July 18, 2025 • 17 mins
🎙️ Episode 79: Cross-Population GWAS and Proteomic Insights in Atrial Fibrillation
🧬 In this episode of PaperCast Base by Base, we explore a cross-population genome-wide meta-analysis combined with proteomic profiling to enhance risk prediction and uncover molecular mechanisms underlying atrial fibrillation.
🔍 Study Highlights:
The authors conducted a meta-analysis of over 168,000 atrial fibrillation cases across diverse ancestries and identified 525 genome-wide significant loci. They integrated genomic data with large-scale proteomic profiling to discover 28 circulating proteins with potential causal roles in atrial fibrillation. Mendelian randomization analyses highlighted modifiable risk factors such as obesity and hypertension and revealed novel protein targets for therapeutic development. A combined polygenic risk score and protein score significantly improved prediction accuracy for atrial fibrillation beyond existing models. Population-specific analyses uncovered both shared and unique genetic correlations with circulatory comorbidities across European and African ancestries.
🧠 Conclusion:
This study demonstrates the power of integrating cross-population genomic and proteomic data to refine atrial fibrillation risk stratification and identify novel therapeutic avenues.
📖 Reference:
Yuan S, Chen J, Ruan X, Li Y, Abramowitz SA, Wang L, Jiang F, Xiong Y, Levin MG, Voight BF, Gill D, Burgess S, Åkesson A, Michaëlsson K, Li X, Damrauer SM & Larsson SC. Cross-population GWAS and proteomics improve risk prediction and reveal mechanisms in atrial fibrillation. Nat Commun. 2025;16:6426. doi:10.1038/s41467-025-61720-2
📜 License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/
On PaperCast Base by Base you’ll discover the latest in genomics, functional genomics, structural genomics, and proteomics.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:14):
Welcome to base by base, the AERcast that brings genomics to you
wherever you are. All right, let's jump straight
into a condition that silently impacts millions globally, often
without them even knowing it. Atrial fibrillation or AF?
Right. And it's not just a minor blip,
is it? No, definitely not.
It's actually the most common heart rhythm disorder out there.
(00:35):
We're talking about, you know, disorganized heartbeats.
Which can lead to some pretty scary symptoms.
Palpitation, certainly, but alsoreduced exercise capacity.
And even more seriously, complications like stroke.
It's a big deal. It really is, and the scale is
just staggering. The Global Burden of Disease
Study back in 2019, they estimated nearly 60 million
(00:56):
people worldwide living with AF.60 million.
Yeah, and that translates to something like 8.4 million
disability adjusted life years every single year.
It's genuinely an increasing epidemic, especially in places
like North America and Europe. Those numbers really hit home,
but imagine if we could sort of flip the script on something so
widespread. How?
Do I mean well? What if we could use that
(01:18):
incredible, unique combination of your genetic blueprint, your
DNA, and the proteins actually circulating in your body to
predict your risk for AF before it even develops, And then maybe
even tailor prevention and treatment just for you?
So uncovering those hidden signals, those molecular clues,
(01:38):
moving from just reacting to thedisease to being proactive,
truly personalized. Exactly.
That's the promise here, to fundamentally revolutionize how
we understand and hopefully manage this condition.
A really exciting prospect. And today, we're diving into
some truly ground breaking work that brings us a lot closer to
that future. We want to celebrate the efforts
(01:59):
of a global research team led byresearchers from the University
of Pennsylvania and Karolinska Institute.
They've really pushed our understanding of atrial
fibrillation forward significantly so.
Let's set the stage a bit. Despite AF being so common,
what's the core problem researchers have been facing?
Well, it's kind of surprising, but we still don't fully grasp
the precise molecular mechanismsdriving it.
(02:21):
You know what's actually going wrong at the biological level.
And that means we don't know thebest therapeutic targets either.
Exactly. There's this urgent need to
uncover its biological basis so we can get much better at
predicting who's at risk, preventing it and, of course,
treating it effectively when it does happen.
