Fred Allendorf or Cant we answer all of our questions by just generating more data 1 For many decades empirical population genetics lagged far behind theory developed by Sewall Wright RA Fisher JBS Haldane etc ID: 687321
Download Presentation The PPT/PDF document "Population genomics: Do we really need t..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Population genomics:Do we really need theory?
Fred Allendorf
or
Can’t we answer all of our questions by just generating more data?
1Slide2
For many decades, empirical population genetics lagged far behind theory developed by Sewall Wright, R.A. Fisher, J.B.S. Haldane, etc.Slide3Slide4
Empirical data
Theory & models
Analysis & interpretation
1920-1970: The early yearsSlide5
Empirical data
Theory & models
Analysis & interpretation
1971-2000: The golden yearsSlide6
Empirical data
Theory & models
Analysis & interpretation
2000-present: The “
omic
” yearsSlide7
Chinook salmon in the Columbia RiverSince not one example of gene frequency has been published for the population in question, a hypothetical (although realistic) model will have to suffice.Ray Simon 1972Genotyped 1,956 individuals at 19,703 loci from 56 sample sites2015
7Slide8
Four million genotyped fish and counting: What we’ve learned about genetic stock identification in salmonids. Lisa Seeb8Slide9
First and foremost, we needed population geneticists. These are geneticists who study DNA sequence variation in a species or a population and, from this, infer what happened to these species or population in the past. They can tell when populations split, whether they exchanged genes, and whether selection acted upon them.9Slide10
New printing available for $
55 at Amazon.com
1970Slide11
Available online for free
We have the same situation in population genomics. People have vast amounts of data and do completely half-ass things with it because they don’t know any better. And, I wish there was some way of persuading people that we need to train students in the development and properties of the methods. And that means population genetics. Slide12
Felsenstein’s timeline of population genetics12Slide13
I am constantly appalled to see people go through the motions of performing HW and LD tests, reporting the results, and then completely ignoring them, even when significant deviations are found. Few people know how to properly interpret multiple testing results, and most seem to have lost track of why it is essential to do these tests before using the data for other analyses.2015Slide14
FIS for 14,000 SNPs in two elk population samples (Luikart et al.)14Slide15
Inbreeding: Confusion between FIS and pedigree F FIS is not an estimate of the mean pedigree F of individuals.A high mean F is expected in small random mating populations because all individuals will be closely related, but the expected FIS is 0. In fact, FIS is expected to be <0 in very small populations because of allele frequency differences between males and females.
201615Slide16
“Filtering loci out of HW proportions will remove all loci under selection.” Anonymous genomicistPapers use LDNE to estimate Ne in population samples known to have LD from sources other than drift (e.g., hybridization). 16Slide17
blue - traditional
red – “genomics” Slide18
Genomic measures
Pedigree
F
2016
18Slide19
Identical by descent (IBD) chromosome tracts
=
F
Runs of Homozygosity
19Slide20
Single pair of chromosomes from Swedish collared flycatcherH = Proportion of heterozygous SNPs in 10Kb windows
~20 Mb, IBD tractEllegren et al., 2012, NatureKawakami et al., 2014, Mol
Ecol
Proportion of the genome IBD can be measured
with
incredible accuracy & precision
using whole
genome
sequences
with a good assembly!
20Slide21
We thus pass from apoint-theory to a strand-theory. 1965 R.A. Fisher21Slide22
The proportion of the genome IBD varies greatly even between full sibs! Pedigree F = 0.25They have all the same “Coefficient of Inbreeding”, but differ in the extent of homozygosity. R.A. Fisher (1965, p. 97)
22Slide23
Expected proportion of the genome IBD in progeny from the last two wolves. 23Slide24
Length distribution of IBD segments from common ancestor 2, 5, and 10 generations back2gens5gens10gensRoH also provide information about the time since the inbreeding event (Fisher 1965).24Slide25
The Wrangel Island mammoth has 20% less heterozygosity than the Oimyakon mammoth.
25Slide26
Wrangel Island mammothOimyakon mammoth: RoH23.3%0.83%26Slide27
Comparison of individuals allows estimation of Ne over time.
IBD
27Slide28
Polyploid salmonid fishes (Allendorf et al. 2015)28Slide29
Centromere Centromere 1 co~50%0 co~50%
19832015(10 loci)
(3,496 loci)
29Slide30
Thank you!
30