Quantitative Genetics MCB 140 12 8 06 1

Quantitative Genetics MCB 140, 12 -8 -06 1

A loose distinction “Qualitative” traits: • Blood groups (ABO) • Coat color in cats • Color vision Quantitative traits: • Height • Weight • Facial features Difference between phenotypes of two individuals can be explained by difference in genotype at a small number of loci (for example, 1 or 2). Mendelian ratios in F 1 The phenotype is determined – to some extent – by genotype, but phenotypic difference between individuals is due to difference in genotype at a large number of loci. No Mendelian ratios in F 1 MCB 140, 12 -8 -06 2

Gene for starting businesses “If you belong to a certain extended family in Seattle, you're probably an entrepreneur. It seems to be about the only career many of the members ever considered. ''It's in our blood'' said Brian Jacobsen, president of Madison Park Greetings, a stationery and gifts company. Mr. Jacobsen's brother, mother, grandfather, two uncles, two cousins and an aunt all started and ran their own companies and say they cannot imagine any other livelihood. Why are so many people in the same clan hooked? Some of them have a theory. They believe that somewhere in their chromosomes lurks an actual entrepreneurial gene -that their bent for business really is in their blood. ” New York Times, Nov. 20, 2003 – p. C 8 MCB 140, 12 -8 -06 3

New York Times, Nov. 20, 2003 – p. C 8 MCB 140, 12 -8 -06 4

Gene for metaphors “AG: Many of your songs include clear, visual images. Do these images come from dreams? Suzanne Vega: My mind works in a metaphorical way. It’s easier for me to say what I see than what I feel. The emotions are expressed in the images. I think it must be genetic, because my daughter, Ruby, thinks the same way. She’ll see smoke coming out of the back end of a car and say, "The smoke is tapdancing. " And if you look at it, you can see what she means. http: //www. acousticguitar. com/issues/ag 110/feature 110. html MCB 140, 12 -8 -06 5

The God Gene “Modern science is turning up a possible reason why the religious right is flourishing and secular liberals aren’t: instinct. It turns out that our DNA may predispose humans towards religious faith. … Dean Hamer, a prominent American geneticist, even identifies a particular gene, VMAT 2, that he says may be involved. People with one variant of this gene tend to be more spiritual, he found. ” N. Kristof, New York Times, 2 -12 -05 MCB 140, 12 -8 -06 6

“Nature vs. nurture”: the curse of “folk wisdom” Is a given human trait in a given person the result of “genes or the environment”? • “She got her brains from her Dad” • “Crime runs in his family – it’s genetic…” • “All Klingons are bellicose” (“Blood tells”) • “… is genetically predisposed to …” MCB 140, 12 -8 -06 7

In a population, phenotypes of individuals for a quantitative trait tend to be normally distributed MCB 140, 12 -8 -06 8

Central limit theorem Carl Friedrich Gauss If a variable is the sum of many independent variables, then its distribution will be normal: MCB 140, 12 -8 -06 9

In this example, a trait (color) is controlled by three loci, A, B, and C, each of which occurs in only 2 alleles. The actions of the alleles is somewhat additive, in other words, an aabbcc organism is all white, whereas an AABBCC organism is dark red. The remarkable thing is, even with such a simple system (three loci with two alleles each!), we can get a remarkably “smooth” distribution of phenotypes! MCB 140, 12 -8 -06 10

Crisper distinction between Mendelian and quantitative traits: For a quantitative trait, the range of phenotypes exhibited by individuals in any given genotypic class is BROADER than the difference between two average individuals of two different genotypes. In contrast, for a Mendelian trait, two individuals of the same genotype will tend to be relatively similar, and all quite strikingly distinct from individuals of a different genotypic class. MCB 140, 12 -8 -06 11

Is trait X heritable? Height – yes. Language – no. Neuroticism – ? MCB 140, 12 -8 -06 12

Pellagra Disorder caused in large part by ? ? ? , and characterized by skin lesions and by gastrointestinal and neurological disturbances--the so-called classical three Ds of pellagra: dermatitis, diarrhea, and dementia. Pellagra Commission (1910 s): simple Mendelian inheritance of pellagra in the US South. “Controversy over the origins of pellagra continued until the mid-1930's…” Pellagra is caused by a dietary deficiency of niacin (nicotinamide and nicotinic acid) – precursor to NAD and NADP Pellagra ran in families because poverty ran in families, and poor families in the South subsisted on corn, which is low in niacin. MCB 140, 12 -8 -06 13

