The pros and cons of GWAS

Summary

This video introduces genome-wide association studies (GWAS), which identify gene loci associated with traits or phenotypes. The host explains the pros and cons of GWAS, emphasizing its hypothesis-free nature and its potential therapeutic benefits. Limitations such as missing heritability and a lack of diversity in the studies are discussed, along with the need for further experiments to enhance accuracy. The video also touches on polygenic risk scores, which predict the likelihood of traits like obesity, and stresses the importance of making these scores actionable and independent of existing information.

Takeaways

• 🧬 Genome-wide Association Studies (GWAS) are used to identify gene loci associated with a trait or phenotype of interest.
• 🔍 GWAS involves scanning the entire genome to find single nucleotide polymorphisms (SNPs) that may be associated with a particular trait.
• 🌟 One of the main benefits of GWAS is that it is hypothesis-free, allowing researchers to identify potential gene associations without prior knowledge.
• 💊 GWAS can have therapeutic potential, particularly when combined with other data like polygenic risk scores, which can predict the likelihood of certain conditions.
• 📈 The data for GWAS is becoming more accessible due to the prevalence of sequencing data, making these studies technically easier to conduct.
• 🔄 Combining GWAS with other studies, like single-cell RNA sequencing, can help overcome some of the limitations and provide a fuller understanding of gene expression.
• ⚠️ A key limitation of GWAS is the issue of 'missing heritability', where identified genetic variants only account for a small percentage of the genetic influence on a phenotype.
• 🔎 GWAS often identifies loci rather than specific genes, which can be challenging as it's the gene function that researchers are typically more interested in understanding.
• 🌐 There is a lack of diversity in many GWAS, which can limit the applicability of findings to different populations and may contribute to 'missing heritability'.
• 🏥 Polygenic risk scores generated from GWAS can be used to predict an individual's risk for certain diseases or conditions, potentially enabling early intervention.

Q & A

What are genome-wide association studies (GWAS)?

Genome-wide association studies (GWAS) are used to identify gene loci associated with a specific trait or phenotype by analyzing genetic variations across the entire genome. These studies look for single nucleotide polymorphisms (SNPs) that may be linked to certain traits.

What is the main advantage of genome-wide association studies?

One major advantage of GWAS is that they are hypothesis-free. Researchers do not need prior knowledge about the genome, making it possible to identify genetic variants that are likely associated with a particular trait.

What are some limitations of genome-wide association studies?

Key limitations include the inability to establish causation, the issue of missing heritability, the challenge of identifying rare variants, and the potential for identifying non-coding regions or loci without clearly linked genes.

What does 'missing heritability' mean in the context of GWAS?

Missing heritability refers to the gap between the genetic variants identified through GWAS and the full genetic explanation for a trait. The variants typically explain only a small percentage of the heritability, leaving most of the genetic factors unaccounted for.

How can combining GWAS with other studies improve results?

Combining GWAS with studies such as single-cell RNA sequencing can help researchers better understand which genes are expressed in specific tissues and confirm the association between genetic variants and phenotypes, enhancing the reliability of findings.

Why is the lack of diversity in GWAS a problem?

Most GWAS have used sequencing data primarily from European populations, which may limit the generalizability of results. Including more diverse ethnic groups could help address missing heritability and provide more comprehensive insights into genetic associations.

What is a polygenic risk score, and how is it related to GWAS?

A polygenic risk score is a measure that combines the effects of multiple genetic variants identified through GWAS to estimate an individual's likelihood of developing a specific trait or disease. It can be used for early disease prediction and prevention.

What are some potential applications of polygenic risk scores?

Polygenic risk scores can help predict the likelihood of developing conditions such as obesity, cancer, or other complex diseases. By identifying individuals at higher risk, early intervention and preventive measures can be implemented.

What is the main challenge with using polygenic risk scores for actionable insights?

The challenge lies in ensuring that polygenic risk scores provide independent and actionable information beyond what is already known. For example, if someone is already obese, a risk score confirming their likelihood of obesity offers little new information.

How can rare genetic variants influence the effectiveness of GWAS?

GWAS typically focus on common genetic variants, but rare variants may have a greater effect size on traits. These rare variants might be missed in standard GWAS but could play a significant role in explaining missing heritability.

