Conservation Genetics

1/27/2025

Darwins postulates

• Individuals in a population vary in traits

• Some trait variation is inherited

• There is competition for resources

• Successful survival and reproduction is dependent on fitness and is not random

For evolution to occur, natural selection must act upon genes/alleles

• Acts upon variation that already exists

• New variants can arise by mutation, and selection can act upon these novel variants

Natural selection: Differential survival and reproduction

• Natural selection acts on individuals most of the time

• Evolution occurs in populations

12/29/2025

Mismatch: changes in our environment, traits we evolved to match the previous environment are now no longer pertinent or helpful

Connection between bacteria and allergies/certain inflammatory diseases

• Our microbiome is no longer the same as it used to be

• We get antibiotics in our food

Ancestors didn’t experience the same stresses as we do

• We are not evolutionarily meant to handle this kind of stress

• Mismatched with our environment, and now we experience chronic stress

Cline: populations that are near each other and genetically similar (more accurate term for cline)

• Evidence: all human genomes are about 95% similar

• Our differences are more of a gradient, our differences are

• The differences that we do find among races are no greater than the differences we find between people of the same race

What have we learned from studies on ancient DNA

• The average person has about 2% Neanderthal DNA

Cheddar man

• The cheddar man is a 7000-year-old skeleton in the Britons

• Actually had dark skin

2/3/2025

Importance of taking an evolutionary and genetic approach in conservation

• Diversity in genotypes

• Phylogenetic approach

• How do we conserve species

• Root of the problem

Why conserve genetic diversity

• susceptibility to genetic disease

• Inability to adapt

• Genetic diversity is the ultimate source of biodiversity

• Raw material for evolution

Categories of topics in conservative genetics

• management and reintroduction of captive breeding programs and restorations of biological communities

• Detecting and predicting the effects of habitat loss

• Description and identification of individuals, genetic population structure, etc

• Detection and prediction of effects of hybridization and introgression

• Understanding the relationships between the fitness of individuals or local adaptation of populations and environmental factors

Problems with small populations

• <1000 two-humped camels in 4 disjunct populations

• Song sparrows of mandate island

  • Small islands population — very unstable with severe crashes

  • their survival was correlated with inbreeding coefficient

• Glanville fritillary butterfly

  • Metapopulations: groups of smaller populations that interact

Management of inbreeding

• florida panther

  • populations inbreed because habitat disappeared

• Captive breeding programs require genetic management

Effective population size

• Not full number of individuals in the population

• Using genetic data from representative sample to estimate effective population size

Population structure and management units

• are all individuals in an area interacting?

Adaptation

• Geographically distinct panda populations differ in the amount of bitter taste receptor genes

• Ones with more bitter receptors eat more leaves

Genetics vs genomics

• Genetics = things we can understand from a single gene/set of genes with known locations on the chromosomes

• Genomics = sequencing the whole genome, all the genes in all the chromosomes in a cell

What do we conserve?

• How do we assign value to a species

• Keystone species — species with more important niches

• Genetic diversity within the population — save species with the potential to evolve more

• Where the species are geographically

• Populations vs species

  • Evolution happens at a population level

2/10/2025

Genetic variation

• Can be adaptive → coding DNA

  • Influenced by selection

• Can be neutral → noncoding DNA

  • Influenced by drift and gene flow

• WE study genetic variation using

  • Microsatellites

  • SNPs

  • Mitochondrial

How do we generate genetic data?

• Polymerase chain reaction (PCR)

  • Revolutionized molecular genetics

  • Amplifies specific fragments of DNA

  • Uses primers specific to genetic markers

  • You can use gel electrophoresis to analyze it — this is just the amplification

  • Examples:

    • Looking at microsatellites — use a primer for specific regions

    • Looking for specific genes in the mitochondrial genome

How to quantify genetic variation

• Heterozygosity within an individual: Proportion of loci that are heterozygous

• Heterozygosity within populations: proportion of heterozygous (polymorphic) loci

2/12/2025

Genomics

• Use genetic data from the whole genome

• More data than with traditional genetic markers

• next-generation DNA sequencing

• Requires bioinformatics to process data

• Better able to infer demographic history

• study both selectively neutral and adaptive genetic variation

• However — analytical, computational, and big data challenges

DNA sequencing

• Method 1: Sanger Sequencing

  1. Denature dsDNA

  2. Make multiple copies of a segment

  3. Attach a primer

  4. Add to four polymerase solutions

  5. Grow complementary chains until termination dye

  6. Denature the grown chains

  7. Electrophorese the four solutions

  • Result: a single fragment amplified - usually >1000 bp

• Method 2: Next generation

  • Whole genome from the cell

  • Chop DNA up into pieces

  • Build DNA libraries

  • Sequence strands, end up with short reads that you have to use to generate contigs

