Transcript Lecture #9 ppt

Quiz
• What were the two most significant
consequences of geographic isolation of some
mangrove stand in Panama?
• In the Hogberg et al paper on Fomitopsis what
were the two most significant findings?
• ------• Why is there a Somatic Compatibility system in
fungi and whay it is a good proxy for
genotyping?
• Why do we talk of balancing selection with
regards to mating alleles and how would you
use mating allele analysis to prove the
relatedness of fungal genotypes
Are my haplotypes sensitive
enough?
• To validate power of tool used, one needs
to be able to differentiate among closely
related individual
• Generate progeny
• Make sure each meiospore has different
haplotype
• Calculate P
RAPD combination
1
2
• 1010101010
• 1011101010
• 1010101010
• 1010111010
• 1010101010
• 1010001010
• 1010101010
• 1010000000
• 1011001010
• 1011110101
Conclusions
• Only one RAPD combo is sensitive
enough to differentiate 4 half-sibs (in
white)
• Mendelian inheritance?
• By analysis of all haplotypes it is apparent
that two markers are always
cosegregating, one of the two should be
removed
If we have codominant markers
how many do I need
• IDENTITY tests = probability calculation
based on allele frequency… Multiplication
of frequencies of alleles
• 10 alleles at locus 1 P1=0.1
• 5 alleles at locus 2 P2=0,2
• Total P= P1*P2=0.02
Have we sampled enough?
• Resampling approaches
• Raraefaction curves
– A total of 30 polymorphic alleles
– Our sample is either 10 or 20
– Calculate whether each new sample is
characterized by new alleles
Saturation (rarefaction) curves
No
Of
New
alleles
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Dealing with dominant
anonymous multilocus markers
•
•
•
•
Need to use large numbers (linkage)
Repeatability
Graph distribution of distances
Calculate distance using Jaccard’s
similarity index
Jaccard’s
• Only 1-1 and 1-0 count, 0-0 do not count
1010011
1001011
1001000
Jaccard’s
• Only 1-1 and 1-0 count, 0-0 do not count
A: 1010011 AB= 0.6
B: 1001011 BC=0.5
C: 1001000 AC=0.2
0.4 (1-AB)
0.5
0.8
Now that we have distances….
• Plot their distribution (clonal vs. sexual)
Now that we have distances….
• Plot their distribution (clonal vs. sexual)
• Analysis:
– Similarity (cluster analysis); a variety of
algorithms. Most common are NJ and UPGMA
Now that we have distances….
• Plot their distribution (clonal vs. sexual)
• Analysis:
– Similarity (cluster analysis); a variety of
algorithms. Most common are NJ and
UPGMA
– AMOVA; requires a priori grouping
Results: Jaccard similarity coefficients
Frequency
P. nemorosa
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.90
0.92
0.94
0.96
Coefficient
1.00
0.98
Frequency
P. pseudosyringae: U.S. and E.U.
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.90
0.92
0.94
0.96
Coefficient
0.98
1.00
Results: P. nemorosa
4175A
p72
p39
p91
1050
P. ilicis
P. pseudosyringae
p7
2502
p51
2055.2
2146.1
5104
4083.1
2512
2510
2501
2500
2204
2201
2162.1
2155.3
2140.2
2140.1
2134.1
2059.2
2052.2
HCT4
MWT5
p114
p113
p61
p59
p52
p44
p38
p37
p13
p16
2059.4
p115
2156.1
HCT7
p106
0.1
P. nemorosa
Results: P. pseudosyringae
P. ilicis
P. nemorosa
4175A
2055.2
p44
= E.U. isolate
0.1
FC2D
FC2E
GEROR4
FC1B
FCHHD
FCHHC
FC1A
p80
FAGGIO 2
FAGGIO 1
FCHHB
FCHHA
FC2F
FC2C
FC1F
FC1D
FC1C
p83
p40
BU9715
p50
p94
p92
p88
p90
p56B
p45
p41
p72
p84
p85
p86
p87
p93
p96
p39
p118
p97
p81
p76
p73
p70
p69
p62
p55
p54
HELA2
HELA 1
P. pseudosyringae
AMOVA groupings
• Individual
• Population
• Region
AMOVA: partitions molecular variance
amongst a priori defined groupings
Example
• SPECIES X: 50%blue, 50% yellow
AMOVA: example
Scenario 1
v
v
Scenario 2
POP 1
POP 2
Expectations for fungi
• Sexually reproducing fungi characterized by high
percentage of variance explained by individual
populations
• Amount of variance between populations and
regions will depend on ability of organism to
move, availability of host, and
• NOTE: if genotypes are not sensitive enough so
you are calling “the same” things that are
different you may get unreliable results like 100
variance within pops, none among pops
The “scale” of disease
• Dispersal gradients dependent on propagule size,
resilience, ability to dessicate, NOTE: not linear
• Important interaction with environment, habitat, and
niche availability. Examples: Heterobasidion in Western
Alps, Matsutake mushrooms that offer example of
habitat tracking
• Scale of dispersal (implicitely correlated to
metapopulation structure)---
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
RAPDS> not used often now
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
RAPD DATA W/O COSEGREGATING MARKERS
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Coco Solo
Mananti
Ponsok
David
Coco Solo
0
237
273
307
Mananti
Ponsok
David
0
60
89
0
113
0
Distances between study sites
White mangroves:
Corioloposis caperata
Forest fragmentation can lead to loss of gene flow among
previously contiguous populations. The negative
repercussions of such genetic isolation should most severely
affect highly specialized organisms such as some plantparasitic fungi.