We know lifestyle factors play arole, sure, but the genetic side
(02:41):
is becoming clearer, right? Absolutely.
The genetic contribution is increasingly recognized.
Previous studies, these genomideassociation studies or GS,
they've already identified over 100 genetic regions or loci
linked to AF risk. OK, so think of your DNA like a
huge instruction manual. We knew a few page numbers or
(03:02):
loci where typos increased risk.Right, A good start, but far
from the whole story. And here's the critical gap.
Most of those earlier studies focus overwhelmingly on people
of European ancestry. Which means we've been missing
potentially crucial insights from other populations around
the world. Precisely, we had significant
blind spots, blanks in our understanding of AF's global
(03:22):
genetic landscape. And that's where the study comes
in. It really tackles that gap head
on. They conducted this massive
multi population genetic analysis.
But they didn't just stop with the genes.
They cleverly combined that genetic data with proteomics,
the study of proteins. AH.
Looking at the proteins too, whyis that important?
It gives you a much clearer, more refined picture.
(03:42):
Genes provide the blueprint, butproteins are often the
functional molecules doing the work.
Looking at both gives you a richer understanding of risk
factors and potential drug targets.
It's a real deep dive into both the genetic and protein
landscape. OK, so methodology.
How did they actually approach this?
It sounds complex. It was incredibly comprehensive.
They started with a cross population Juwar's meta
(04:05):
analysis, basically pooling genetic data from many studies,
and then they followed up with extensive downstream analysis to
really dig into what the data meant.
And you said massive earlier. How big are we talking?
We are talking huge. The GU2S meta analysis included
data from over 2,000,000 individuals, 2 million yeah, and
within that 168,007 confirmed AFcases.
(04:28):
But crucially, they made a real point of including diverse
populations European, East Asian, African and add mixed
American ancestries. That diversity is key, isn't it,
Filling in those gaps we mentioned.
Absolutely. It's a huge step forward for
making genetic research more equitable and frankly, more
accurate globally. So they have all this genetic
data, these potential risk locations or loci, how do they
(04:51):
figure out which specific genes at those locations were likely
involved? Good question.
They used a very systematic framework, not just one method,
but they integrated 6 different computational approaches to find
the most confident links betweena gene, the specific
instruction, and the risk locus it sits.
In OK, so multiple lines of evidence pointing to the same
(05:13):
gene. Exactly.
And once they had this list of prioritized genes, they used
something called pathway enrichment analysis.
What does that tell you? It helps identify the biological
jobs or pathways these genes areinvolved in.
So it confirmed known links to things like muscle contraction
and heart development, which we expected for AF, right?
But, and this is really interesting, it also flags some
(05:33):
new, potentially surprising connections athways we hadn't
strongly associated with AF before.
Intriguing. We'll come back to those.
They also used a technique called Mendelian randomization,
or MMR. Sounds fancy.
It's powerful. Think of it like this.
The genetic variants you inheritare essentially random, like a
natural lottery at conception. Mr. uses these variants as
(05:57):
natural experiments. Natural experiments.
Yeah, it helps figure out if an exposure, say smoking or the
level of a certain protein, actually causes a disease like
AF, rather than just being correlated with it.
It helps get around some of the biases you see in standard
observational studies. So moving beyond correlation to
potential causation, that's crucial for finding things we
(06:20):
can actually change. Exactly, they used it here to
pinpoint modifiable risk factors, lifestyle things, and
also which circulating proteins might have a causal role in AF.
Got it. And lastly, they developed some
predictive scores. Right, 2 main ones, a polygenic
risk score, the PGS built from all the combined genetic data
and a protein score pro S, whichthey constructed using
(06:41):
sophisticated machine learning on individual level protein data
and. They tested these scores.
Critically, yes. They tested them in independent
patient data sets to make sure they actually worked in
predicting AF risk in different groups of people.
That validation is key for real world use.
OK, that's a massive amount of work.
Let's get to the payoff. What were the key findings?
(07:01):
What did they actually discover?Well, First off, the sheer
number of genetic loci. They identified a remarkable 525
genetic locations significantly associated with AF. 525 That's
way more than the 100 or so we knew before.