What to do with this information

1. create a reference genome (entire sequence of DNA form an organism)

2. Genome annotation — you need a transcriptome (full RNA sequence)

3. Assembled genome = most or all DNA assembled into contiguous sequence, ideally scaffolded onto chromosomes

4. Annotated genome = associates DNA sequence regions with genes

How do you get the SNP

• Whole genome resequencing

  • Tons of info

  • Data storage issues

• Reduced representation sequencing

  • A subset of the genome

  • Can be random or targeted

  • RAD seq (restriction site associated sequencing)

    • Barcodes to identify individual sequences

• Targeted sequence capture

  • Sequence pre-selected regions of the genome

  • NOT random

• Transcriptomics

  • RNA sequencing and RNA microarrays

    • Reveals info about gene expression

Meta genomics

• sequencing all the DNA in a sample containing multiple species

• Example: microbiome- soil or gut

• Environmental DNA: water, soil, fecal sample

Epigenetics

• Heritable changes that regulate gene expression but do not change DNA sequence

  • DNA methylation

2/19/2025

Heterozygosity

• Observed heterozygosity: proportion of heterozygous loci or individuals at a locus

• Expected heterozygosity: Proportion expected from HWE

  • With 2 alleles at each locus: He=2pq

  • With >2 alleles at a locus: He=1- (sum #alleles i=1) pi2

Small populations

• Small populations violate hardy weinberg

• Genetic drift acts on small populations

  • Genetic drift: random (stochastic) fluctuations in allele frequencies within a population

  • Primary evolutionary force causing frequency change across the genome over time

  • Happens in large populations but tends to be subtle

  • Alleles fluctuate each generation due to random sampling

Changes in allele frequency

• Direction of change is less predictable in small populations — but you can predict the amount it will change

• deltaq=(q(1-q))/2Ne = amount allele frequency will change per generation

  • deltaq=variance

  • Sqrt(variance)=std deviation = actual amount the allele frequency will change in each generation

• Example: if q=0.5, what is the change in q for a population of 10 prairie chickens vs a population of 200 prairie chickens

  • (.5)*(.5)/20=0.0125

    sqrt(0.0125)=0.112

  • (.5)*(.5)/400=.000625

    sqrt(.000625)=0.025

• Predicting allele frequency rage in next generation

  • p’=p+-2sqrt[(pq)/(2N)]

  • p’=allele frequency in next generation

  • p=allele frequency in current generation

• Example: if p=0.5 and N=10, p in the next generation will range

  • 0.5+-2(0.112)= 0.28 — 0.72

  • 0.5 +- 2(0.025)= 0.45 — 0.55

effect of allele frequency on allele loss

• more even distribution = less likely to reach fixation

• Rate of heterozygosity loss per generation

  • deltah=-1/(2N)

  • Ht=H0[1-1/(2N)]^t

    • H0 = initial heterozygosity

    • Ht = heterozygosity at time t

    • t= # of generations

2/24/2025

Problems with small populations

• Stochasticity

  • Demographic — changing sex ratio

  • Environmental — storms, unpredictable changes in conditions

  • Genetic — drift, inbreeding, etc.

• Genetic drift: results in fluctuations in allele frequencies AND heterozygosity

  • Both depend on population size

• Heterozygosity is maximized with moderate allele frequencies

• Loss of heterozygosity is an increase in homozygosity

  • This can happen solely from genetic drift, without inbreeding

  • Inbreeding can exacerbate the increase in homozygosity

  • Measured as a function of population size

  • delta h is a rate — percentage of the starting heterozygosity lost in each generation

Consequences of genetic diversity loss

• Population bottleneck

  • Specific type of population change — sudden and rapid decline

  • Stay at low population number for a while — population can either then recover or go extinct

  • Consequences depend on

    • How long population stays at low numbers (duration)

    • How small does the population get

    • Populations <50 considered severe bottleneck

• Genetic effects of bottleneck

  • Single-generation bottleneck must be severe to have a large effect (Ne<50)

  • Theoretical predictions

    • If N=2: decrease in He by 25% in 1 generation

    • If N=25: 2% decrease

    • If N-100, 0.5% decrease

• Examples

  • California COndors

    • Recessive lethal allele causes dwarfism

    • Consequence of bottleneck/founder effects in captivity

    • Would have been really rare in a large wild population, but in a small captive breeding population, it is a problem