Is a trait heritable? Well, let’s mate individuals from the two extremes of the distribution, place their progeny into a common, controlled environment, and look. If the progeny follow a distribution that is skewed from the parental one towards the end of the curve their parents came from, the trait is heritable. If the progeny follow the same distribution as the original one, the trait is not heritable. MCB 140, 12 -8 -06 14

Mapping of quantitative trait loci (QTLs) “Search for genes responsible for variance in quantitative traits” MCB 140, 12 -8 -06 15

Ulrike Heberlein: the inebriometer Moore et al. (1998) Cell 93: 997. MCB 140, 12 -8 -06 16

QTL mapping, version 1 Testing alleles of candidate genes (forward genetics): pick a gene (serotonin receptor) that may have something to do with the trait under study (neuroticism) and look for allelic forms thereof that may occur more frequently in individuals who are neurotic. How do you measure someone’s level of neuroticism? http: //cac. psu. edu/~j 5 j/test/ipipneo 1. htm MCB 140, 12 -8 -06 17

“Mood disorders”: anxiety • Serotonin (5 -HT) – neurotransmitter (motor activity, food intake, sleep, reproductive activity, cognition, emotional states, including mood anxiety). • Serotonin transporter (5 -HT uptake) – two alleles (long and short) – former transcribed more efficiently • “NEO personality inventory” – people with short allele have greater anxiety-related personality characteristics • Polymorphism explains 7 -9% of genetic variance Lesch et al. (1996) Science 274: 1527. MCB 140, 12 -8 -06 18

“Serotonin transporter genetic variation and the response of the human amygdala” “Individuals with one or two copies of the short allele of the serotonin transporter (5 -HTT) promoter polymorphism, which has been associated with reduced 5 -HTT expression and function and increased fear and anxiety-related behaviors, exhibit greater amygdala neuronal activity, as assessed by BOLD functional magnetic resonance imaging, in response to fearful stimuli compared with individuals homozygous for the long allele. These results demonstrate genetically driven variation in the response of brain regions underlying human emotional behavior and suggest that differential excitability of the amygdala to emotional stimuli may contribute to the increased fear and anxiety typically Hariri et al (2002) Science 297: 400. associated with the short SLC 6 A 4 allele. ” MCB 140, 12 -8 -06 19

Willis-Owen Biol. Psychiatry 2005 MCB 140, 12 -8 -06 20

Willis-Owen Biol. Psychiatry 2005 MCB 140, 12 -8 -06 21

Living With Our Genes D. Hamer and P. Copeland (1998) “In the future, a person who complains of depression or anxiety could have a DNA test to check the serotonin genes. People with compulsive behaviour such as gambling, drinking, drugs, or promiscuous sex, would be checked for dopamine genes. Eating disorder or obesity? Look at the genes for leptin, the leptin receptor, or its targets. … Doctors won’t be the only ones to read this information. Insurance companies … would be very interested in genetic predispositions toward addiction or mental disorders. The military … might want to know about genes for rebellious temperament. Employers might be interested in genes for loyalty. Religious orders would be wise to discourage high novelty seekers, while the maker of sports cars would want to target them with ads. Dating services would have revealing new ways to match people. Imagine how excited certain school administrators would be to track students who are bright, troubled, or aggressive. ” MCB 140, 12 -8 -06 22

QTL mapping, version 2 Mapping by linkage: Take pedigrees with some frequency of individuals affected by a trait (e. g. , schizophrenia), and scan their entire genome for “linkage” with some polymorphic marker (e. g. , a RFLP). MCB 140, 12 -8 -06 23

Mapping a qualitative trait by linkage MCB 140, 12 -8 -06 24

11. 24 MCB 140, 12 -8 -06 25

Brockmann and Bevova (2002). Trends Genet. 18: 367. MCB 140, 12 -8 -06 26

MCB 140, 12 -8 -06 27

The ob mouse MCB 140, 12 -8 -06 28

Well-said “The fact that single genes can be modified and produce obese mice conclusively shows that genes contribute to obesity. However, the effect of a tested gene – especially if it contributes to complex traits – depends on the genetic background (i. e. , the effects of other genes – and thus might or might not produce the expected phenotype, or a milder phenotype. ” Brockmann and Bevova (2002). Trends Genet. 18: 367. MCB 140, 12 -8 -06 29