AFLP study on single spores
Coriolopsis caperata on
Laguncularia racemosa
Site
# of isolates
# of loci
% fixed alleles
Coco Solo
11
113
2.6
David
14
104
3.7
Bocas
18
92
15.04
Coco Solo
Coco Solo
Bocas
David
0.000
0.000
0.000
Bocas
0.2083
0.000
0.000
David
0.1109
0.2533
0.000
Distances =PhiST between pairs of
populations. Above diagonal is the Probability
Random d istance > Observed distance (1000
iterations).
Spatial autocorrelation
0.6
0.5
0.4
Moran's I
0.3
0.2
0.1
0
-0.1
-0.2
1
10
100
1000
10000
100000
1000000
Mean Geographical Distance (m)
Moran’s I (coefficient of departure from spatial randomness) correlates
with distance up to Distribution of genotypes (6 microsatellite markers)
in different populations of P.ramorum in California
32
Genetic analysis requires
variation at loci, variation of
markers (polymorphisms)
• How the variation is structured will tell us
– Does the microbe reproduce sexually or clonally
– Is infection primary or secondary
– Is contagion caused by local infectious spreaders or by a longdisance moving spreaders
– How far can individuals move: how large are populations
– Is there inbreeding or are individuals freely outcrossing
CASE STUDY
A stand of adjacent trees is infected by a disease:
How can we determine the way trees are infected?
CASE STUDY
A stand of adjacent trees is infected by a disease:
How can we determine the way trees are infected?
BY ANALYSING THE GENOTYPE OF THE MICROBES: if the
genotype is the same then we have local secondary
tree-to-tree contagion. If all genotypes are different then primary
infection caused by airborne spores is the likely cause of
Contagion.
CASE STUDY
WE HAVE DETERMINED AIRBORNE SPORES (PRIMARY
INFECTION ) IS THE MOST COMMON FORM OF INFECTI
QUESTION: Are the infectious spores produced by a
spreader, or is there a general airborne population of spores
may come from far away ?
HOW CAN WE ANSWER THIS QUESTION?
If spores are produced by a
local spreader..
• Even if each tree is infected by different
genotypes (each representing the result of
meiosis like us here in this class)….these
genotypes will be related
• HOW CAN WE DETERMINE IF THEY
ARE RELATED?
HOW CAN WE DETERMINE IF
THEY ARE RELATED?
• By using random genetic markers we find
out the genetic similarity among these
genotypes infecting adjacent trees is high
• If all spores are generated by one
individual
– They should have the same mitochondrial
genome
– They should have one of two mating alleles
WE DETERMINE INFECTIOUS
SPORES ARE NOT RELATED
• QUESTION: HOW FAR ARE THEY COMING FROM?
….or……
• HOW LARGE IS A POPULATION?
Very important question: if we decide we want to wipe out
an infectious disease we need to wipe out at least the
areas corresponding to the population size, otherwise
we will achieve no result.
HOW TO DETERMINE
WHETHER DIFFERENT SITES
BELONG TO THE SAME POP
OR NOT?