• Loss of heterozygosity as a function of population size

  • Can calculate future heterozygosity from the initial heterozygosity and N (really Ne)

  • H_t = H₀ [1-(1/2N)]^t

    • H₀ = initial heterozygosity

    • H_t = heterozygosity at time t

    • t = number of generations

  • Bottlenecl in Mauritius kestrel

    • 1974: single pair

    • 1997: N=400-500

    • 57% loss of gen diversity

  • Example:

    • H₀ = 0.23

    • N=2

    • t=1

    • H_t = 0.23[1-(1/2×2)]¹ = 0.173

    • After 3 generations it would be

      • H_t = 0.23[1-(1/2×2)]³

      • H_t = 0.097

Founder effects

• Lose genetic diversity as just a few individuals found new populations

• depends on

  • How many founders there are

  • Their genetic diversity

  • Are they a representative sample of the parent pop

• Example: Scandinavian wolves and ISle Royale wolves

  • Last pair of wolves on Isle Royale — half-siblings and father-daughter

  • In 2018, people reintroduced wolves from Canada

Patterns

• Patterns of allele frequencies can indicate past bottlenecks or founder effects

• You have fewer rare alleles in populations that have been bottlenecked

• Heterozygosity is lossed slower than allelic diversity

Inbreeding

• ⁽Inbreeding: mating between close relatives

  • Can be measured at the individual level (the extent to which its genome is homozygous due to shared genetic ancestry) or at the population level (level of similarity among individuals due to relatedness and non-random mating)

• Inbreeding depression: Affect of inbreeding on fitness

  • Inbred individuals have higher mortality rates

• Identical by descent: 2 copies of the same recessive lethal allele

• Autozygous: IBD at a locus, have 2 copies of the same allele that can be traced to a common ancestor

• Allozygous: both alleles derive from 2 different ancestors

2/26/2025

Quantifying inbreeding

• Inbreeding coefficient (F): many types and definitions

  • Proportion of the genome that is IBD

  • Always ranges 0-1

• Pedigree inbreeding coefficient (F_p): Measures increase in homozygosity for inbred individuals relative to others in the population

  • Figure it out by looking at pedigrees

  • Assume pedigree with A,B,C,D,E and X. Assume ancestor has no inbreeding (Fa=0)

  • X’s parents (D and E) are half siblings (both have A as mom)

  • To determine Fp of X, trace path from X back through the common ancestor back to X

  • There are 3 individuals in that path — N=3

  • To calculate pedigree inbreeding coefficient

    • Fp=(1/2)ⁿ(1+Fp(ca))

    • Only is Fp(CA) if you know the F other common ancestor

    • So if not, Fp=(1/2)ⁿ

    • F=(1/2)³=0.125

• Common co-ancestry

  • Coancestry coefficient = probability that 2 alleles, one from each of 2 individuals, are identical by descent

  • Not the same as the coefficient of relatedness

• Individual heterozygosity: proportion of loci that are heterozygous in each individual

  • IR: internal relatedness

  • Multilocus heterozygosity

• Runs of Homosygosty (F_ROH)

  • From the genome

  • Demonstrated on manhattan plot

  • Gaps in the plot = run of homozygosity

  • Can be used on individual or population level

Inbreeding depression

• Reduction in fitness of offspring produced by mating of relatives

• 2 direct causes

  • Unmasking of deleterious recessive alleles

  • Increased homozygosity more generally can reduce fitness, for many loci have heterozygote advantage

• Examples

  • Lower fecundity

  • Lower fertility

  • Higher mortality in young

  • Decreased population growth rate

• Effects of inbreeding are greater in stressful environments

  • Greater effects in wild vs captivity

  • Experimental evidence is there that increeding increases extinction risk

    • 80-95% of populations die out in short term in laboratory studies on mice

• Lethal equivalents = # of deleterious alleles with fatal consequence

  • 1 lethal equivalent = 1 allele with 100% fatality when homozygous

    • OR 2 alleles with probability of 0.5 of causing death when homozygous

    • OR 10 alleles with probability of 0.1 of causing death when homozygous

  • # of LE relates to the genetic load

  • Can be estimated from relationship of fitness and F

Is inbreeding ok sometimes

• Does inbreeding occur in the wild in healthy populations?

