Topic 5: Quantitative Genetics PDF

Topic 5: Quantitative Genetics Content Owned By University of Sydney Notes By Miranda Memmolo  ➔ Understand what quantitative traits are and contrast to qualitative traits ➔ Interpret experimental results, such as Edwards East's Experiment in Nicotiana longiflora...

Topic 5: Quantitative Genetics Content Owned By University of Sydney Notes By Miranda Memmolo  ➔ Understand what quantitative traits are and contrast to qualitative traits ➔ Interpret experimental results, such as Edwards East's Experiment in Nicotiana longiflora ➔ Understand how experimental and genetic factors impact on an observed trait Thinking About Height How to describe phenotype? Does it make sense to categorise? ○ Tall, less tall, medium, short, very short? How many genes might affect height? Other factors affect height? Mendel did look at height of pea plants, but in that particular case, the plants were easily categorised as either tall or short, rather than continuous like you would see in other examples, like these meerkats. The reason for there being these two categories in the pea plants is because height is determined by alleles in one locus (Gibberellin 3 beta-hydroxylase) whereas you will see in this lecture what happens when there are more. Example: Ladybug Spots Alleles – In terms of Mendel’s Laws One Locus Two Locus Three Locus To keep simple Same idea, now with two loci Three loci: A, B, and C No dominance Two loci: A and B Seven phenotypes Five phenotypes Each uppercase allele leads to an increase to some trait E.g. for the ladybug, it adds one spot, indicated as 1+ per allele One locus: A Three phenotypes When there was one locus, there were not many phenotypes. As you get more loci, with different level of effects, you get an increase in the number of phenotypes to the point that it becomes a continuous distribution Ignoring dominance, imagine an example where No need to memorise this! Just appreciate that some each allele indicated with a capital letter results in traits are controlled by many loci and that the number of phenotypes will increase very quickly as an increase in some trait, e.g., pigment, height, number of loci increases. flower length. Make up whatever example you It’s more complicated if it is not just additive. Other want. factors? (e.g., effects of environment as you will see In this case, we can think of just one spot on a in the next few slides) ladybug per allele. Over the next two slides, we can see the different genotypes and phenotypes as the number of loci increases Real Example: Kernel Colour of Wheat Alleles at two loci Nilsson-Ehle’s experiment A bit of luck: ○ Only 2 or 3 loci ○ Genes were all additive ○ Almost no effect of environment You May Notice This gets complicated very quickly In reality, different genes can also have different effect sizes, i.e. effect of A is not necessarily the same as effect sizes i.e. effect of A is not necessarily the same as effect of B Allele frequencies can also vary Edward East’s Experiment: Flower Length in Nicotiana longiflora What factors influence corolla ( flower) length? 1. Generate two pure-breeding lines: one with short corolla, and one with long corolla 2. Crossed these to generate F1 Result: lengths between that of the parents, and with similar variation in phenotypes 3. Crossed F1 to generate the F2 Result: average length was about the same as F1 but greater variation in phenotypes 4. Selected the plants from the upper and lower ends for three generations (F5) Result: lengths similar to parental. Again, greater variation in phenotypes It’s important to understand what was done in this experiment and what the results mean. The figure describes Edward East’s experiment in Nicotiana longiflora where he bred these plants and measured the corolla ( flower) length. He started by breeding the plants for many generations to generate two pure-breeding lines: one with short corolla (around 40 mm), and one with long corolla (around 90 mm). He crossed these together for the F1, generating plants with corolla lengths of about 65 mm. Variance was similar. He then crossed the F1 to generate F2. generating plants where the average length was about the same as F1, with a greater variance. After generating F2, East bred the plants from the upper and lower ends over three generations to generate plants with short and long corolla. These lengths were similar to the parental. So to answer the previous slide, genotype and environment influence flower length. And no, there are no distinct categories of flower length. Even for a single genotype, there are a range of lengths due to variation within the environment. When I say genetics are identical, I mean in the short corolla lines, those plants are identical. In the long corolla lines, those plants are also identical. Then the F1 is heterozygous of these, so again, this group of plants is genetically identical. We can see that there are environmental factors influencing the length (e.g., location, water, light), as seen by the fact that there is not just a single phenotype for the pure-breeding lines and F1. Contrast this to what you saw previously with the flie. Alleles segregate and assort independently so by chance, some individuals might have more of the alleles associated with longer corolla, and some might have alleles associated with shorter corolla Effects of Environmental Variation Due to environmental variation: A single genotype can give a continuous distribution of phenotypes But the same phenotype can be a result of different genotypes If individuals of a population have different phenotypes: Due to genetic differences? Or environmental differences? The variance (the spread on the frequency distribution) of the phenotype increases as the variance of the environment increases. For example, in controlled conditions versus in the wild. Like what you saw previously with qualitative traits. You get discrete phenotypes for each genotype. If there is some environmental variation, e.g., access to resources for growth, then you get variations in the phenotype even if the genotype is the same. Very important to understand how much is due to genetic variation when considering breeding. Phenotypic Variance (VP) Phenotypic variation is dependent on genetic and environmental variation. Note about the different generations: Parents: The genetic variance becomes tiny after all these generations of breeding to generate these two pure-breeding lines, so VG = 0 in each of these lines. The different phenotypes are due to the environmental variation. Even in controlled conditions, it is hard to get a situation where VE = 0. F1: As he has just crossed the two pure-breeding lines, VG = 0 here also. Similar effect of environmental variation. F2: There is also genetic variation in this generation as alleles segregate and assort independently (VG is not 0). As before, environmental variation still present. Hence, the increase variation in lengths. F5: Selecting the upper and lower ends led to two lines where the lengths were similar to the parental. As with F2 there is genetic variance in the F5 (VG is not 0) as the alleles of different genes segregate and assort independently. There are also the environmental factors. We can see that there are environmental factors influencing the length (e.g., location, water, light), as seen by the fact that there is not just a single phenotype for the pure-breeding lines and F1. Genetic-Environmental Interaction Effect of environment can differ depending on genotype Example: ○ AA grows bigger than aa in dry environments ○ aa grows bigger than AA in wet environments VGxE = variance of genetic-environmental interaction VG×E = 0 if there is no interaction. All the examples I will use assume it is 0, but just noting here so you are aware of this. 24.16 Genetic–environmental interaction variance arises when the effect of a gene depends on the specific environment in which the organism is found. In this theoretical example, the genotype affects plant weight, but the environmental conditions determine which genotype produces the heavier plant. Summary Qualitative vs Quantitative Traits Everything you have seen so far have been qualitative traits. These are traits that you can categorise and count. No effect from environment so phenotypes are discrete and populations only differ in frequencies of genotypes and phenotypes Like the example with East’s experiment, quantitative traits often depend on the environment. Genetics can be similar but still result in different phenotypes. Can’t just look at genotype and phenotype frequencies – need to look at means and variances. We will see how we work with these in upcoming lectures. Studying Quantitative Traits Cannot count allele frequencies Need quantitative measurements ○ height/length ○ Weight And statistics ○ Mean (average) ○ Variance/standard deviation ○ Analysis of Variation ○ Correlation/Regression You will not be expected to calculate these or perform these tests in this unit – just understand how to work with and interpret given values. Cannot use Hardy-Weinberg here because you can’t count allele frequencies. Statistics is really important in science. Whenever we test something, there is only so much we can do, whether it be sample size, number of times we repeat an experiment, etc. Statistics gives us information about how confident we can be about what we are observing. Sampling must, of course, be random! Can start to see the bell curve from the previous example, with 6+ and 0+ at the extremes and 3+ in the mean. This is without considering environmental effects. Another example is the frequency distribution of heights in men and women. There will be genetic and environmental effects leading to this continuous distribution of heights. Note also that distributions may not necessarily be normal distributions, e.g., may be skewed or bimodal Example: Phenotypic Variance 1. You have two pure-breeding lines of Nicotiana: ➔ One with an