A huge increase. It gives us an incredibly
detailed map of of the genetic landscape.
Now most of these 483 were foundmainly in the European ancestry
(07:25):
group, which was the largest. But here's a really crucial
finding. 2 specific loci near genes called PIT X2 and ZF HX3
were strongly associated with AFacross all four diverse
populations they studied so. Those two seem universally
important regardless of your ancestry.
That's the take away. They represent a common genetic
thread for AF risk across humanity.
(07:47):
They also highlighted 6 specificgenetic variants that had
particularly strong effects, increasing risk by over 30%.
These are near genes like SRCS 3and PLD one.
Definitely areas needing more research.
Fascinating. So universal factors and some
really strong individual signals.
What about the mechanisms? Did the pathway analysis turn up
(08:07):
anything unexpected? It did.
Beyond confirming the known roles of muscle development and
heart contraction, it revealed some surprising insights.
For example, pathways related tothe cellular response to
transforming growth factor beta or TGF out.
TGF was that involved in? It's heavily involved in tissue
repair and crucially, fibrosis scarring.
(08:28):
This suggests AF might not just be an electrical rhythm problem,
but perhaps also involves structural changes like scarring
in the heart tissue itself. Wow, that could change how we
think about treating. It it really could, they also
found links to artery formation and how cells communicate
definitely opens up new avenues for understanding the biology.
And did they see differences between the populations?
Yes, that was another important finding.
(08:49):
They looked at how AF genetically correlates with
other circulatory conditions, things like abdominal aortic
aneurysm or deep vein thrombosis, and they found
significant differences in thesecorrelations between European
and African ancestry populations.
It suggests that the way AF links up with other health
problems might actually vary depending on ancestry.
(09:11):
So the comorbidities might have different genetic roots in
different groups. Potentially yes.
It really highlights why we needdiverse study populations and
why A1 size fits all approach might not work for managing
related conditions. Makes sense.
Now, what about those modifiablerisk factors identified using
Mendelian randomization? Did it confirm what we generally
(09:31):
suspect? It did, but with much stronger
causal evidence. Things like obesity measured by
BMI and waisted hip ratio, type 2 diabetes, high blood pressure,
even high levels of thyroid stimulating hormone, TSH, we're
all causally linked to higher AFrisk.
And lifestyle factors. Yep, smoking, alcohol
consumption, even things may be less obvious like high amounts
(09:54):
of leisure screen time and insomnia showed causal links to
increased AF risk. So this really reinforces the
importance of lifestyle changes for prevention, but now with
stronger genetic backing for causality.
Exactly. It provides clear, actionable
targets for prevention efforts that individuals can discuss
with their doctors. And Mr. also looked at proteins,
(10:15):
right? Anything exciting there for
potential treatments? Definitely.
They identified 28 circulating proteins with potential causal
roles in AF. And here's the exciting part for
drug development. Seven of these proteins already
have drugs that target them. Drugs already exist.
Yes, drugs either in clinical trials or even approved for
other conditions, just not specifically for AF yet.
(10:36):
This could potentially fast track the repurposing of these
drugs for AF treatment. That could save years of
development time. Huge potential.
Absolutely. But there was also a really
intriguing paradox they found with one specific protein, NT
Pro BNP. NT Pro BNP.
I know that one doctors often measure it in heart failure
patients and it's usually high in AF patients too, right?
(10:57):
Exactly. It's often elevated and we use
it as a marker. But here's the twist.
The Mr. analysis found that genetic liability to AF was
actually associated with lower predicted levels of NT PRO BNP
lower. That seems completely backward.
It does, but what it suggests isreally profound.
It suggests the high levels we see in AF patients might
actually be a consequence of thedisease, the heart being under
(11:20):
stress rather than a primary cause driving the AF itself.
Whoa. So the high levels are a result,
not a 'cause that's a major aha moment for how we interpret that
biomarker. A complete game changer for
diagnostics and maybe even monitoring.
