pedigrees permits a more precise measurement which is referred to as ‘narrow-sense heritability’ (h2):

h2=VA/(VA+VD+VI+VE)

Although narrow-sense heritabilities are preferable, they are more difficult to obtain because some type of pedigree information for the offspring is required. The most common approaches to obtaining narrow-sense estimates include: (1) measuring traits in parents and offspring in the wild or in lab populations over several generations (e.g., Réale et al., 2003), (2) measuring offspring that were produced by experimental matings between known parents according to a specific design (e.g., Etterson and Shaw, 2001), or (3) from response to artificial selection measured over several generations (e.g., Lenski, 2001).

If replicated genotypes (e.g., clones) or broods obtained from natural matings (e.g., nestlings or seed collected by maternal plant) are measured, a coarser heritability estimate, referred to as ‘broad-sense heritability’ (H2), can be estimated. Because clones and siblings share parents, they are also expected to share some variance arising from dominance and epistatic interactions. These effects, however, are not transmitted to subsequent generations and therefore will not generally contribute to evolutionary change. Consequently, broad-sense heritability is a coarser upper-bound estimate of heritability that is confounded with dominance variance:

What does epigenetics tell us about heritability?

Are epigenetic effects believed to be heritable?

What is epigenetics and why is it important for understanding how traits are inherited?