average length of 40.47 mm (VP = 3.12 mm) ➔ Another with average length of 93.75 mm (VP = 3.87 mm) 2. You cross this to generate F1: ➔ Average length of 63.90 mm (VP = 4.74 mm) 3. You cross F1 to generate F2: ➔ Average length of 68.72 (VP = 47.70 mm) Assuming there is no genetic-environmental interaction 1. What is the environmental variance (VE) of this trait? VG = 0 in the pure-breeding parental and the F1 heterozygotes. Thus, VP = VE in these generations, to get VE of 3.12, 3.87 and 4.74. Taking the average of these values gives the estimate VE = 3.91 mm 2. Determine the genetic variance (VG) in the F2. In F2 VG ≠ 0 so VP = VG + VE. We are given VP = 47.70 and we calculated VE = 3.91. Thus rearranging VG = VP – VE = 47.70 – 3.91 so VG = 43.79 mm Practise Short Answer Question Q: define the basic model of quantitative genetics and explain the meaning of its components. A: The basic model of quantitative genetics is represented by the equation P = G + E, where P represents the phenotypic value, G represents the genotypic value, and E represents the environmental value. This model describes how the phenotype of an individual is determined by the interaction between its genotype and the environment in which it develops. Comment about this response? Do you see any problems? (Hint: there are!) Note: I would never ask for the ‘basic model’ in an exam because it’s vague. What does ‘value’ mean? It should be phenotypic variance, genotypic variance and environmental variance. Important: it is about variances! When you have a population and look at a particular trait and see variances in that trait, e.g., weight, that variation in the phenotype will be due to variation in the environment and variation in the genotype (unless it is a pure breeding line, for example)  Heritability ➔ Explain what heritability is (both broad-sense and narrow-sense) ➔ Understand how trait heritability can be estimated ➔ Predict response to selection using heritability values Genetic or Environmental Variation? How much of this is genetic variation and how much is environmental variation? Recall from previous lecture: phenotypic variation is a combination of genetic and environmental variation. Here are a few examples of families who have excelled in various areas. How much of it is genetic variation and how much is environmental variation (e.g., how they were raised, resources, etc.) 1. The Curie family has five Nobel Prizes! (Marie Curie in physics and chemistry, Pierre Curie physics, Irène Joliot-Curie in chemistry, Frédéric Joliot-Curie in chemistry.) 2. László started teaching his daughters at a young age as an experiment to see whether they could be brought up to be prodigies. Not entirely sure if he thought he could get anything conclusive with just one family, but still interesting! Judit Polgár is the strongest female chess player of all time. 3. Parents and aunt also canoeists Considerations for Selective Breeding How effective will our attempts to selectively breed be? If we have individuals with a desired phenotype and breed them, how much will the offspring resemble the parents? How much of the phenotypic variation is due to genetic variation? If a desired phenotype, e.g., a particular fur colour, is not due to genetic variation, but is instead due to environmental variation, then breeding individuals with that phenotype will not be helpful Heritability What proportion of the phenotypic variation is due to genetic variation? Relates to a specific trait, in a specific population, in a specific environment Heritability will always be in reference to a particular population in the particular environment Heritability can be described using: ○ Broad-sense heritability, H2 ○ Narrow-sense heritability, h2 Gives us information about whether the phenotype e.g. big tomatoes is due to variation in genetics or variation in the environment. Different traits will give different values for heritability. The same trait can have different heritability values in different populations and environments, as genetics and environment will be different. We are looking for patterns in populations, not individuals. Heritability for a particular population is also not fixed and can also change as the environment changes. E.g., for a trait with heritability of 1 does not mean that it cannot be affected by the environment if it changes. Also important to note that we are looking at environmental and genetic variation. These values do not tell us what the genes are doing. Higher heritability means greater proportion from genetic component. Relatives will more likely share that trait. Lower means more environmental. Relatives won’t necessarily share that trait because it is dependent on the environment. For example, for the large tomatoes, are the genetics of those plants different to those of the smaller tomatoes? Does breeding the plants with large tomatoes lead to large tomatoes? Or is it variation in the environment, such as watering, light, fertiliser, etc.? Broad-sense