It shows how genetics can help untangle cause and effect.
Absolutely incredible. OK, let's talk about risk
prediction. You mentioned the PGS and Pro S
(11:41):
scores. How much better are they?
Significantly better the new polygenic risk score, the PGS
derived from this massive, diverse data set showed superior
predictive performance compared to older scores based mostly on
European data. How much better could you
quantify that? Sure, they often use a metric
called AUC area under the curve.Higher is better, with one point
(12:03):
O being perfect prediction. This new PGS achieved an AUC of
.780 compared to around .767 forprevious scores.
It might sound small, but it's adefinite improvement in accuracy
and. What does that mean for an?
Individual, well, individuals inthe top 10% for this new PGS had
about a six fold increased risk of developing AF compared to
(12:25):
those in the bottom 10%. That's a substantial difference
in risk stratification. But here's the real kicker when
they combine the genetic score, PGS with the protein score.
Its approach exactly. Combining them significantly
boosted the predictive power. The combined score reached an
impressive AUC of .823. Wow, .823, that's getting pretty
(12:46):
good. It's a very strong indicator
that looking at both genes and proteins together gives you a
much clearer picture of someone's individual risk.
It's a major leap forward for prediction.
So stepping back, this study hasreally refined our whole picture
of AF genetics, hasn't it? Finding both those universal
risk factors and population specific ones.
It's like sharpening the focus dramatically, going from a
(13:07):
somewhat blurry understanding toa much higher resolution image
of the genetic architecture. And it really hammers home the
need for more diversity in genetic research, doesn't it?
Absolutely. It underscores that urgent need
to get the full picture to ensure discoveries benefit
everyone. We simply must increase
representation from diverse populations in future studies.
(13:29):
We're still missing pieces otherwise.
Definitely. And the new biological insights
like the TGF pathway link, that opens up completely new
therapeutic avenues, right? Exciting new avenues, yes.
Thinking about fibrosis, about cell communication.
It suggests potential new drug targets and anti arrhythmic
strategies we hadn't considered before, or as we discussed,
repurposing existing drugs targeting some of those causal
(13:52):
proteins. And on the prevention side, the
solid causal evidence for those modifiable risk factors obesity,
blood pressure, smoking, even sleep.
It gives really clear, actionable guidance.
It empowers individuals and doctors with stronger evidence
to target prevention efforts effectively.
Lifestyle changes, medical interventions.
(14:12):
We now have more causal backing for their importance in reducing
AF incidents. Which all leads towards that
goal of personalized medicine. The improved prediction using
combined genetic and protein scores seems like a huge step.
It really is. It opens the door for much more
tailored prevention and management.
Imagine a future where your doctor can calculate your
(14:34):
specific risk using your unique genetic and protein profile, and
then devise a strategy just for you.
That's the future we're heading towards.
Now, of course, no study is perfect.
Were there limitations acknowledged?
Yes, and it's important to mention them.
Despite their efforts, the statistical power for the non
European ancestry groups was still lower just because the
sample sizes were smaller. So some population specific
(14:56):
associations might still have been missed.
We need even larger diverse cohorts.
OK, so still more work to do there.
Definitely. Also, some of the gene
assignments relied on computational predictions.
While sophisticated, these ideally need follow up
experimental validation in the lab to confirm the biological
function. Right, the computational work
points the way, but lab work needs to confirm it.
(15:17):
Exactly, this study was primarily in silico using
computational methods. Functional studies are the
crucial next step to confirm thereal world biological relevance.
Understood. So wrapping up, what's the main
take home message from this hugeundertaking?
I think the central insight is that this landmark cross
population study, by integratingmassive genomic and proteomic
(15:40):
data, has not only significantlybroadened our understanding of
AF's underlying biology, but hasalso delivered clear, actionable
insights for prevention and potential new treatments.
And its key contribution. Identifying those critical
modifiable risk factors with causal evidence, finding novel
protein targets for potential drugs, and, perhaps most
excitingly for the future, dramatically enhancing
(16:02):
individual risk prediction by combining mining, genetic and
protein scores. It's a powerful stride towards
personalized medicine for AF. It really sets the stage.