• Most animals avoid mating with relatives

• Naked mole rats have a queen and she mates with a ton of men — high relatedness

  • No observes inbreeding depression

    • Until one type of enteric coronavirus wiped out populations if inbred naked mole rats

  • Purging: maybe they have purged the deleterious alleles from their genomes over time

    • influenced by rate of inbreeding, population size, time

Genetic Rescue

• Population augmentation via translocation

• Used in Adders in Sweden that were isolated for a really long time

• And the Florida panthers

• Survivorship is a good measure of if genetic rescue worked

• Heterosis: offspring of outbred parents, offspring are more fit

• Genetic rescue: Decrease in extinction probability of a population due to gene flow

  • Measured by increase in population growth rate

  • If you din’t have growth rate data, look at increased fitness

  • Increase in heterozygosity is good, but doesn’t tell you much about demographic effect

    

  • Stochastic resilience: population has better resistance to disease, environmental factors, etc.

• 2 aspects to why heterozygosity is important

  • Harmful recessive alleles — remasking these alleles reduces the genetic load

  • Heterozygote advantage

• Adding genetic variation means there is more for selection to act upon → increases adaptive potential

  • BUT it is important to balance new allelese/adaptive potential with local adaptation

• Genetic rescue may also be an important process in the wild, especially for metapopulation

Concerns with genetic rescue

• Genetic swamping with new alleles

  • Solution: only bring in a few immigrants at a time

• Outbreeding depression: the introduced individuals reduce the fitness of the population that is being saved

  • Happens when the two populations are too different

  • Disrupts local adaptation

  • Decreased fitness of hybrids

    • Within your genome, certain genes are coadapted. Disrupting this disrupts adaptation

  • Solution: seek a genetically similar source population

• Current consensus: genetic rescue of small, inbred, recently isolated populations by large source populations should be the default

• Heterosis is known to be maximal in the first generation — how effective is genetic rescue over time?

• How frequently do populations need to be rescued?

  • Most studies have been short term

  • Florida panthers long term results

    • Big change in heterozygosity that increased over time

    • Survival was corellated with heterozygosity and genetic ancestry

    • Positive population growth

    • Lowered extinction probability

    • Concluded that management was still needed over long term, but nothing drastic

• How can genomics aid implementing genetic rescue

  1. Identifying source populations

  2. Identifying best individuals to translocate

  3. Monitering the outcomes of genetic rescue

Effective population size (N_e)

• Size of an idealized population that would act the same (have the same genetic consequences) as the population under consideration

• “gene pool”

• The actual size of the population overestimates the gene pool

• Genetic drift depends on the effective population size, not the census size

• The rate of genetic diversity loss (heterozygosity loss) depends on Ne

  • delta h=-1/2Ne

• Why might effective population not be the same as the census population

  • Unequal ratio of males to females

  • Nonrandom mating

  • Not everyone mates

  • Unequal offspring production

  • Population sizes fluctuate

What is an ideal population

• All individuals contribute equally to the gene pool

• Population size stays the same

• nonoverlapping generations

• Equal sex ratio

Ne VS N

• Ne<N (~.1-.5 in most wild populations)

• Some examples of ratios as low as 0.001 → marine fish, marine invertebrates

Unequal sex ratio

• N_e=(4N_f * N_m)/(N_f+N_m)

  • If sex ratio is equal (N_f = N_m). N_e=1

  • N_f = Number of females

  • N_m = Number of males

• Example: skewed sex ratios in hunted ungulate populations

  • Some suggest that 1 male per 25-100 females is sufficient to maintain population growth in hunted ungulate populations

  • Effective size is only about 4

• In the most extreme case, Ne ~ 4 * the rarer sex

• Generally, unequal sex ratio does not have a big effect unless highly skewed

Nonrandom number of offspring

• Variation in reproductive success

• V_k=(sum(k_i-avgk)² /N

• N_e=(4N-2)/(2+V_k)

• If every parent produces the same number of offspring, the variance (V_k) = 0

Variance in reproductive success

• Same equation as nonrandom number of offspring

Fluctuating population size

• Because H is lost at a rate of 1/2 Ne, years with small population sizes will have large effect on H loss

• As a result, average population size is not representative of H lost over years with fluctuating population size

• Example: effect of fluctuating population size on H

  • Consider 3 generations with N1=100, N2=2, and N3=100

  • At what rate will H be lost in each generation?