Heritability Narrow-sense Heritability An estimate of the proportion of the phenotypic An estimate of the proportion of the phenotypic variation that is due to total genetic variation: variation that is due to additive genetic variation: Examples: Additive genetic variance is mainly what ○ Coat colour in cattle = H2 close to 1 determines resemblance  ○ Milk production in Holsteins = H2 between If we are selecting using homozygotes, we can 0 and 1 focus on the addictive part and use narrow-sense Cattle coat colour is simpler, closer to 1. i.e., variations in coat colour are heritability almost entirely due to genetic variations Now we are looking at the proportion that is due to additive variance VA only More variation in milk production, heritability can vary. If 0 then it’s due to environmental variance, if 1 then it’s due to genetic variance. by dividing that value by VP. Breeding homozygotes, such as in selective breeding, means that we can also just focus on additive variance. Response to selection is dependent on additive genetic variance so it is useful in selective breeding. If we want to improve a crop or agricultural species, then if we have something with a desirable phenotype that we want more of in our population, if we know it has a high h2, then that individual is valuable for breeding. If it is low, then it could be due to some environmental effect. E.g., if we know a particular plant is growing taller because of lots of water rather than genetics, then breeding it without giving it lots of water will not produce tall plants. Important to inform how a breeder would breed their populations. Genetic Variation Genetic variation can be broken down into: ○ Additive genetic variance, VA = additive effect of the alleles ○ Dominance genetic variance, VD = i.e. where Aa does not give an intermediate between AA and aa ○ Gene interaction variance, VI = epistatic effects In agriculture and breeding: will breeding individuals with a desired trait lead to offspring with this trait? Selective breeding is performed with homozygotes ○ Breeding using heterozygotes will also give you homozygotes for alleles you do not want If complete dominance, then AA and Aa give the same phenotype. Think back to what you learnt in Emily’s lectures around haplosufficiency. Otherwise, Aa may give some intermediate between aa and AA (depending on the gene). Selecting using homozygotes is more effective because you are just working with the alleles you want. If you have the heterozygotes you will also end up with homozygotes of the alleles that you don’t want. A simplified way to visualise this is to think about all the combinations you get when you draw a punnett square for a dihybrid cross of heterozygotes. Now, scale this up to the number of loci involved in quantitative traits Additive genetic variance is mainly what determines resemblance between parent and offspring so if we are selectively breeding, we can just look at additive genetic variance and select for the additive alleles. Example: Heritability You have calculated the following variance from your herd of cattle: 1. Calculate the broad-sense (H2) and narrow-sense (h2) heritabilities of each trait. Body Size Milk Production 2 23.2 2 16.8 𝐻 = 51.3 = 0. 45 𝐻 = 42.5 = 0. 40 2 10.18 2 12.32 ℎ = 51.3 = 0. 20 ℎ = 42.5 = 0. 30 2. What would you recommend to a breeder considering this breed for milk production? Given h2 is 0.30 for milk production it means there is potential for selective breeding. In other words 30% of the phenotypic variation is due to additive genetic variation. Note About h2 No need to know what the “cut-off” is for heritability for selective breeding Just understand that if it is really low (close to 0), then you could not use selective breeding: i.e. if h2 is close to 0, then the variation you see in the phenotypes is almost completely due to variations in the environment. Other genetic variation also plays a part but these effects are smaller. Any exam questions about this will be clearly very close to 0, or not. Estimating h2 Breed individuals of a population in a given environment and look at a specific trait e.g. mean, distribution If you breed individuals from the extremes, what does the distribution of the offspring look like? i i.e. looking at correlation between parental and means of offspring Remember, we are estimating for a specific trait, population, and environment. Looking at the result of breeding and using mean, distribution -> stats to estimate h2. Visualised with Llamas If the mean of your population looks like this: And you select the larger ones to breed: And the mean of the offspring shifts to bigger llamas, then it suggests h2 is NOT low I say not low because it doesn’t necessarily mean it’s high, but it’s enough for the trait to be heritable. Taking individuals from the extremes, look at the distribution of offspring. If it shifts, then suggests a genetic component. Higher h2 will shift more. If the mean of your population