So the big questions remain. What does this leap in
predictive power truly mean for developing personalized
prevention and treatment for atrial fibrillation down the
road? And perhaps broader still, how
might these multi emic approaches looking at genes,
(16:25):
proteins, maybe other Ohms together transform how we
understand and manage other complex diseases beyond AF?
Lots to think about. A truly fascinating deep dive.
This episode was based on an Open Access article under the
CCBY 4 Point O license. You can find a direct link to
the paper and the license in ourepisode description.
If you enjoyed this analysis, the best way to support Base by
(16:48):
Base is to subscribe or follow in your favorite podcast app and
leave us a five star rating. It only takes a few seconds but
makes a huge difference in helping others discover the
show. Thanks for listening and join us
next time as we explore more science base by Base.
Welcome to base by base, the AERcast that brings genomics to you
wherever you are. All right, let's jump straight
into a condition that silently impacts millions globally, often
without them even knowing it. Atrial fibrillation or AF?
Right. And it's not just a minor blip,
is it? No, definitely not.
It's actually the most common heart rhythm disorder out there.
(00:35):
We're talking about, you know, disorganized heartbeats.
Which can lead to some pretty scary symptoms.
Palpitation, certainly, but alsoreduced exercise capacity.
And even more seriously, complications like stroke.
It's a big deal. It really is, and the scale is
just staggering. The Global Burden of Disease
Study back in 2019, they estimated nearly 60 million
(00:56):
people worldwide living with AF.60 million.
Yeah, and that translates to something like 8.4 million
disability adjusted life years every single year.
It's genuinely an increasing epidemic, especially in places
like North America and Europe. Those numbers really hit home,
but imagine if we could sort of flip the script on something so
widespread. How?
Do I mean well? What if we could use that
(01:18):
incredible, unique combination of your genetic blueprint, your
DNA, and the proteins actually circulating in your body to
predict your risk for AF before it even develops, And then maybe
even tailor prevention and treatment just for you?
So uncovering those hidden signals, those molecular clues,
(01:38):
moving from just reacting to thedisease to being proactive,
truly personalized. Exactly.
That's the promise here, to fundamentally revolutionize how
we understand and hopefully manage this condition.
A really exciting prospect. And today, we're diving into
some truly ground breaking work that brings us a lot closer to
that future. We want to celebrate the efforts
(01:59):
of a global research team led byresearchers from the University
of Pennsylvania and Karolinska Institute.
They've really pushed our understanding of atrial
fibrillation forward significantly so.
Let's set the stage a bit. Despite AF being so common,
what's the core problem researchers have been facing?
Well, it's kind of surprising, but we still don't fully grasp
the precise molecular mechanismsdriving it.
(02:21):
You know what's actually going wrong at the biological level.
And that means we don't know thebest therapeutic targets either.
Exactly. There's this urgent need to
uncover its biological basis so we can get much better at
predicting who's at risk, preventing it and, of course,
treating it effectively when it does happen.
We know lifestyle factors play arole, sure, but the genetic side
(02:41):
is becoming clearer, right? Absolutely.
The genetic contribution is increasingly recognized.
Previous studies, these genomideassociation studies or GS,
they've already identified over 100 genetic regions or loci
linked to AF risk. OK, so think of your DNA like a
huge instruction manual. We knew a few page numbers or
(03:02):
loci where typos increased risk.Right, A good start, but far
from the whole story. And here's the critical gap.
Most of those earlier studies focus overwhelmingly on people
of European ancestry. Which means we've been missing
potentially crucial insights from other populations around
the world. Precisely, we had significant
blind spots, blanks in our understanding of AF's global
(03:22):
genetic landscape. And that's where the study comes
in. It really tackles that gap head
on. They conducted this massive
multi population genetic analysis.
But they didn't just stop with the genes.
They cleverly combined that genetic data with proteomics,
the study of proteins. AH.