Outlines

00:00

🧬 Introduction to Genome-Wide Association Studies

This paragraph introduces genome-wide association studies (GWAS), explaining their purpose and methodology. GWAS are used to identify gene loci associated with a specific trait or phenotype by scanning the entire genome for single nucleotide polymorphisms (SNPs) that may correlate with the trait. The paragraph provides an example of how a SNP could be associated with tallness. It also mentions previous GWAS examples, such as those related to chronotype and obesity, and a recent study on same-sex sexual behavior. The pros of GWAS include being hypothesis-free, having therapeutic potential, and being technically easy to perform due to the prevalence of sequencing data. The paragraph also touches on the importance of combining GWAS data with other studies, like single-cell RNA sequencing, to gain a fuller understanding of the genetic associations.

05:02

🔎 Limitations and Potential of Genome-Wide Association Studies

The second paragraph delves into the limitations of GWAS, such as the challenge of associating genetic variants with causation rather than just correlation. It discusses the issue of 'missing heritability,' where the sum of identified genetic variants only accounts for a small percentage of the likelihood of a phenotype. The paragraph also addresses the problem of identifying common variants over rare ones, which might have a greater effect size. It suggests that combining GWAS with other experiments, like RNA sequencing, can help overcome some of these limitations. The paragraph also highlights the importance of including diverse populations in GWAS to improve the understanding of genetic associations and heritability. It concludes by discussing the potential of GWAS in generating polygenic risk scores, which can predict the likelihood of developing certain conditions or phenotypes, and the importance of these scores being actionable and independent of other available information.

10:04

📝 Conclusion on Genome-Wide Association Studies

The final paragraph summarizes the discussion on GWAS, emphasizing their benefits and current limitations. It reiterates the potential of GWAS in conjunction with other studies to advance genetic research and the importance of addressing the limitations to fully harness their capabilities. The paragraph ends on a note of thanks to the audience for their attention, indicating the conclusion of the video script.

Keywords

💡Genome-wide Association Studies (GWAS)

Genome-wide Association Studies (GWAS) are research methods used to identify genetic variations (often single nucleotide polymorphisms, or SNPs) associated with a particular trait or phenotype across the entire genome. In the video, GWAS is described as a tool to find loci related to traits like height or disease susceptibility without needing prior hypotheses. This technique is key to understanding genetic influences on traits.

💡Single Nucleotide Polymorphisms (SNPs)

Single Nucleotide Polymorphisms (SNPs) are genetic variations where a single nucleotide (A, T, C, or G) in the genome is altered. In the context of GWAS, SNPs are the primary focus when identifying genetic markers associated with traits like obesity or chronotype. SNPs are used as markers to locate regions in the genome that correlate with specific phenotypes.

💡Phenotype

A phenotype is an observable trait or characteristic, such as height, eye color, or susceptibility to a disease. GWAS aims to connect specific genetic variants (SNPs) to particular phenotypes. For example, the video mentions traits like obesity or same-sex sexual behavior as phenotypes studied using GWAS.

💡Polygenic Risk Scores

Polygenic risk scores estimate an individual's likelihood of developing a certain trait or condition based on the sum of multiple genetic variants. The video explains that polygenic risk scores can predict traits like body mass index (BMI) by analyzing multiple SNPs identified through GWAS, potentially aiding early disease identification and prevention.

💡Loci

Loci are specific, fixed positions on a chromosome where a particular gene or genetic marker is located. In GWAS, researchers look for loci where genetic variations (like SNPs) are associated with traits of interest. The video provides an example of loci linked to traits such as height or obesity.

💡Missing Heritability

Missing heritability refers to the gap between the genetic variations identified in studies like GWAS and the total heritability of a trait. In the video, it is explained that even when loci are identified through GWAS, they often account for only a small percentage of a trait’s heritability, leaving much of it unexplained. This phenomenon is known as missing heritability.

💡Hypotheses-Free

Hypotheses-free means that researchers do not need a preconceived idea or theory about which genes are linked to a trait when conducting GWAS. This is one of the benefits of GWAS mentioned in the video, as it allows for broad, unbiased searches across the genome to identify new loci associated with traits.

💡Therapeutic Potential

Therapeutic potential refers to the possibility of using research findings, such as those from GWAS, to develop new treatments or interventions for diseases. The video suggests that identifying genetic loci associated with diseases through GWAS could lead to targeted therapies in the future, especially when combined with polygenic risk scores.

💡Linkage Disequilibrium

Linkage disequilibrium describes the non-random association of alleles at different loci. It is mentioned in the video as a challenge in GWAS, where multiple genetic variants in a region may appear associated with a trait, making it difficult to pinpoint the exact variant causing the phenotype. It complicates the process of understanding the genetic basis of traits.

