CH 5 - Genetic Linkage and Mapping

Syntenic genes

• genes on the same chromosome

• Alleles on these genes can cross over to produce recombinant chromosomes

• They can be close that they can’t sort independently

  • They’re linked genes

  • Not all of them though

Genetic Linkage Map

Plots the positions of genes and their relative distances from each other on chromosomes

Unlinked genes assort independently

• Alleles from each gene move independently to gametes

• Each gamete has a frequency of 25% bc there’s 4

  • 2 are the same as the parents (parental)

  • 2 are non-parental (has recombined)

• Has a RF = 50% (Recombination Frequency)

Complete Genetic Linkage

• Genes are located so close they’re never separated by crossing over

• The offspring thus only display the parental traits

• This is rare in nature but can be seen in male drosophila where crossing over doesn’t occur

• These genes are thus in linkage disequilibrium

• Has a RF = 0% (Recombination Frequency)

Incompletely Linked Genes

• Come recombination occurs

• However there’s more of an abundance of non-recombined (parental) gametes

  • Linkage Disequilibrium

• It’s less likely for them to be separated, so it’s like parental genes

• Has a RF = < 50% (Recombination Frequency)

How do you detect linkage?

• Quantify how alleles are associated in gametes/offspring

• Compare this to expectations based on independent assortment of alleles at each gene

• So perform chi-squared test (expected is 9:3:3:1)

  • You’ll see p « 0.05 meaning there’s a significant difference

  • So there’s another system at play than 9:3:3:1

  • The ratios then suggest linkage and not independent assortment

Recombination Frequency ( r )

• r = # recombs / total # progeny

• It’s (+) correlated with physical distance between genes on a chromosome

• Longer distance means more recombination

• When r is small, the closer the genes are

Two-Point Test-Cross Analysis

• Two-Point: 2 genes

• Test-Cross: Hetero x homo rec

• Determines whether two genes are linked

• Estimate the distance between them on a chromosome.

Two-Point Test-Cross Analysis: Drosophila Example

• Examines recombination between 2 genes in the female during oogenesis

  • w (eye colour), m (wing size)

• In fruit flies, recombination doesn’t occur in spermatogenesis

• Parents: w+m+/wm (red eye large wing mom) and w+m+/Y (white eye small wing dad)

• So the progeny has geno/phenotypes as depicted by the diagram

  • Unrecombined

    • Females: w+m+/wm OR wm/wm

    • Males: w+m+/Y OR wm/Y

  • Recombined

    • Females: w+m/wm OR w+m/Y

    • Males: wm+/wm OR wm+/Y

• Dad’s one X chromosome contribution never recombines, as shown in one of the daughter’s X

Two-Point Test-Cross Analysis: Drosophila Chi-Squared Linkage Analysis

• Under independent assortment, you’d expect 1:1:1:1 ratio

• So for expected value, you use ¼ to find it

• df = 3 and P < 0.005

• So the null hypothesis that the genes assort independently is rejected

• So they must be linked

How do you quantify genetic linkage?

• Find the rates for each trait in comparison to another until you’ve compared them all

• Find the recombination rates for each trait in comparison to the other

  • Lower r = closer together

  • Find the closest ones and start there

• 1% recombination = 1 map unit = 1 cM (centiMorgan)

Notation for Linked Genes

• AB/ab

  • The two dominaint alleles are in the cis conformation

  • Coupling

• Ab/aB

  • The two dominant alleles are in the trans conformation

  • repulsion

• Ab/aB ; cd/CD

  • Alleles are on seperate chromosomes

Three-Point Test-Cross

• Parent 1 is trihybrid, Parent 2 is homo rec for all loci

• The rarest progeny phenotypes result from double crossovers

  • Only the middle is different

  • It’s rare bc event one AND event 2 has to happen

  • If RF between allele 1 and 2 is 0.18

  • And RF between allele 2 and 3 is 0.12

  • then the probability is (0.18)(0.12) = 0.02

  • This means out of 1000, only 20 display it

  • 10 for each reciprocal phenotype

Recombination Interference

• In trihybrid crosses, the # of observed double crossovers is less than expected

• This is caused by recombination interference (I)

• Higher values of I indicate more interference

Interference Calculation

I = 1 - Coefficient of coincidence

Coefficient of Coincidence Calculation

Coefficient = # of observed double crossovers / # of expected double crossovers

Why is the map distance based off observed recombination events less than the actual distance?