Problem For “simple” traits, the genotype-phenotype relationship is fairly unambigious, issues of penetrance and expressivity aside. If you are homozygous for Hb. S, you will get SCA, no question about it. It is largely irrelevant, what environment you grow up in – you’ll get anemia (although medical care will help you lead a better life, no question about that). For qualitative traits, the relationship between genotype and phenotype is impossible to express in a simple statement, so the general public, eager for simple solutions that fit its short attention span, has become enamored of the word “tendency. ” MCB 140, 12 -8 -06 30

Sigh “Genetic predisposition to…”: (you name it) MCB 140, 12 -8 -06 31

A pernicious line of reasoning I did X (or “I am Y”) I have a gene that caused me to do X (or to become Y) It’s not my fault, it’s the gene’s fault (or, “I am genetically superior”) MCB 140, 12 -8 -06 32

Misunderstanding a fundamental distinction between quantitative and qualitative traits Let’s say a driver with colorblindness (a mutation on the X chromosome) ran a streetlight and killed a person. Well, on some level, we will be less prone to blame that person (“he can’t help being colorblind”) for that act than someone with normal color vision who got drunk and ran that light. The leap here is to go from such “assignment of blame” in the case of qualitative traits, with their fairly unambiguous genotype phenotype relationship, to similar logic for quantitative traits (“I ran the streetlight because I was scared of being late to work, and I carry the allele of the serotonin receptor gene that makes me genetically scared”). MCB 140, 12 -8 -06 33

Norm of reaction A plot of carefully measured phenotype in large pool of genetically identical individuals grown under a range of environments. MCB 140, 12 -8 -06 34

Achillea millefollium (yarrow) MCB 140, 12 -8 -06 35

Take 7 yarrow plants, grow cuttings from each one at different elevations. Measure each “child” at each elevation. MCB 140, 12 -8 -06 36

MCB 140, 12 -8 -06 37

Risk of breast cancer and physical exercise in BRCA 1/2 mutation carriers: an example of how the norm of reaction illuminates the modification of a “genetic tendency” by environment “Physical exercise and lack of obesity in adolescence were associated with significantly delayed breast cancer onset. ” M. -C. King et al. Science 2003 MCB 140, 12 -8 -06 38

The reality of “Nature vs. nurture”: a group of individual of identical genotype, when placed in a normally distributed range of environments (X axis) will yield a population with a normally distributed range of phenotypes (Y axis), but the shape of the distribution will be a function BOTH of that genotype’s norm of reaction and of the distribution of environments. MCB 140, 12 -8 -06 39

Nature vs. Nurture: Wrong Question! 1. 2. Each organism is the subject of continuous development throughout its life. The environment’s effects on the organism will vary depending on when in development the effect is exerted (e. g. , PKU). The developing organism is not under the effect of genes and the environment acting separately. It is under the effect of mutually interacting genes and the environment. In the resulting organism, the effects of nature and nurture as “separable” as effects of flour and of the oven in the baked muffin. MCB 140, 12 -8 -06 40

And therefore … For any given individual, assigning percentages to “genes” and “environment” in determining the phenotype for a given quantitative trait has no meaning. For each given quantitative trait, each human being (a specific, given genotype), has a norm of reaction. The “norm of reaction” describes exactly how the trait will develop when very many individuals of exactly identical genotype will be placed in many different, very carefully controlled environments. Careful analysis of the norm of reaction may, perhaps, tell us a little bit about how the genes responsible for that quantitative trait interact with the environment in that trait’s development. No percentage values will be assigned to “nature or nurture” at any point in time—they are meaningless in this context. MCB 140, 12 -8 -06 41