💡RNA Sequencing

RNA sequencing is a method used to analyze the quantity and sequences of RNA in a sample, providing insights into which genes are active in different cells and tissues. In the video, RNA sequencing is suggested as a complementary tool to GWAS for better understanding gene expression and verifying the relevance of the loci identified, improving confidence in the results.

Highlights

Introduction to genome-wide association studies (GWAS) and their role in identifying gene loci associated with traits or phenotypes.

GWAS are hypothesis-free, meaning they don't require prior knowledge to identify gene loci associated with a phenotype.

GWAS can have therapeutic potential, particularly when combined with polygenic risk scores to predict disease likelihood.

Advancements in sequencing technology have made it easier to access the data required for GWAS, increasing their prevalence.

Combining GWAS with other experiments, such as single-cell RNA sequencing, provides a more comprehensive understanding of gene expression and its relationship with traits.

One major limitation of GWAS is that identifying an association between a gene and a trait does not imply causation.

The concept of 'missing heritability' refers to the fact that GWAS often explain only a small percentage of the genetic variance for complex traits.

GWAS tend to miss rare variants, which may have a greater effect on the trait of interest compared to common variants.

Another limitation of GWAS is that they often identify loci without identifying the gene itself, making it harder to understand the function and role of the gene.

There is a lack of diversity in GWAS, with many studies relying predominantly on European populations, which can limit the generalizability of findings.

Polygenic risk scores, derived from GWAS, can be used to estimate the likelihood of an individual developing a certain disease or phenotype.

Polygenic risk scores could enable earlier detection of diseases and more personalized treatments based on genetic risk.

The actionable potential of polygenic risk scores is still being evaluated, as they must offer independent predictive value beyond existing information.

An example of actionable polygenic risk scores is the use of these scores to assess the risk of obesity, which could lead to targeted interventions.

The future of GWAS and polygenic risk scores depends on addressing limitations such as rare variant identification, heritability gaps, and study diversity.