looks like this: And you select the larger ones to breed: And the mean of the offspring remains similar to the mean of parental generation, then it suggests a low h2 Low h2 means differences are due to environmental variation, e.g., access to food Estimating h2 with Relatives Plot to get regression (b) or correlation (r) As we are interested in resemblance between relatives, we can use data from parents or siblings: ○ Midparent (x) vs offspring (y): b = h2 ○ Parent (x) vs offspring (y): b = 0.5h2 ○ Siblings: r = 0.5h2 ○ Half Siblings: r = 0.25h2 ○ You will make heritability estimates in your practical Since h2 is about resemblance between relatives, you don't necessarily need to look at parents. Looking at both parents (i.e., the average) against the offspring, the regression or slope will give the value for heritability One parent will share half the genetic material, so the slope gives half the value of heritability Correlation can be used to estimate heritability from siblings. Full siblings share an average of 50% of their genes, and half siblings share an average 25% of genes. Limitation is that these assume the environmental variation is similar to unrelated individuals, which it wouldn’t be if they live together. Response to Selection How much the trait changes from the mean Response to selection (R), is dependent on: ○ Heritability, h2 ○ Selection differential, S = how strongly you select The relationship to heritability: Look at how effective selection will be for a certain trait. If h2 is low, then it is very much environmental. No point trying to breed big plants, for example, if it is just because of something environmental such as watering. Also how strongly you select. Even if you have high h2, you need to pick the right ones to breed. E.g., picking the really big plants (away from the mean) to breed. S (selection differential) is the difference between mean of selected parents and mean that you started with/use to breed. R (response) is the difference between mean of resulting offspring and starting mean. h2 is how far your offspring is from the mean, relative to how far from the mean you initially selected. E.g., If response is only half of the selection differential (only goes half way), then h2 is 0.5. R = h2S gives you information about response. If you know heritability (using your data from the whole population, breeding history over the last few years to get pattern between parents and offspring) and how strongly you select (overall mean and mean of the parents you are using, e.g., upper end), you can predict response -> how much can you select for the trait in your population In conservation and wildlife, if you can identify the parents (e.g., from looking at genotypes), you can then look at the correlation with the offspring to estimate heritability of traits such as those related to survival. From this you can get information such as what parents to breed and how this population is likely to respond to selection Example: Estimating Heritability in Llamas (→) Practise Short Answer Question Q: Imagine a farmer who grows two varieties of corn, A and B, on two separate fields. Both fields have the same soil, climate, and are treated with the same farming practices. However, at harvest time, the farmer notices that variety A consistently produces larger and more abundant ears of corn than variety B. How might a geneticist explain this difference in terms of heritability? A: A geneticist might explain the difference in corn yield between variety A and variety B in terms of heritability by suggesting that the genes of variety A have a higher heritability for traits related to ear size and abundance. This means that the genetic makeup of variety A has a stronger influence on these traits than environmental factors, resulting in larger and more abundant ears of corn even when grown in the same conditions as variety B. Comment about this response? Do you see any problems? (Hint: there are!) Summary Heritability is the proportion of phenotypic variation that is due to genetic variation Values will differ depending on what trait, population and environment we are looking at Narrow-sense heritability (h2) considers only the additive genetic variation Useful in selective breeding ○ Predict how much the offspring resemble the parents in the traits that we want ○ We expect greater similarity for traits with higher h2  Applications ➔ Define heterosis and transgressive segregation and explain the relevance in agriculture ➔ Describe the approaches to identify markers associated with a quantitative trait ➔ Explain how data such as markers and heritability values are useful in agriculture, health and conservation Agriculture: Cattle Breeding Understanding of genetics allows us to make informed decisions to get phenotypes of interest. Going back to the example from the previous lecture where h2 = 0.3 for milk production: What might happen if you continually selected the cattle with highest milk production? Consider: ○ The alleles that are passed on ○ The effect on genetic variance ○ The effect on h2 What about if you moved the cattle to a different farm? Selecting these cattle for breeding would lead to offspring having these alleles. Over time the genetic variance would decrease as you are selecting for these alleles (there will be less of the other alleles in the population), and therefore heritability will also decrease. Remember that heritability values are specific to the environment – we cannot predict the heritability values in a different environment, nor the phenotypes. These particular alleles may be favourable for milk production in this environment, but it may be different in another. Similarly, if the environment changes, e.g., climate change, that can also impact heritability. Agriculture: Crop Breeding E.g. selectively breeding corn to increase oil percentage Notice heritability (in brackets) decreases through the generations of selection Can be used in combination to get phenotypes of interest. All this from selective breeding and quantitative genetics and choosing the right individuals to breed. If it is bred until a pure breeding line is obtained, the value of h2 will be 0. Examples of h2 in Chickens Remember heritability is in reference to a specific trait, in a specific population in a specific environment H2 varies between populations and traits Notice h2 is low for viability – heavy selection to improve these traits Some of these traits have a range of heritability values. Different populations may have different allele frequencies so genetic variance can be different. Different farms, farming systems, etc. will also change environmental variance. Heritability for viability is low, and there is a low range. To reiterate, this does not mean that there are no genes that affect these treats. The low values are due to heavy selection to improve survival. These are the ones that survive and so this reduces the genetic variation as the genetics are optimised. More examples of h2 Similarly, notice that h2 is low for traits that are important to survival Estimated Breeding Values (EBVs) Important to breed the “right” individuals to get desired traits EBVs estimate the genetic merit for traits that affect profitability e.g. weight, fertility/calving, carcase Compared against the average within a breed E.g. if you have individual 1 with EBV of +40 kg for 600 day weight and individual 2 with an EBV of +10 kg: ○ The difference between them is 30 kg ○ Individual 1 will be expected to produce offspring that is 15 kg heavier then individual 2 at 600 days Carcase is kind of like body composition. Livestock weights at different time points depending on what it will be used for. In the example, the difference between the animals is 30 kg. Each individual contributes half the alleles to the offspring, so the difference in the offspring is half, giving 15 kg on average Selection and Quantitative Traits Recall in East’s Experiment: Mean of F1 was an intermediate of the parents ○ No heterosis Range of F2 to F5 within the range of the parents ○ No transgressive segregation Remember, the parents are pure breeding lines. When crossed, the combination of alleles can result in these effects Heterosis Mean of F1 higher or lower than the parents Also referred to as hybrid vigour F1 hybrid breeding, e.g. : ○ corn/maize ○ Vegetables ○ Chickens ○ Pigs ○ Cattle Instead of an intermediate of the pure breeding parents (recall: pure breeding lines are obtained through lots of backcrosses so are mostly homozygous), it goes above/below parents ‘best’ can mean different things for different traits. Can be higher or lower. Essentially, it is past the range of the parents rather than an intermediate. Note that F1 is mostly heterozygous and genetically similar across the many genes. Whatever variation we see in F1 must be from variation in environment. This is not something that can be ‘forced’ to happen, but if it is known that a particular combination of pure breeding parents lead to these desirable traits in the (heterozygous) offspring, then these pure breeding parents can be bred to produce these. I have previously seen students suggest that heterosis could be used to improve outcomes in selective breeding, but it only works if the combinations of alleles lead to these improved phenotypes. It is also important to note that we cannot expect to see similar improvements if we breed the F1 as these phenotypes result from being heterozygous in a number of genes. Transgressive Segregation Range of F2 and later generations greater than the range of the parents If these traits are desirable, they can be bred → generate purelines with these traits Recall that from F2, you start to get some genetic variation in the offspring as this is the result of crossing heterozygotes. This may result in increased phenotypic variation, where the ends of the distribution are beyond the range of the parents. Can use the end of that range (e.g., upper) to keep