Looking at the proteins too, whyis that important?
It gives you a much clearer, more refined picture.
(03:42):
Genes provide the blueprint, butproteins are often the
functional molecules doing the work.
Looking at both gives you a richer understanding of risk
factors and potential drug targets.
It's a real deep dive into both the genetic and protein
landscape. OK, so methodology.
How did they actually approach this?
It sounds complex. It was incredibly comprehensive.
They started with a cross population Juwar's meta
(04:05):
analysis, basically pooling genetic data from many studies,
and then they followed up with extensive downstream analysis to
really dig into what the data meant.
And you said massive earlier. How big are we talking?
We are talking huge. The GU2S meta analysis included
data from over 2,000,000 individuals, 2 million yeah, and
within that 168,007 confirmed AFcases.
(04:28):
But crucially, they made a real point of including diverse
populations European, East Asian, African and add mixed
American ancestries. That diversity is key, isn't it,
Filling in those gaps we mentioned.
Absolutely. It's a huge step forward for
making genetic research more equitable and frankly, more
accurate globally. So they have all this genetic
data, these potential risk locations or loci, how do they
(04:51):
figure out which specific genes at those locations were likely
involved? Good question.
They used a very systematic framework, not just one method,
but they integrated 6 different computational approaches to find
the most confident links betweena gene, the specific
instruction, and the risk locus it sits.
In OK, so multiple lines of evidence pointing to the same
(05:13):
gene. Exactly.
And once they had this list of prioritized genes, they used
something called pathway enrichment analysis.
What does that tell you? It helps identify the biological
jobs or pathways these genes areinvolved in.
So it confirmed known links to things like muscle contraction
and heart development, which we expected for AF, right?
But, and this is really interesting, it also flags some
(05:33):
new, potentially surprising connections athways we hadn't
strongly associated with AF before.
Intriguing. We'll come back to those.
They also used a technique called Mendelian randomization,
or MMR. Sounds fancy.
It's powerful. Think of it like this.
The genetic variants you inheritare essentially random, like a
natural lottery at conception. Mr. uses these variants as
(05:57):
natural experiments. Natural experiments.
Yeah, it helps figure out if an exposure, say smoking or the
level of a certain protein, actually causes a disease like
AF, rather than just being correlated with it.
It helps get around some of the biases you see in standard
observational studies. So moving beyond correlation to
potential causation, that's crucial for finding things we
(06:20):
can actually change. Exactly, they used it here to
pinpoint modifiable risk factors, lifestyle things, and
also which circulating proteins might have a causal role in AF.
Got it. And lastly, they developed some
predictive scores. Right, 2 main ones, a polygenic
risk score, the PGS built from all the combined genetic data
and a protein score pro S, whichthey constructed using
(06:41):
sophisticated machine learning on individual level protein data
and. They tested these scores.
Critically, yes. They tested them in independent
patient data sets to make sure they actually worked in
predicting AF risk in different groups of people.
That validation is key for real world use.
OK, that's a massive amount of work.
Let's get to the payoff. What were the key findings?
(07:01):
What did they actually discover?Well, First off, the sheer
number of genetic loci. They identified a remarkable 525
genetic locations significantly associated with AF. 525 That's
way more than the 100 or so we knew before.
A huge increase. It gives us an incredibly
detailed map of of the genetic landscape.
Now most of these 483 were foundmainly in the European ancestry
(07:25):
group, which was the largest. But here's a really crucial
finding. 2 specific loci near genes called PIT X2 and ZF HX3
were strongly associated with AFacross all four diverse
populations they studied so. Those two seem universally
important regardless of your ancestry.
That's the take away. They represent a common genetic
thread for AF risk across humanity.
(07:47):
They also highlighted 6 specificgenetic variants that had
particularly strong effects, increasing risk by over 30%.
These are near genes like SRCS 3and PLD one.
Definitely areas needing more research.
Fascinating. So universal factors and some
really strong individual signals.
What about the mechanisms? Did the pathway analysis turn up
(08:07):
anything unexpected? It did.