• Double recombination events are often not detected

  • Only middle changes

  • If you look at 2 traits on the edges of the replaced portion, it’ll look the same

  • So you don’t see it and don’t count it with observed recombination events

• Longer the distance, the greater the difference between observed and expected

  • more distance means more opportunity for multiple crossovers

  • Each of those hide a recombination event bc it looks like parental phenotype

  • So observed recombination frequency is less than expected bc you don’t count it

Heterogenous Recombination Landscape

• Recombination hotspot: Genomic regions where recombination happens more frequently

• Recombination coldspot: Genomic regions without frequent recombination events

• It can be influenced by env factors

  • age, temp, diet

• Recombination rate may differ between sexes

  • More during oogenesis than spermatogenesis

• Recombination locations differ between sexes

  • More frequent on chromosome tips in males than females

NS and Recombination

• Natural selection may affect recombination

• So higher recombination rates may be adaptive

• A favoured mutant may arise in a population

• It eventually sweeps throughout the population bc it’s good (through NS)

• This mutation may be inherited with a gene it’s linked to

  • Recombination can seperate them if the linked “hitchhiker” is bad

Sex and Recombination

• Genome-wide recombination rate is lower in males

• So males have a smaller genetic map BUT the physical maps are the same

  • Physical map: measured actual # of bp for precise DNA length (exact distance)

  • Genetic map: measures relative distance based on frequency of crossing over (relative distance)

Allelic Phase

• Refers to which alleles are physically attached to eachother on same chromosome

• Disease-causing genes can be identified by looking at these linked polymorphisms

• Ex. disease allele D might be frequently associated with A1/B1

  • Sometimes it may not bc recombination, but it usually is

Genetic Linkage of Disease-Causing Allele

• A deleterious mutation may arrive on an individual

• It will be passed down through genetic drift

  • It’s close to allele A so we can tell it’s more likely to be near A than C

• Recombination may change that fact so it’s not always associated with A

Mapping variants associated with a phenotype

• A marker locus can be closely linked to a disease-causing allele

• Specific alleles at that marker locus would be significantly associated with the disease allele throughout a population

• GWAS can be used to locate these associations

GWAS

• Genome-wide association study

• It discovers associations between certain variations in our genetic code and a certain phenotype of interest

How is a GWAS performed?

• You sequence the complete genomes of as many people as you can

  • Controls → health people

  • Cases → people all with the same disease

• You test whether single nucleotide polymorphisms (SNP) tend to be significantly found with the disease

  • Each SNP is plotted, can be millions

  • The P-value (significance) is found

  • The slope (effect size of disease likelihood) is found

• This is done computationally accounting for covariates (like age or sex)

Axes:

• x: the possible allele combinations at one SNP

• y: the associated phenotype

Factors Affecting GWAS Results

• Statistical power (sample size)

  • Bigger studies = more chance to accurately detect associations

• Variation in biology

  • Some diseases can have big effects with 1-2 genes contributing

  • Other diseases have many genes with small effects adding up

• Environmental Influence

  • Traits heavily influenced by the env, the weaker genetic signal

  • Low heritability means the env explains most of the variation

Manhattan Plot

• A graph used in GWAS to show which genetic variants (usually SNPs) are associated with a trait.

• x-axis: Chromosomal Position

• y-axis: -log(P-value)

  • strength of association

  • The P value must bc reallyyyy small for it to be significant bc there’s a high chance of false (+)

  • So these values are easier to work with on a log scale

• The green peaks indicates stronger statistical evidence that this SNP is associated with the trait

  • Can involve one (coronery artery disease)

  • Can be polygenic (Crohn’s disease)