Genetics of “intelligence” MCB 140, 12 -8 -06 42

“The Bell Curve” (1994) R. J. Herrnstein and C. Murray Claims (according to Hn. M, based on data they present): 1. African-American and Latinos in the US score on the average lower on IQ tests that whites or Asian. Americans, and lower IQ contributes to greater crime, poverty, illegitimacy, welfare dependency, unemployment, workplace injury. 2. IQ is substantially heritable and the higher fecundity of Blacks and Latinos leads to dysgenesis – a decline in the population’s potential for high IQ. 3. Forecast: establishment of a caste-like, reproductively isolated, cognitive elite, maintained through intermarriage, and ruling over the cognitively limited masses. MCB 140, 12 -8 -06 43

Public vs. private dialogue HM: “Here was a case of stumbling onto a subject that had all the allure of the forbidden. Some of the things we read to do this work, we literally had to hide when we were on planes and trains. ” In other words, the data really support all the claims, and people are afraid to admit it. Problem: >99% of all people in the US are not qualified to understand whether the data support the claims. New York Times: “But this reviewer is not a biologist and will leave the arguments to the experts” (10/27/94). MCB 140, 12 -8 -06 44

Calling a spade “[The] taboo being violated is … that of the great war against Nazi Germany. It’s not the taboo against unflinching scientific inquiry, but against pseudoscientific racism. Of all the world’s taboos, it is the one most deserving of retention. ” John Judis Hearts of Darkness (The Bell Curve Wars, New York, Basic Books 1995) MCB 140, 12 -8 -06 45

For the record 1. In my opinion, Drs. Herrnstein and Murray are pseudoscientific racists. That is, they just don’t like people of other races, and hide that dislike behind data, or what they claim to be data. 2. In addition, they deliberately misinterpret those data, and their conclusions, therefore, are wrong. 3. Problem: while no education is necessary to see the validity of item 1, item 2 requires a large amount of specialized knowledge. MCB 140, 12 -8 -06 46

What do the data show? What is IQ and how heritable is it? How severe is the threat of dysgenesis for IQ? What is the correlation between low IQ and societal ills? To answer these question, we must examine the scientific meanings of the words “IQ” and “heritable, ” the procedures for measuring IQ heritability in human populations, or the heritability of anything in any population, and how correlations between things are actually measured. MCB 140, 12 -8 -06 47

Point 1 – the reification of IQ “Among the experts, it is by now beyond much technical dispute that there is such a thing as a general factor of cognitive ability on which human beings differ and that this general factor is measured reasonably well by a variety of standardized tests, best of all by IQ tests designed for the purpose” H+M “Extraordinary obfuscation” (SJG, The Mismeasure of Man) MCB 140, 12 -8 -06 48

“The tendency has always been strong to believe that whatever received a name must be an entity or a being, having an independent existence of its own. And if no real entity answering to the name could be found, men did not for that reason suppose that none existed, but imagined that it was something peculiarly abstruse and mysterious. ” John Stuart Mill MCB 140, 12 -8 -06 49

Speed Rodgers Rop (Kenya) 2 hrs. 9 min. marathon Tim Montgomery (USA) 9. 78 sec. 100 m Nolan Ryan (USA) 100. 9 mph 8/20/74 MCB 140, 12 -8 -06 50

Point 2 – the heritability of the “trait” that IQ tests measure “The genetic component of IQ is unlikely to be smaller than 40% or higher than 80%. … We will adopt a middling estimate of 60% heritability … the balance of the evidence suggests that 60% may err on the low side” (p. 105) MCB 140, 12 -8 -06 51

Environmental vs. genetic variance in quantitative traits aka “How extremely technical it will very rapidly get” MCB 140, 12 -8 -06 52

20. 13 MCB 140, 12 -8 -06 53

20. 13 MCB 140, 12 -8 -06 54

Two sources of variance In a given population, variance will be due to individuals having different genotypes (genetic variance) and experiencing different environments: Please note, though, that the equation makes a claim about the population and NOT about the “relative contributions of genes and environment” to the phenotype of an individual. MCB 140, 12 -8 -06 55

Broad heritability is NOT, repeat, Nancy, Oliver, Tango, a general characteristic of this trait. In fact, for a trait, heritability can be 0 (if no genetic variation exists). Note that broad heritability being equal to zero does NOT, repeat, November, October, Typhoon, imply that the trait has no genetic basis. If H 2=0, then there is no variation in this population for genotypes that affect this trait, that’s all. Nothing can be said about whether this trait “is in the genes” or not. MCB 140, 12 -8 -06 56