Transcripts

hello and welcome back to the cheeky

sign jus so in today's video we're going

to look at the pros and cons of

genome-wide Association studies so

before I jump straight in and tell you

about the pros and cons I'll first

introduce you to what are genome-wide

Association studies and then after

looking at the pros and cons we'll look

at what polygenic risk scores are and

how genome-wide Association studies are

used to look at putting at risk scores

so firstly what are genome-wide

Association studies so genome-wide

Association studies are used to identify

gene loci associated with a trait or

phenotype of interest so the idea is

that you look across the entire genome

and try to identify single nucleotide

polymorphisms as you can see here with T

a and G and whether or not the T the a

witha G is associated with a phenotype

of interest so let's say people who had

a T instead of an a were also had the

phenotype of being really tall and so

that could be a low save associated with

the traits of being tool that make sense

and so if you do this across the entire

genome you'll find that certain leucite

are more likely to be associated with

that phenotype than others and ones that

surpass the threshold are of interest

because it suggests that that is a low

site why there was a correlation between

having that genetic variant and the

traits so in previous videos I've

already given examples of where

genome-wide Association studies have

been done for example I gave an example

of where it's been done with chronotype

and also with obesity but this paper

here is a recent paper that's forces on

a genome-wide Association study with

Lisa associated with same-sex sexual

behavior and so my point is is that they

are happening all the time and can be

done with many different phenotypes but

why do them what are the benefits of

genome-wide Association studies so

firstly the beauty of genome-wide

Association studies are that they are

hypotheses free

you don't need to know anything

beforehand and so starting from nothing

you can identify a site that have genes

that are likely associated with the

phenotype that you're studying and then

these genes can be further characterized

and understood to understand what causes

the phenotype of interest

wow that was a long-winded explanation

but genome-wide Association studies may

also have therapeutic potential as we'll

see later when we look at polygenic

Briscoes and lastly the data for

genome-wide Association studies requires

sequencing data which is becoming ever

more prevalent and so is technically

quite easy to get now or at least easier

than it was previously

and these datasets can be even more

informative when they're combined with

other studies as well and actually

combining genome-wide Association

studies with other studies helps they

become some of the limitations that I've

seen with genome-wide Association

studies so one experiment that I mean is

single-cell RNA sequencing data because

with that information you can see which

genes are being expressed in a cell and

which shells from assassin tissue and so

if you can see the same subset of genes

being expressed as the same subset of

genes identified from a genome-wide

Association study you have a fuller

understanding and better confidence that

the genes that you're looking at are

possibly associated with the phenotype

that you're studying it would make more

sense probably if you had an example but

bear with me for now but there are

downsides to genome-wide Association

studies the first thing the keywords

Association Association of valais site

with the traits doesn't necessarily mean

causation and so the genes that are

being identified may have new world

value in terms of understanding the

cause of a phenotype so the next

limitation with genome-wide Association

studies is the missing heritability and

so what I mean by this is that if you

took all the low-side that surpassed the

thresholds of being associated with that

phenotype and you added them all

together the total sum of all those

those I in terms of the likelihood of

you getting that phenotype condition

that sees whatever you're looking at is

only around like two to three percent

maximum for most of these genome-wide

Association studies so what about you

know that the other 97% where is this

missing heritability and so part of this

issue comes from the fact that the low

site with like a strong link with the

phenotype I'm being identified and that

comes down to a kind of a flaw we're not

really a floor-by-floor of the

genome-wide Association studies that you

have your population of different whole

genome sequences that you're looking at

and you're trying to find lo site in all

of them all of the people with the

condition that have that trains that

have you know what I mean and the

problem is is if you're looking for

common variants you're going to miss the

rare variants that let's say one or two

people who have the condition also have

that variance and it seems possible that

these rare variants are the ones that

have a greater effect size as you can

see in the graph here but there is still

hope that these rare variants can be

identified given that now we have the

tools to easily from whole genome

sequencing and with this knowledge that

we could be missing them we're more

likely to look for the reference so

another limitation with genome-wide

Association studies is that it often

identifies loose I where you have a

genetic variant but not a gene why are

it's the genes that were more interested

in because then we can understand the

function of the gene where the genes

expressed I understand how that links it

to the phenotype and so certain reasons

could be that the low size and a

non-coding region in which case it could

be affecting the expression also not

just one gene but multiple genes that

altogether could be influenced in the

phenotype um there's also a linkage

disequilibrium whereby there might be

more than one gene

in that same region and/or there's so is

more to do the fact that there could be

multiple variants within a certain low

sign is determining which of those

variants is responsible and then lastly

is looking at which tissues important in

terms of the gene so as I said there

might be a gene that you've identified

but in might only be expressed and

Sasson tissues and so to get answers to

these different points you need to

combine genome-wide Association studies

with other experiments such as the RNA

sequencing data that I spoke about

earlier and another important issue that

is now being better addressed is the

lack of diversity in the studies at the

moment so a lot of the genome-wide

Association studies have used sequencing

data from mainly European populations

and say by including this including more

ethnicities into these studies could

also explain why there seems to be a

lack of heritability in the results and

so it would be valuable to include them

and so if some of these limitations can

be addressed there's great excitement

for genome-wide Association studies for

being used to generate polygenic riscos

whereby they can take your genetic

information and look at which less

variants you've got and therefore your

risk or likelihood of getting a certain

disease or phenotype and so this could

be the likelihood that you could get a

certain type of cancer and this

predictive information kids enable

better identification and early

identification of certain diseases which

will leads a better time available for

treatment and so by taking body mass

index as an example the greater your

polygenic risk or the greater the

prediction of your BMI is and so if you

did this across a population you would

see a bell curve for the distribution or

Fresco's and so the risk score could

have some

kind of predictive value for assessing

people who are high risk of having a

high BMI and getting obesity and

therefore could be targeted through

effective treatment but that's the key

thing that I need to emphasize it's

about whether or not a risk score can be

actionable and so this was actually

already done in a recent paper whereby

they used a genome-wide polygenic risk

score to quantify the inherited

susceptibility to obesity and they did

this to see if they could identify

adults at risk of severe obesity and the

idea is that if you can then

you can try and help to prevent that

from happening but the other key thing

is that these polygenic risk scores have

to be independent of other information

that you can already get so if you were

going to help somebody and that person

was already obese and getting a probably

done at risk score that tells you that

they're going to be obese is already

very much help then is it because it's

already gonna be obvious so it's all

about whether or not play gentle risk

scores are going to be actionable and

can be independent of other sources of

information that we can already attained

so I hope this video has given you a

good introduction to what genome-wide

Association studies are and their

benefits and the current limitations

with them but also what we can to you

with them so thanks for listening