Quantitative Genetics

Question 1

what are complex traits?

Answer 1

traits that are controlled by multiple genes

- you have to measure these traits rather than categorise

Question 2

when does variation follow a normal distribution?

Answer 2

• multiple loci are evolved
• each locus has about equal effect size
• each locus acts independently

Question 3

what is a quantitative trait locus?

Answer 3

a locus (section of DNA) which correlates with variation of a quantitative trait in the phenotype of a population or organisms

Question 4

what is QTL mapping?

Answer 4

uses the populations derived from bi-parental crosses to identify QTL

Question 5

what is GWAS?

Answer 5

use populations of diverse (not closely related) individuals to identify QTL

Question 6

what do both QTL mapping and GWAS use?

Answer 6

use nearby markers (eg SNP markers) to map QTL

Question 7

what happens when the marker is closer?

Answer 7

the closer the marker the more often its co-inherited with the QTL, looking to find markers closest to the gene of interest

Question 8

what is the process of QTL mapping?

Answer 8

• begins with bi-parental cross between parents of different phenotype
• F1 will be identical

Question 9

what happens in an F1 x F1 in QTL mapping?

Answer 9

• recombination will occur during meiosis
• reshuffles genes in the games
• phenotypes segregate at F2

Question 10

what happens when the parents and F2 are genotyped in QTL mapping?

Answer 10

• can identify which sections of the chromosome have been inherited from each parent
• can see what phenotypes they have
• can find associations between phenotype and section of DNA inherited
• sections that correlate with the phenotype = QTL

Question 11

what are the advantages of F2 and BC (F1 x Parent) populations?

Answer 11

quick and simple

good for preliminary mapping

Question 12

what are the disadvantages of F2 and BC (F1 x Parent) populations?

Answer 12

a single individual represents each genotype
replications overtime/space can’t be carried out
low resolution

Question 13

what are recombinant inbred lines?

Answer 13

selfing F2 populations through more generations

Question 14

what are the advantages of recombinant inbred lines?

Answer 14

• very homozygous - can fix the alleles
• dont have a single individual, there are populations of individuals of each genotype
• immortal, can get large populates and breed to get more of the same genotype

Question 15

what are the disadvantages of recombinant inbred lines?

Answer 15

time consuming to produce, there are more generations

Question 16

what is advanced backcross population?

Answer 16

several rounds of backcrossing and selfing

Question 17

what are the advantages of advanced backcross population?

Answer 17

• useful for simultaneous QTL analysis
• breeding QTL into elite lines
• can be use to produce near isogenic lines (NILs) for additional analysis

Question 18

what are near isogenic lines?

Answer 18

nearly identical apart from small sections of DNA from one of the parents

Question 19

what are the disadvantages of advanced backcross population?

Answer 19

time consuming

Question 20

how can we identify QTL?

Answer 20

• single marker analysis
• interval mapping (IM)
• composite interval mapping

Question 21

what is single marker analysis?

Answer 21

• looks at every marker-trait combination (t test or ANOVA)

- simple and quick but low precision

Question 22

what is interval mapping (IM)?

Answer 22

• uses information from pairs of consecutive markers to estimate the likelihood of a QTL being between them
• expect if you have a marker either side of a mutation then both of those markers are more likely to be inherited then those further apart
• more powerful and gives a slight more precise QTL location

Question 23

what is composite interval mapping?

Answer 23

• tests for QTL using IM but simultaneously controls for variance
• tries to estimate the effect that each individual locus has on the trait
• allows you to separate two regions that are close together
• most accurate but statistically complicated and requires more computational power

Question 24

what are the limitations of QTL studies?

Answer 24

• resolution is often to low to identify candidate genes without further fine mapping
• can be time consuming to build populations
• QTL detection limited to the genetic variation between the two parents
• not always possible to build a population

Question 25

how can you carry out a QTL study without building a population?

Answer 25

- existing families can be studied - there are constraints on population size and experimental design - limits statistical power

Question 26

what is GWAS?

Answer 26

``` identifies markers (usually SNPs) that are significantly associated with a trait of interest across a diversity panel of individuals - don't want them to be closely related ```

Question 27

how does GWAS work?

Answer 27

- phenotype the indiduals and look for genome wide SNP phenotypes - identify SNP alleles associated phenotype - each of the marker is normally presented according to its position in the genom - find not only signle marker but multiple markers - forms peaks

Question 28

why do you get multiple markers?

Answer 28

because there are markers of linkage disequilibrium

Question 29

what does QTL analysis rely on?

Answer 29

linkage - the physical state of being linked due to the chromosomal organisation of the genome

Question 30

what is linkage disequilibrium?

Answer 30

refers to the presence of statistical association between allelic variants, is this marker statistically associated with his marker - the degree of non-random association of alleles at two or more loci

Question 31

whats linkage disequilibriums role in GWAS?

Answer 31

determines the resolution of mapping

Question 32

what is long distance LD?

Answer 32

- mapping the centimorgan (cM) distance - markers across a large amount of chromosome - low resolution

Question 33

what is short distance LD?

Answer 33

- mapping at the base pair (gene) distance | - get to a candidate gene more easily

Question 34

when are alleles in perfect LD?

Answer 34

- if alleles are always seen together in a population

Question 35

what is linkage equilibrium?

Answer 35

- if all combinations of alleles are seen at a random in a population - recombination breaks down allele combinations over time - more distant loci will be broken down more quickly

Question 36

when does LD decay?

Answer 36

decays with distance between markers | markers can be co-inherited even if they're far away

Question 37

what are the major factors of LD?

Answer 37

- recombination rates - population size - inbreeding - mutation rate - selection - population structure

Question 38

what are recombination rates in LD?

Answer 38

- changes arrangement of haploytypes | - creates new haplotypes

Question 39

what is population size in LD?

Answer 39

- LD may increase in small populations as haplotypes are lost through genetic drift - rapid population growth reduces drift and LD

Question 40

what is inbreeding in LD?

Answer 40

- decay of LD is reduced in selfing populations

Question 41

what are mutation rates in LD?

Answer 41

- rapidly mutatin loci may show low LD even in the absence of recombination

Question 42

what is selection in LD?

Answer 42

- fixation of beneficial alleles through positive selection, get regions of inflated LD

Question 43

what is population structure in LD?

Answer 43

- population stratification may shape patterns of LD across population

Question 44

how can population structure effect data?

Answer 44

- can cause false positives or type 1 errors in association analysis - some individuals will be more closely realted than others - are markers really associated with the causative gene or are they just common in some populations

Question 45

how can we correct from population structure?

Answer 45

principle component analysis (PCA) - estimates how the individuals are related from their SNPs

Question 46

how would design a GWAS experiment to optimise analysis?

Answer 46

1. maximise resolution (diverse panel of accession and plenty of accessions) 2. saturate with markers (dependent on extnet of LD and size of genome, SNP clips/next generational sequencing) 3. control false positives (use linear models that incorporate population structure)