… and therefore It is scientifically incorrect to use high withingroup broad-sense heritability of a trait to make claims about the “reasons” for differences between groups. MCB 140, 12 -8 -06 57

What H 2 does NOT mean Highly heritable traits (height, for example) are not necessarily “set in stone. ” For example, height is highly heritable. In the third world, people tend to be shorter due to malnutrition. If we take those children and move them to Marin county, they will grow much taller. MCB 140, 12 -8 -06 58

Problem H 2 is simply not a very useful number, however wide attempts at its (mis)use may be. It does not say much except something about the amount of genetic variation in a given population for a trait. Note that Fig. 20. 15 talks about how to measure H 2 in humans using twin studies. What one would really like is some quasiexperimental measure of the extent to which we can expect a given trait to vary when we vary the genotype of the organism. I. e. , can we breed “a bigger artichoke”? MCB 140, 12 -8 -06 59

Narrow-sense heritability, h 2 = a really useful number that is really hard to measure properly MCB 140, 12 -8 -06 60

A closer look at genetic variance In a population, the total variance is due to individuals differing in genotype, and to individuals experiencing different environments: This sg 2 term, in fact, hides 3 different things! MCB 140, 12 -8 -06 61

Let’s take a “complex” trait controlled by 10 loci (A through J), each with two alleles (A and a through J and j). For any given locus, variation between individuals for this trait can be attributed to 3 different things: 1. The particular allele the individual inherited (e. g. , A tends to make you taller, and a tends to make you shorter). 2. The overall genotype for this locus (if you’re AA, you’ll be very tall; if you’re Aa, intermediate; if aa, short). 3. The genotype for the other loci – there always are epistatic interactions (for example, an AA BB person will tend to be taller than an AA bb person). Different traits – i. e. , height vs. neuroticism – may be expected to experience these 3 different variances differently. For example (hypothetical), neuroticism may be particularly susceptible to epistasis (item 3) – that is, it doesn’t really matter what specific allele of any given gene you’ve inherited, what you need is a particular combination. On the other hand, other traits may be more “additive” – hypothetically, height may be due to algebraic additive effects of individual genes – if you’re AA for the first gene pair and homozygous recessive for the other 9 loci, you’re 5 inches shorter that an AA BB person, and that person is 5 inches shorter than an AA BB CC person, etc. From an agricultural perspective, traits amenable to change by selective breeding are those that are susceptible to the “additive” process (“the more right alleles of different genes you get, the bigger the artichoke”). Items 2 and 3 reflect individual genotypes, on different chromosomes, are ripped apart by meiosis, and are impossible to systematically recreate in agriculture. Thus, traits in which it’s important what allele pair occurs at a given locus, or what combination of alleles are found at a number of different loci, are hard to manage via selective breeding, unless they all occur on the same short stretch of 1 chromosome! MCB 140, 12 -8 -06 62

Components of variance 1. Additive: what allele of what gene you inherited. 2. Dominance: what is the other allele for that locus. 3. Interaction: what other alleles of other genes you inherited. MCB 140, 12 -8 -06 63

The useful part of genetic variance In quantitative genetics, how extensive is the additive genetic variance for a trait is a measure of how much we can expect the trait to respond to changes in genotypic composition. MCB 140, 12 -8 -06 64

A definition of narrow-sense heritability The fraction of total variance in the population that is due to additive genetic variance: By the way, did anyone notice how rapidly technical it got, and away from “This kid is real bright, just like his old man was. Must be genetic…” ? MCB 140, 12 -8 -06 65

Fact and problem In agriculture, narrow-sense heritability is a valuable number to know, because it illuminates the extent to which a quantitative trait may respond to selective breeding. For example, is it feasible to breed a cow that gives more milk? How on Earth do you measure “additive genetic variance”? MCB 140, 12 -8 -06 66

Estimates of narrow-sense heritability from regression of offspring on parents MCB 140, 12 -8 -06 67

Regression of y on x MCB 140, 12 -8 -06 68

For a given quantitative trait, regression of offspring on parents yields a numerical estimate of additive genetic variance, i. e. , the elusive and highly interesting narrowsense heritability, h 2. MCB 140, 12 -8 -06 69

