Topic 8: Assignment tests Flashcards
What are assignment tests?
- tests used to assign individuals into a subpopulation, based on their genotypes. Will assign them where their genotype has a higher probability of occurring
- can tell us if individuals migrated from one population to another
- assumes that sample locations are genetic groups
Highly separated populations on an assignment scatterplot tell us….
- that there is a high Fst (large genetic difference between the two groups)
- that the two populations are very genetically differentiated, and we would have a good chance of assigning individuals correctly to their population of origin.
the number of cross-assigned individuals (migrants) is a measure of _____ ______ between populations in assignment tests
genetic divergence
What are some uses for assignment tests?
- assigning individuals to natal populations
- determining how many subpopulations there many be
- measuring genetic differentiation
- identifying immigrants/gene flow
What are alternative methods to assignment tests?
bayesian methods
What are bayesian methods?
- more dynamic tests than assignment tests
- dont assume that sampling locations are genetic groups, instead build genetic clusters based on the data/genetic profile
- would use this if we suspect that populations are genetically structured differently than the geography would suggest
- cluster individuals in a way to maximize the differences between groups (Fst) and minimize differences within groups (Fis)
- bayesian methods also allow for admixture between genetic groups
What are admixture plots?
- graph that visualizes each sampled individuals genetic assignment
- every individual appears as a column
- genetically distinct clusters are plotted in different colours
- genetic composition of each individual (column) shows how much of their genetic background can be assigned to each of the clusters
How will migrants appear (visually) in admixture plots? admixed/hybrid individuals?
- migrants will have different colours that the population in which they are sampled in.
- hybrids will have their bar broken into more than one colour
When determining K for admixture plots ( how many genetic groups), underestimating the structure can lead to …
- management actions at the wrong spatial scale
- inflating population size estimates and preventing protection under legislations
- incorrect grouping when conducting association studies
When determining K for admixture plots ( how many genetic groups), overestimating the structure can lead to …
- costly conservation actions for ‘rare’ taxa, such as translocations or habitat restoration to unnecessarily promote gene flow
How do you determine the best ‘k’ number?
- either the greatest likelihood value
- or biggest rate of change in likelihood ( ex. if there is no increase after 5, choose 5)
what is the difference between local ancestry and global ancestry?
- global ancestry is a single genome-wide estimate of admixture proportion
- local ancestry is a population assignment of individual loci or regions of the genome
What can we determine from local ancestry?
- migration processes
- identify regions of selection
- infer recombination rates
What is PCA?
- Principle component analysis
- takes complex data sets and simplifies it, transforms it to reduce complexity
- similar to assignment tests, but can have more than 2 populations included
- attempts to show as much genetic variability as possible in a few axes
- individuals closer to one another on the axis are more genetically similar than those further away
What are the pros and cons of bayesian methods?
pros:
- can be used to find k
- can identify admixed individuals
cons:
- assumes HWE within populations
- assumes linkage equilibrium between loci
- computationally intensive