• Sample the sites and run the genetic markers
• If sites are very different:
– All individuals from each site will be in their own exclusive clade, if two
sites are in the same clade maybe those two populations actually are
linked (within reach)
– In AMOVA analysis, amount of genetic variance among populations will
be significant (if organism is sexual portion of variance among
individuals will also be significant)
– F statistics: Fst will be over ) 0.10 (suggesting sttong structuring)
– There will be isolation by distance
Levels of Analyses

Individual
•

identifying parents & offspring– very important in
zoological circles – identify patterns of mating between
individuals (polyandry, etc.)
In fungi, it is important to identify the "individual" -determining clonal individuals from unique individuals that
resulted from a single mating event.
Levels of Analyses cont…
• Families – looking at relatedness within colonies
(ants, bees, etc.)
• Population – level of variation within a
population.
– Dispersal = indirectly estimate by calculating
migration
– Conservation & Management = looking for
founder effects (little allelic variation),
bottlenecks (reduction in population size leads
to little allelic variation)
• Species – variation among species = what are
the relationship between species.
• Family, Order, ETC. = higher level phylogenies
What is Population
Genetics?
 About microevolution (evolution of species)
 The study of the change of allele frequencies,
genotype frequencies, and phenotype
frequencies
Goals of population genetics
• Natural selection (adaptation)
• Chance (random events)
• Mutations
• Climatic changes (population expansions and contractions)
•…
To provide an explanatory framework to describe the evolution
of species, organisms, and their genome, due to:
Assumes that:
• the same evolutionary forces acting within species
(populations) should enable us to explain the differences we see
between species
• evolution leads to change in gene frequencies within
populations
Pathogen Population Genetics
• must constantly adapt to changing environmental
conditions to survive
– High genetic diversity = easily adapted
– Low genetic diversity = difficult to adapt to changing
environmental conditions
– important for determining evolutionary potential of a pathogen
• If we are to control a disease, must target a population
rather than individual
• Exhibit a diverse array of reproductive strategies that
impact population biology
Analytical Techniques
– Hardy-Weinberg Equilibrium
• p2 + 2pq + q2 = 1
• Departures from non-random mating
– F-Statistics
• measures of genetic differentiation in populations
– Genetic Distances – degree of similarity between
OTUs
•
•
•
•
Nei’s
Reynolds
Jaccards
Cavalli-Sforza
– Tree Algorithms – visualization of similarity
• UPGMA
• Neighbor Joining
Allele Frequencies
• Allele frequencies (gene frequencies) =
proportion of all alleles in an all individuals
in the group in question which are a
particular type
• Allele frequencies:
p + q = 1
• Expected genotype frequencies:
p2 + 2pq + q2
Evolutionary principles: Factors
causing changes in genotype
frequency
• Selection = variation in fitness; heritable
• Mutation = change in DNA of genes
• Migration = movement of genes across populations
– Vectors = Pollen, Spores
• Recombination = exchange of gene segments
• Non-random Mating = mating between neighbors rather
than by chance
• Random Genetic Drift = if populations are small
enough, by chance, sampling will result in a different
allele frequency from one generation to the next.
The smaller the sample, the
greater the chance of deviation
from an ideal population.
Genetic drift at small population
sizes often occurs as a result of
two situations: the bottleneck
effect or the founder effect.
Founder Effects; typical of
exotic diseases
• Establishment of a population by a few individuals can profoundly
affect genetic variation
– Consequences of Founder effects
•
•
•
•
Fewer alleles
Fixed alleles
Modified allele frequencies compared to source pop
GREATER THAN EXPECTED DIFFERENCES AMONG POPULATIONS
BECAUSE POPULATIONS NOT IN EQUILIBRIUM (IF A BLONDE FOUNDS
TOWN A AND A BRUNETTE FOUND TOWN B ANDF THERE IS NO
MOVEMENT BETWEEN TOWNS, WE WILL ISTANTANEOUSLY OBSERVE
POPULATION DIFFERENTIATION)
Bottleneck Effect
• The bottleneck effect occurs when the numbers of
individuals in a larger population are drastically reduced
• By chance, some alleles may be overrepresented
and others underrepresented among the survivors
• Some alleles may be eliminated altogether
• Genetic drift will continue to impact the gene pool
until the population is large enough
Founder vs Bottleneck
Northern Elephant Seal:
Example of Bottleneck
Hunted down to 20 individuals in
1890’s
Population has recovered to over
30,000
No genetic diversity at 20 loci
Hardy Weinberg Equilibrium
and F-Stats
• In general, requires co-dominant marker system
• Codominant = expression of heterozygote
phenotypes that differ from either homozygote
phenotype.
