Lightnews — Scholar-powered news

Reposted by Dennis Bontempi, Ph.D.

fbm-unil.bsky.social @fbm-unil.bsky.social · Aug 7

Dirigée par @zkutalik.bsky.social et @rjhfmstr.bsky.social à la @fbm-unil.bsky.social, au @sib.swiss et à #unisanté, une nouvelle étude publiée dans @nature.com présente une méthode computationnelle innovante pour étudier ces effets liés à l’origine parentale.
🔗 www.nature.com/articles/s41...

Parent-of-origin effects on complex traits in up to 236,781 individuals - Nature

A novel multistep strategy reveals how parent-of-origin effects shape complex traits in large-scale biobanks.

www.nature.com

1 4 8

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jun 3

I understand your point and these data are definitely not for everyone. But at the end of the day, if communicated clearly, this is not that dissimilar to The Framingham’s Score for cardiovascular diseases.

Nobody hates blood tests for that, and it actually turns out it saves tons of lives!

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jun 3

A very similar case: genetic testing for variants that increase the risk of certain breast cancers. It’s mostly about good communication of what (additive) “risk” actually means. Lots of people end up making lifestyle changes that improve their outcome (and risk profile, and life!) quite a lot!

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

Finally, thanks to everyone who stopped by my poster and all the friends and colleagues that contributed to making #ESHG25 a blast!

Can’t wait to do it all again next year in 🇸🇪🇪🇺

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

Such a different message from what we are used to hearing nowadays.

* I actually found out I had her book in my Goodreads to-read since last June but never got to it. So I got a copy on my way back and I am going to dive in it in the next few days 😎

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

“Don’t focus on what you cannot change; if they fire you, don’t waste time feeling sorry for yourself. Focus on what’s next, because that’s what you can change. And don’t hold a grudge against anybody for that. It poisons you, while the other person won’t even remember”.

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

Yet - she is one of the people behind the mRNA tech that helped bring the world out of the 2020 pandemic, and got a Nobel prize for it. And, as if that wasn’t enough, she’s also the mother of a two-time Olympic gold medalist!

Some lives are pure inspiration.

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

Being new to the field, I didn’t know her (though probably I should have)*.

She worked behind the iron curtain for years, was let go from multiple positions and forced to move countries several times. She basically never won a grant.

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 27

This year’s #ESHG 🧬 in Milan 🇮🇹 couldn’t have wrapped up better.

After days of delicious food and even tastier science, I got to be part of something special: a talk by Katalin Karikó that ended with a room full of scientists standing up and clapping to her incredible career.

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 26

And don’t get me wrong. Interesting study.

But sometimes I wonder if the problem with someone consuming 30% of their kcal in proteins (mostly from animal sources) while exercising 45 minutes/week at 30 BMI is really the g of proteins (and not the source, or being sedentary, or their lipids/BMI) 😅

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 26

Inclusion criteria included BMI ≥ 25.0 and ≤35.0 and less than 90 min of exercise per week (21-70yo).

I am really not sure how these findings generalise to a… healthier population. The numbers above are definitely not those of anyone who would normally be considered “fit” or close to it.

1 3

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 26

Particularly interesting talk from @rjhfmstr.bsky.social @ #ESHG25. Impressive work coming out of @unil.bsky.social 🧬🚀

1 5

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 9

... This has been a long time coming - super happy to finally have it published 🙃

📢 If you are interested in helping with the great research my colleagues are working on, expanding on this, you are welcome to read more here! aim.hms.harvard.edu/tryfaceage

#BiologicalAge #DeepLearning #AI #cancer

Try FaceAge — AIM - Harvard

aim.hms.harvard.edu

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 9

🏥 This moves us closer to bring an objective tool to prognostic assessments in oncology, which are often shaped by subjective judgment. Our results show that FaceAge improves validated models for clinical decision making, and can meaningfully support treatment planning.

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 9

🤳 To try and tackle this, we developed FaceAge, a deep learning system that estimates biological age from a simple photo of your face. In our study, we show FaceAge is a powerful predictor of health outcomes for cancer patients.

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 9

🎂 Biological age isn’t just the number of birthdays you’ve had - it’s a reflection of your health and aging process. Several promising biological aging clocks were developed in the last few decades, but most of them require invasive and expensive procedures 🩸🧬

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · May 9

✨ What if your face could tell something about how old your body really is?

Excited to share our latest paper just published in The Lancet Digital Health (open access!)

👉 www.thelancet.com/journals/lan...

FaceAge, a deep learning system to estimate biological age from face photographs to improve prognostication: a model development and validation study

Our results suggest that a deep learning model can estimate biological age from face photographs and thereby enhance survival prediction in patients with cancer. Further research, including validation...

www.thelancet.com

2 1 3

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 24

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

Why should we care? 🧠 ✨

Our research suggests that there is still much to learn about #cardiovascularhealth and that looking at the combination of both cLD and cIMT may be better than either one alone (... more on this coming soon 👀 )

#science #research #genetics #ultrasound #AI #UKBiobank

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

👉 We also investigated the ratio of cIMT to cLD, looking for a new perspective on the relationship between arterial wall thickening and lumen diameter that might lead us to a deeper understanding of arterial health!

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

👉 We observed that, although cLD and cIMT are only moderately correlated phenotypically, they have a much higher genetic correlation. This suggests they are influenced by some common genetic factors, but are also shaped by distinct processes!

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

👉 We identified several genes and pathways associated with cLD, and found that cLD has a substantially higher heritability than cIMT!

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

Here's a quick rundown of our study:

👉 We developed a phenotyping pipeline to measure cLD from ultrasound images using deep learning and classical computer vision. This allowed us to automate the measurement on a dataset from the UK Biobank of roughly 20k individuals!

1 1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

In response to plaque build-up or increased arterial stiffness, the inside of the common carotid artery — the Carotid Lumen Diameter (cLD) — can expand in an attempt to preserve the regular blood flow.

Contrary to the cIMT, the genetics of cLD haven’t been explored on a large scale... until now 😎

1

Dennis Bontempi, Ph.D. @denbonte.bsky.social · Jan 13

Classic risk factors for cardiovascular health — like high blood pressure, obesity, and smoking — have been linked to the thickening of the wall of the common carotid artery.

This phenotype — the Carotid Intima-Media Thickness (cIMT) — is a well-established marker of atherosclerotic burden.

1