Gasp! Regression of offspring on midparents yields a line the slope of which is the elusive narrowsense heritability!!! 20. 16 MCB 140, 12 -8 -06 70

Resemblance between relatives (a simplification) 1. 2. 3. Resemblance between relatives is one of the basic genetic phenomena displayed by metric characters. It is, of course, due to shared genes and shared environment. In a given population, however, if a parent-child pair diverges from the population mean (the taller person in the village has a taller child), this may be due to the fact that they are the only ones with access to lead-free water, but, in a general, averaged sense, it’s more likely that their divergence is due to sharing of alleles of genes that contribute to the development of the trait. This means that the covariance of parents and their offspring for a given trait – i. e. , the tendency of both parents and their children to diverge from the population mean to the same extent in the same direction – is reflective of their common genetic makeup. D. Falconer Quantitative Genetics MCB 140, 12 -8 -06 71

20. 14 MCB 140, 12 -8 -06 72

MCB 140, 12 -8 -06 73

The importance of the womb What Herrnstein and Murray “overlooked” in “calculating” the narrow-sense heritability of IQ MCB 140, 12 -8 -06 74

Nature, July ’ 97: “The heritability of IQ” IQ heritability, the portion of a population's IQ variability attributable to the effects of genes, has been investigated for nearly a century, yet it remains controversial. Covariance between relatives may be due not only to genes, but also to shared environments, and most previous models have assumed different degrees of similarity induced by environments specific to twins, to non-twin siblings (henceforth siblings), and to parents and offspring. We now evaluate an alternative model that replaces these three environments by two maternal womb environments, one for twins and another for siblings, along with a common home environment. Meta-analysis of 212 previous studies shows that our 'maternal-effects' model fits the data better than the 'family-environments' model. Maternal effects, often assumed to be negligible, account for 20% of covariance between twins and 5% between siblings, and the effects of genes are correspondingly reduced, with two measures of heritability being less than 50%. The shared maternal environment may explain the striking correlation between the IQs of twins, especially those of adult twins that were reared apart. Devlin et al. (1997) Nature 388: 468. MCB 140, 12 -8 -06 75

Devlin et al. (1997) Nature 388: 468. MCB 140, 12 -8 -06 76

Hmmmm…. “Adoption designs are a popular means of estimating IQ heritability. Associated analyses, however, usually assume negligible maternal effects. By contrast, our results show that 20% of twin and 5% of sibling covariance may be attributable to maternal effects. These results have two implications: a new model may be required regarding the influence of genes and environment on cognitive function; and interventions aimed at improving the prenatal environment could lead to a significant increase in the population's IQ. Moreover, some of Herrnstein and Murray's conclusions regarding human evolution such as the development of cognitive castes and IQ dysgenics, arise from their belief that IQ heritability is at least 60%, and is probably closer to the 80% values obtained from adoption studies. Our results suggest far smaller heritabilities: broad-sense heritability, which measures the total effect of genes on IQ, is perhaps 48%; narrow-sense heritability, the relevant quantity for evolutionary arguments because it measures the additive effects of genes, is about 34%. Herrnstein and Murray's evolutionary conclusions are tenuous in light of these heritabilities. ” Devlin et al. (1997) Nature 388: 468. MCB 140, 12 -8 -06 77

Dysgenesis, huh? Amount that phenotype changes from generation to generation: R Selection differential: difference between the mean phenotype of the selected parents and the mean phenotype of the population before selection – s R = h 2 · s MCB 140, 12 -8 -06 78

A perspective on the allure of the “dream of blue blood” Why does it matter so much to people whether IQ is “hereditary” or not? E. L. Thorndike (1905): “In the actual race of life, which is not to get ahead, but to get ahead of somebody, the chief determining factor is heredity” [emphasis added] R. Lewontin (1995): “In the three-quarters of a century that have since passed, the central effort of human behavioral and psychological genetics has been to put a firm foundation under Thorndike’s claim. ” MCB 140, 12 -8 -06 79

Further reading on this subject 1. 2. 3. 4. D. Falconer Quantitative Genetics M. G. Bulmer Principles of Statistics R. Lewontin The Triple Helix S. J. Gould Mismeasure of Man MCB 140, 12 -8 -06 80