• AA, Aa, aa
Hardy-Weinberg Equilibrium
• Null Model = population is in HW
Equilibrium
– Useful
– Often predicts genotype frequencies well
Hardy-Weinberg Theorem
if only random mating occurs, then allele frequencies
remain unchanged over time.
After one generation of random-mating, genotype
frequencies are given by
AA
Aa
aa
p2
2pq
q2
p = freq (A)
q = freq (a)
Expected Genotype
Frequencies
• The possible range for an allele frequency or
genotype frequency therefore lies between ( 0 –
1)
• with 0 meaning complete absence of that allele
or genotype from the population (no individual in
the population carries that allele or genotype)
• 1 means complete fixation of the allele or
genotype (fixation means that every individual in
the population is homozygous for the allele -i.e., has the same genotype at that locus).
ASSUMPTIONS
1) diploid organism
2) sexual reproduction
3) Discrete generations (no overlap)
4) mating occurs at random
5) large population size (infinite)
6) No migration (closed population)
7) Mutations can be ignored
8) No selection on alleles
IMPORTANCE OF HW
THEOREM
If the only force acting on the population is random
mating, allele frequencies remain unchanged and
genotypic frequencies are constant.
Mendelian genetics implies that genetic variability
can persist indefinitely, unless other evolutionary
forces act to remove it
Departures from HW Equilibrium
• Check Gene Diversity = Heterozygosity
– If high gene diversity = different genetic sources due
to high levels of migration
• Inbreeding - mating system “leaky” or breaks
down allowing mating between siblings
• Asexual reproduction = check for clones
– Risk of over emphasizing particular individuals
• Restricted dispersal = local differentiation leads
to non-random mating
Pop 3
Pop 4
FST = 0.30
Pop 2
Pop 1
FST = 0.02
Pop1
Pop2
Pop3
Sample
size
AA
20
20
20
10
5
0
Aa
4
10
8
aa
6
5
12
Pop1
Pop2
Pop3
Freq
p
(20 + 1/2*8)/40 (10+1/2*20)/40 (0+1/2*16)/40
= 0.60
= .50
= 0.20
q
(12 + 1/2*8)/40 (10+1/2*20)/40 (24+1/2*16)/40
= 0.40
= .50
= 0.80
Local Inbreeding Coefficient
• Calculate HOBS
– Pop1: 4/20 = 0.20
– Pop2: 10/20 = 0.50
– Pop3: 8/20 = 0.40
• Calculate HEXP (2pq)
– Pop1: 2*0.60*0.40 = 0.48
– Pop2: 2*0.50*0.50 = 0.50
– Pop3: 2*0.20*0.80 = 0.32
• Calculate F = (HEXP – HOBS)/ HEXP
• Pop1 = (0.48 – 0.20)/(0.48) = 0.583
• Pop2 = (0.50 – 0.50)/(0.50) = 0.000
• Pop3 = (0.32 – 0.40)/(0.32) = -0.250
F Stats
Proportions of Variance
• FIS = (HS – HI)/(HS)
• FST = (HT – HS)/(HT)
• FIT = (HT – HI)/(HT)
Pop Hs
HI
p
q
1
0.48 0.20 0.60 0.40
2
0.50 0.50 0.50 0.50
3
0.32 0.40 0.20 0.80
HT
FIS
FST
FIT
Mea 0.43 0.37 0.43 0.57 0.49 0.12 0.24
n
0.14
Important point
• Fst values are significant or not
depending on the organism you are
studying or reading about:
– Fst =0.10 would be outrageous for humans,
for fungi means modest substructuring
Host islands within the
California Northern Channel
Islands create fine-scale
genetic structure in two
sympatric
species of the symbiotic
ectomycorrhizal fungus
Rhizopogon
Rhizopogon occidentalis
Rhizopogon vulgaris
Rhizopogon sampling & study
area
• Santa Rosa, Santa
Cruz
– R. occidentalis
– R. vulgaris
• Overlapping ranges
– Sympatric
– Independent
evolutionary histories
Local Scale Population
Structure
Rhizopogon occidentalis
FST = 0.26
N
5 km
T
B
FST = 0.24
Populations are similar
Grubisha LC, Bergemann SE, Bruns TD
Molecular Ecology in press.
FST
W
E
8-19 km
FST = 0.33
= 0.17
Populations are different