Beyond confirming the known roles of muscle development and
heart contraction, it revealed some surprising insights.
For example, pathways related tothe cellular response to
transforming growth factor beta or TGF out.
TGF was that involved in? It's heavily involved in tissue
repair and crucially, fibrosis scarring.
(08:28):
This suggests AF might not just be an electrical rhythm problem,
but perhaps also involves structural changes like scarring
in the heart tissue itself. Wow, that could change how we
think about treating. It it really could, they also
found links to artery formation and how cells communicate
definitely opens up new avenues for understanding the biology.
And did they see differences between the populations?
Yes, that was another important finding.
(08:49):
They looked at how AF genetically correlates with
other circulatory conditions, things like abdominal aortic
aneurysm or deep vein thrombosis, and they found
significant differences in thesecorrelations between European
and African ancestry populations.
It suggests that the way AF links up with other health
problems might actually vary depending on ancestry.
(09:11):
So the comorbidities might have different genetic roots in
different groups. Potentially yes.
It really highlights why we needdiverse study populations and
why A1 size fits all approach might not work for managing
related conditions. Makes sense.
Now, what about those modifiablerisk factors identified using
Mendelian randomization? Did it confirm what we generally
(09:31):
suspect? It did, but with much stronger
causal evidence. Things like obesity measured by
BMI and waisted hip ratio, type 2 diabetes, high blood pressure,
even high levels of thyroid stimulating hormone, TSH, we're
all causally linked to higher AFrisk.
And lifestyle factors. Yep, smoking, alcohol
consumption, even things may be less obvious like high amounts
(09:54):
of leisure screen time and insomnia showed causal links to
increased AF risk. So this really reinforces the
importance of lifestyle changes for prevention, but now with
stronger genetic backing for causality.
Exactly. It provides clear, actionable
targets for prevention efforts that individuals can discuss
with their doctors. And Mr. also looked at proteins,
(10:15):
right? Anything exciting there for
potential treatments? Definitely.
They identified 28 circulating proteins with potential causal
roles in AF. And here's the exciting part for
drug development. Seven of these proteins already
have drugs that target them. Drugs already exist.
Yes, drugs either in clinical trials or even approved for
other conditions, just not specifically for AF yet.
(10:36):
This could potentially fast track the repurposing of these
drugs for AF treatment. That could save years of
development time. Huge potential.
Absolutely. But there was also a really
intriguing paradox they found with one specific protein, NT
Pro BNP. NT Pro BNP.
I know that one doctors often measure it in heart failure
patients and it's usually high in AF patients too, right?
(10:57):
Exactly. It's often elevated and we use
it as a marker. But here's the twist.
The Mr. analysis found that genetic liability to AF was
actually associated with lower predicted levels of NT PRO BNP
lower. That seems completely backward.
It does, but what it suggests isreally profound.
It suggests the high levels we see in AF patients might
actually be a consequence of thedisease, the heart being under
(11:20):
stress rather than a primary cause driving the AF itself.
Whoa. So the high levels are a result,
not a 'cause that's a major aha moment for how we interpret that
biomarker. A complete game changer for
diagnostics and maybe even monitoring.
It shows how genetics can help untangle cause and effect.
Absolutely incredible. OK, let's talk about risk
prediction. You mentioned the PGS and Pro S
(11:41):
scores. How much better are they?
Significantly better the new polygenic risk score, the PGS
derived from this massive, diverse data set showed superior
predictive performance compared to older scores based mostly on
European data. How much better could you
quantify that? Sure, they often use a metric
called AUC area under the curve.Higher is better, with one point
(12:03):
O being perfect prediction. This new PGS achieved an AUC of
.780 compared to around .767 forprevious scores.
It might sound small, but it's adefinite improvement in accuracy
and. What does that mean for an?
Individual, well, individuals inthe top 10% for this new PGS had
about a six fold increased risk of developing AF compared to
(12:25):
those in the bottom 10%. That's a substantial difference
in risk stratification. But here's the real kicker when
they combine the genetic score, PGS with the protein score.