  • H is lost at a rateof 1/2N per generation

• Averages do not reflect the situation when there is drastic fluctuation

  • We need harmonic mean

  • N_e=t/sum(1/N_i)

  • t=generations

• harmonic mean is much lower than the regular average

Overlapping (discrete) generations

• Reduces Ne if there is variance in reproductive success and some individuals reproduce over many generations

• Heterozygosity loss is calculated over generations

• Generation interval = mean age of parents

Estimating Ne from genetic data

• Based on genetic drift

1. Temporal change in heterozygosity

   • Needs at least 2 time points

   • Look at changes in allele frequency

2. Measure linkage disequilibrium (LD)

   • Genetic linkage: physical association of genes on a chromosome within an individual: distance-based

   • Linkage disequilibrium: non-random associations between genes of different loci within a population

     • By chance, genes on different chromosomes can be linked due to genetic drift. Their allele frequencies will be correlated

     • Rate of linkage disequilibrium increases with decreasing Ne

   • Ne may not correlate with heterozygosity when there is gene flow (dispersal)

What is a target Ne for conservation

• 50-500 rule for minimum viable populations

• Short term: Ne>50

  • Based on theoretical predictions of heterozygosity loss and inbreeding risk

  • To minimize inbreeding depression by keeping delta f <1% per generation

• Long term: Ne>500

  • To maintain adaptive potential and prevent accumulation of deleterious recessive alleles

3/10/2025

Population subdivision

• As populations become subdivided, genetic consequences are a result of opposing forces of genetic drift and gene flow

• Panmixia: everything is connected completely in a population

• Isolation: opposite of panmixia, there are distinct subdivisions with no gene flow

• Effective dispersal = gene flow (AKA migration)

  • Mixing homogenized effects of genetic drift

Population subdivision in red -cockases woodpeckers

• Very little gene flow among sites

• Small populations have lower heterozygosity than large population

• With subdivisions, heterozygosity decreases within subpopulations (increased genetic drift), and differentiation occurs among the subpops (decreased gene flow)

• Bigger populations are more similar to each other

• Really small populations are more divergent, and the ones that are far from large populations are the most divergent

Relationship of genetic and geographic distance varies among species

• As geographic distance increases, so does genetic distance

• What are some plausible explanations for this variation?

  • Could be something blocking dispersal

• Why is it important to understand patterns of population divergence?

  • Important for translocations

  • Understanding local adaptations

  • Identifying management units

  • Identifying if a population needs intervention

How do we quantify subdivision

• Using F statistics

  • Partitioning variance withinand among subpopulations of the whole population

• Population-level inbreeding coefficient: individuals within local population will be inbred (have more common ancestry) relative to individuals drawn randomly from larger population as a whole

• Measure reduction in heterozygosity relative to HWE

  • F=1-(H0/He)

  • He is expected heterozygosity under HWE

  • H0 is observed heterozygosity

• F_IS=1-(H0/Hs)

  • Looks at relationship between individuals and their subpopulation

  • When FIS is > 0

    • Excess of homozygotes

    • Inbreeding within a local population

    • assortative mating

  • When FIS is < 0

    • Typically only occurs in very small populations

    • Exceess of heterozygotes

    • Indicates bottleneck - weird

• Fst=1-(Hs/Ht)

  • Inbreeding due to genetic divergence among subpopulations relative to the total populations

  • Metric of gene flow

  • Can be taken for an overall population by averaging the Hs for the subpopulations

  • If there was complete mixing, it would be 0

• Ranges more different from other subpopulations

• When you have a high Fis, you usually have a high Fst

• Population size matters for within population diversity as well as between!

Partitioning heterozygosity

• H_s=2pq — where p and q are the alleles in a sub population

  • Expected heterozygosity within a subpopulation

• H_t=2avg(p)avg(q) — calculates heterozygosity in a whole population from p and q in subpopulations

• H0= observed heterozygosity within the subpopulation

Gene flow

• Gene flow homogenizes effects of drift

• Reduces differences between populations

• Island model of gene flow

  • Assumes each population contributes the same number (m) of migrants to the gene pool

  • With this model:

    Fst ~ 1/(4mN + 1)

  • Tells us how Fst relates to gene flow

  • When Nm=1, FST = 0.2

  • “one migrant per generation” rule

  • it takes very little gene flow to connect populations — you might only need a single individual migrating between populations ber generations

• Stepping stone model of gene flow

  • Assumes that dispersal and gene flow only happens in series — from one stone to the next

  • This model highlights the importance of intermediate habitats, as they facilitate the movement of individuals and genetic material, ultimately enhancing the genetic diversity of isolated populations.

  • Gene flow (m) is greater between nearby subpopulations than those farther apart

  • Takes more migration than the island model to cause the same amount of genetic exchange

• Isolation by distance

  • Correlation of genetic and geographic distance

  • Results from stepping stone model

  • Relationship of Fst and geographic distance reveals scale of connectivity