Molecular Evolution and Polygenic Inheritance

A. Recombination (see lecture 2)
B. Intragenic mutation (see lecture 3)

II. Origin of new genes or gene sequences

A. Eucaryotic organisms differ dramatically in amount of DNA and the number of functional genes. More than 90% of eucaryotic DNA is functionless. How can we explain this variation?

B. Unequal crossing over - can cause gene duplications and deletions

1. Results when chromosomes pair out of alignment

2. Occurs infrequently, but may explain origin of gene families, e.g.

a. Globins: alpha and beta, fetal and adult, myoglobin
b. Trypsins, chymotrypsin, thrombin
c. HLA - human leukocyte antigen - (MHC - multihistocompatibility loci in other vertebrates) - up to 70 alleles/locus, up to 12 loci on same chromosome. MHC is involved in recognition, binding and destruction of viral antigens. Also apparently important for mate recognition in humans and mice. Humans dissassortatively choose t-shirts based on HLA genotype. Mice dissassortatively choose mates based on MHC type. Best explanation is inbreeding avoidance.
d. Homeobox genes

C. Replication slippage
D. Gene conversion
E. Exon shuffling - all human genes evolved from 1000 unique exons
F. Retroviruses

1. can cause horizontal transmission of genes between organisms
2. used by drosophila geneticists to transform fly genomes

III. Polygenic inheritance

A. Many traits are continuously distributed, not categorical

1. Height
2. bristle number
3. litter size
4. survival

B. Galton

1. Franic Galton (Darwin's cousin) noted that children resemble their parents, although not exactly. Children at the extremes tend to "regress" toward the mean. Galton used the term "regression" to describe this linear relationship.

2. Galton started the field of biometrics. From 1880-1920 biometricians debated with Mendelians about the nature of genetic factors.

C. Fisher

1. Resolution came in 1918 when Fisher showed that a particulate genetic system can give rise to continuous traits without losing variation. Even though mean parents predict mean offspring, variation is replenished every generation through segregation and assortment.

2. One locus produces 3 genotypes, 3 loci produce 8 genotypes, etc.

3. If each locus adds a small increment to the trait, expect a continuous distribution

4. In addition, genetically identical animals often show phenotypic variation. This variation is due to environmental effects during development.

D. Heritability

1. We can measure the fraction of all variation that must be due to additive genetic effects by the heritability. This is useful because it can be used to predict the rate of evolutionary change in response to artificial or natural selection.

2. The heritability can be calculated from the resemblance among relatives. The simplest case is comparing average parent to average offspring. The slope of Galton's regression line then directly estimates the fraction of variation due to genetic effects, i.e. the heritability

3. If the slope is zero, the trait is not influenced by any genes, if it is 1 then all variation can be attributed to genetic variation. For most traits, it is intermediate.

4. Heritability is a population measure, like an allele frequency or a mean. It does not refer to an individual. It measures the extent to which variation in a population is caused by genetic differences.

5. Heritabilities depend on the environment in which they are measured because while the genetic variation may remain constant, the phenotypic variation can change. You cannot assume heritabilities remain constant in different environments

6. Heritabilities can be estimated from comparisons among other classes of relatives, but the estimates must be adjusted by the degree of relatedness among the relatives

a. One parent shares half their genes with their offspring. Therefore, if all variation is genetic, then a parent-offspring regression will have a slope of 0.5, i.e. slope x 2 = 1. Consequently, heritability = 2* slope.

b. Monozygotic twins share all their genes while dizygotic twins share only half their genes. Thus, twice the difference in similarity between these types of twins estimates the heritability.

c. However, in both of theses cases, parents and offspring or twins may be similar because they have grown up in similar environments or even learned similar habits. Thus, these estimates do not cleanly separate environmental effects from genetic effects. A better method is to compare adopted children with their biological or adoptive parents. The degree to which adopted children resemble their adoptive parents indicates environmental effects while the degree to which they resemble their biological parents represents genetic and prenatal environmental effects.

E. Examples

1. IQ - most studies indicate heritability around 0.5. This means that half of the variation in IQ is due to genetic variation while the remainder is due to environmental variation. In no sense does this mean IQ is genetically determined. One way to interpret this number is to imagine that most IQ estimates range between 75 and 125. If we use the difference as our measure of variation, then half this difference reflects the environmental variation, i.e. 25. This means that a child from parents with IQ's of 100, would be expected to have an IQ of 100 but environmental factors could move it between 112 and 88. Furthermore, segregation, could cause it to take virtually any value. Thus, predicting an individual's IQ is questionable.

2. IQ studies can also be questioned on ethical grounds because they have been used in sterilization programs in Virginia as recently as 20 years ago. Between 1927 and 1972 7500 people were sterilized because they had low IQ scores - primarily on white unwed mothers, prostitutes, petty criminals, and juvenile delinquints.

3. The "Bell Curve" was the most recent misinterpretation of these data. Data show that IQ is heritable in caucasions and african americans, but the average african american is 20 points below the average caucasion. Is this a genetic effect? Heritability studies do not provide the answer. Such a mean difference could simply be due to the environment.

4. Adopted study in France showed that IQ could be changed by an equal amount depending on both environment and parentage

IV. Gene localization of complex traits -> Quantitative Trait Loci (QTL)

A. By using many genetic markers throughout the genome, it is now possible to locate genes which contribute partially to the expression of a character using recombination frequencies

1. Examine pedigrees to locate nonrandom associations between markers and incidence of disease or complex trait

2. Test putative association with other pedigrees to insure generality of result.

3. Alternatively, examine afflicted and nonafflicted individuals for the presence of a suspected disease causing allele.

B. Examples

1. Schizophrenia
2. Alcoholism
3. Sexual orientation - some evidence for an x-linked effect which causes male homosexuality
4. breast cancer (BRC1)

C. Be cautious of media claims.

1. QTLs often only explain a fraction of the variation in any trait. Consequently, a single allele does not determine the expression of the character.
2. Also, studies need to be replicated in multiple environments. If suspected QTL does not appear in every environment, something more complicated, like genotype x environment interaction, may be occurring.