breeding to generate pure lines with these high values. Quantitative Trait Loci (QTLs) Recall: quantitative traits can be dependent on many loci Although complicated, it is possible to identify these QTLs are regions of the chromosome containing loci that correlate to a traits Mapping QTLs 1. Start with 2 strains with large differences in phenotype e.g. tall plant and short plant 2. Cross to generate F1 heterozygotes 3. Cross F1 to generate F2 4. Measure the trait to identify markers that are associated Similar idea to week 4-6 practicals 24.20 Mapping quantitative trait loci by linkage analysis can identify genes that help determine differences in quantitative traits. Genotypes at the C locus are associated with the inheritance of differences in plant height, indicating that a QTL for height (the T locus) is closely linked to the C locus. Recall at the end of your week 5 practical, you compared the inheritance of the white marks phenotype to the different microsatellite loci. The loci that were associated were the ones with the same inheritance pattern as the observed phenotype. Example: QTL in Tomatoes Mapping QTLs is useful in agriculture for plants and livestock A QTL called fw2.2 was identified to be associated with variation in tomato fruit size Can produce transgenic crops with desired traits Figure 1 (A) Fruit size extremes in the genus Lycopersicon. On the left is a fruit from the wild tomato species L.pimpinellifolium, which like all other wild tomato species, bears very small fruit. On the right is a fruit from L. esculentum cv Giant Red, bred to produce extremely large tomatoes. (B) Phenotypic effect of the fw2.2 transgene in the cultivar Mogeor. Fruit is from R1 progeny of fw107 segregating for the presence (+) or absence (−) of cos50 containing the small-fruit allele. Mogeor has the large-fruit allele of fw2.2 but this is partially recessive. Transforming with the small-fruit allele reduced the size in B. Genome-wide Association Studies Revisited Another approach to identify associated loci Remember, these do NOT give you any information about causation If causation then we can look for potential drug targets There are advantages and disadvantages/limitations to each method that I won’t go into (therefore, not assessable). Just appreciate there are different approaches to study these. Role of Genetics and Environment in Disease Many diseases have genetic and environmental contributions e.g. ○ Cardiovascular disease ○ Diabetes mellitus ○ Cancer Actions taken by an individual may affect outcomes e.g ○ Diet and exercise ○ Regular check-ups Applications: Conservation To aid conservation efforts, it is important to understand how endangered species can respond to selection. Especially true as climate change leads to rapid changes in selection pressures Locations of Hihi Previously found across North Island of New Zealand Reduced population from predators and habitat loss → bottleneck Small population used to reintroduce → founder effect Genetic drift in these small populations → risk of losing alleles over time Species used to be found all over the north island of New Zealand. Severe bottlenecks have resulted from introduced mammals, predators and habitat loss (Recall: bottleneck is a massive reduction in a population, leading to changes in allele frequencies). A small population managed to survive on Little Barrier Island and were used to reintroduce to other locations, marked in orange. (Recall: founder effect e.g., from taking a small population to establish these new populations - lead to changes in allele frequencies due to sampling). Genetic drift: changes in allele frequencies in small populations by chance. Some alleles may be lost, leading to reduced genetic variation. Low Genetic Diversity Genetic diversity are lower in hihi for both measures No need to learn about these tests Comparing three hihi populations to other similar birds using different ways to measure genetic diversity (represented in light and dark bars – no need to learn specifically what they’ve done). Lower genetic diversity in hihi for both measures. Low Heritability of Traits Almost all the traits related to survival and breeding have very low heritability (h2) as they have already been selected for ➔ Variation in these traits are driven by environmental variation Looked at narrow sense heritability (look for similarity between relatives, do some stats). Shown are estimates of heritability with error bars. Recorded morphological traits that might be important for survival: mass, tarsus length (leg bone length), head-bill length. Longevity (lifespan) heritability is low – no correlation between relatives. Variation in longevity is largely driven by environmental variation, e.g., ability to acquire food, seek shelter, etc. Recruitment: whether offspring can find a territory and a partner to breed. Lay date: small genetic component, but still mainly environmental. Time to fledge: how long it takes between hatching to when it can fly