Its approach exactly. Combining them significantly
boosted the predictive power. The combined score reached an
impressive AUC of .823. Wow, .823, that's getting pretty
(12:46):
good. It's a very strong indicator
that looking at both genes and proteins together gives you a
much clearer picture of someone's individual risk.
It's a major leap forward for prediction.
So stepping back, this study hasreally refined our whole picture
of AF genetics, hasn't it? Finding both those universal
risk factors and population specific ones.
It's like sharpening the focus dramatically, going from a
(13:07):
somewhat blurry understanding toa much higher resolution image
of the genetic architecture. And it really hammers home the
need for more diversity in genetic research, doesn't it?
Absolutely. It underscores that urgent need
to get the full picture to ensure discoveries benefit
everyone. We simply must increase
representation from diverse populations in future studies.
(13:29):
We're still missing pieces otherwise.
Definitely. And the new biological insights
like the TGF pathway link, that opens up completely new
therapeutic avenues, right? Exciting new avenues, yes.
Thinking about fibrosis, about cell communication.
It suggests potential new drug targets and anti arrhythmic
strategies we hadn't considered before, or as we discussed,
repurposing existing drugs targeting some of those causal
(13:52):
proteins. And on the prevention side, the
solid causal evidence for those modifiable risk factors obesity,
blood pressure, smoking, even sleep.
It gives really clear, actionable guidance.
It empowers individuals and doctors with stronger evidence
to target prevention efforts effectively.
Lifestyle changes, medical interventions.
(14:12):
We now have more causal backing for their importance in reducing
AF incidents. Which all leads towards that
goal of personalized medicine. The improved prediction using
combined genetic and protein scores seems like a huge step.
It really is. It opens the door for much more
tailored prevention and management.
Imagine a future where your doctor can calculate your
(14:34):
specific risk using your unique genetic and protein profile, and
then devise a strategy just for you.
That's the future we're heading towards.
Now, of course, no study is perfect.
Were there limitations acknowledged?
Yes, and it's important to mention them.
Despite their efforts, the statistical power for the non
European ancestry groups was still lower just because the
sample sizes were smaller. So some population specific
(14:56):
associations might still have been missed.
We need even larger diverse cohorts.
OK, so still more work to do there.
Definitely. Also, some of the gene
assignments relied on computational predictions.
While sophisticated, these ideally need follow up
experimental validation in the lab to confirm the biological
function. Right, the computational work
points the way, but lab work needs to confirm it.
(15:17):
Exactly, this study was primarily in silico using
computational methods. Functional studies are the
crucial next step to confirm thereal world biological relevance.
Understood. So wrapping up, what's the main
take home message from this hugeundertaking?
I think the central insight is that this landmark cross
population study, by integratingmassive genomic and proteomic
(15:40):
data, has not only significantlybroadened our understanding of
AF's underlying biology, but hasalso delivered clear, actionable
insights for prevention and potential new treatments.
And its key contribution. Identifying those critical
modifiable risk factors with causal evidence, finding novel
protein targets for potential drugs, and, perhaps most
excitingly for the future, dramatically enhancing
(16:02):
individual risk prediction by combining mining, genetic and
protein scores. It's a powerful stride towards
personalized medicine for AF. It really sets the stage.
So the big questions remain. What does this leap in
predictive power truly mean for developing personalized
prevention and treatment for atrial fibrillation down the
road? And perhaps broader still, how
might these multi emic approaches looking at genes,
(16:25):
proteins, maybe other Ohms together transform how we
understand and manage other complex diseases beyond AF?
Lots to think about. A truly fascinating deep dive.
This episode was based on an Open Access article under the
CCBY 4 Point O license. You can find a direct link to
the paper and the license in ourepisode description.
If you enjoyed this analysis, the best way to support Base by
(16:48):
Base is to subscribe or follow in your favorite podcast app and
leave us a five star rating. It only takes a few seconds but
makes a huge difference in helping others discover the
show. Thanks for listening and join us
next time as we explore more science base by Base.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.