away from its nest Hatching success: proportion of eggs that hatch Fledgling success: proportion of hatchlings that survive to fly away from the nest Thinking back to the earlier examples, traits that are important for survival have low heritability because it has already been selected for. There is no selection differential as the population is already surviving the best it can in that environment. Low because already at optimum. The ones that survive are the ones that are still here. Error bars represent 95% credible intervals – not necessary to know exactly what this is. It is just one of the many ways to express uncertainty. Heritability and Adaptation From previous slide, heritability for laying date is low Do you think the hihi will shift the laying date? Low genetic diversity means it is unlikely that the hihi will adapt. Effect of environmental change? E.g. climate change Fitness (number of recruited offspring) plotted against laying date. Size of circles represent the number of individuals at a given fitness and laying date. Red line is an estimation of optimal breeding time, with the dotted line at the peak (optimum of fitness), and shading for 95% credible interval. The blue line indicates time when most individuals are breeding (mode). If heritability was high then the mean would gradually shift towards the value it is being selected for/the optimum. The low heritability ( from the previous slide) indicates that it is mostly environmental variation that contributes to the variation in laying date, rather than variations in genetics. Therefore, lay date is unlikely to shift and hihi are unlikely to be able to adapt to the optimum lay date. Low heritability likely is due to the low genetic diversity (if there is low genetic variation, then the heritability will be low). Need genetic diversity for species to adapt, i.e., there needs to be different alleles in the population. That way, the individuals with the more favourable alleles will be more likely to survive and reproduce and pass these on to offspring Lay date may change in response to change in environment, e.g., climate change Working with Genetic and Biological Data Identify issues and potential strategies to overcome them? Make predictions e.g. what will be the long-term consequence for conservation? How might they respond to intervention or environmental changes? Are similar processes occurring in other threatened species? Can we predict which species will be most vulnerable? How can we protect them? Quantitative genetics data: predict how the population is likely to respond Biological data: from habitat, the species and interactions in the environment and with other species The late lay date will not necessarily lead to extinction, e.g., if they are laying enough eggs, but this is worth looking into and make predictions about what might happen long-term. The heritability of lay date is low in this population and environment. Why is October the optimum? If we can identify that, we may be able to find a more suitable habitat, for example. Can model to predict what species are most vulnerable, such as other examples of where the observed is far from the optimum. If the observed is far from optimum, there would be strong selection pressure (high S) but what is the response (R)? Practise Short Answer Question Q: A farmer is considering crossbreeding two different varieties of wheat to increase crop yield. Define heterosis and transgressive segregation and explain their relevance to the farmer’s decision. A: Heterosis is when offspring exhibit superior traits compared to their parents. This can result in increased yield, growth rate, or resistance to disease. Transgressive segregation refers to the occurrence of extreme phenotypes in offspring beyond the range of either parent. This can result in new traits that are not present in either parent. Understanding these concepts can help farmers make informed decisions when crossbreeding crops. Comment about this response? Do you see any problems? (Hint: there are!) Not sure about the idea of ‘new traits’ but rather, more extremes of the trait. Important to relate these to the parents, e.g., heterosis is means of F1 being higher or lower, transgressive segregation is the range of F2 and later generations being larger. For later generations, this is about genetic variation so can cross these Summary Mean of the F1 offspring is not necessarily an intermediate of the parents – can be higher or lower (heterosis) Range of F2 and later generations are also not necessarily within the range of the parents (transgressive segregation) ○ Can select for the extreme phenotypes Loci associated with a quantitative trait can be identified by linkage analysis or genome-wide association studies Conservation can be challenging as there may be low genetic diversity in the population ○ Heritability values for traits important for survival may be low ○ These data such as heritability can be helpful for making informed decisions