Thomas Lemberger

A new impact framework for investors in life sciences bridges financial returns and patient outcomes - Nature Biotechnology A new model for quantifying the impact of capital investments in life science interventions could help investors, policymakers and healthcare stakeholders to assess, monitor and optimize the societal ...

Two metrics designed for measuring the impact of investments in the specific context of life sciences (in brief: the metrics assess the potential increase in life years attributable to the investment & the improvement in quality of life) ➡️ www.nature.com/articles/s41...

YouTube video by Human Technopole Trust, Truth and AI in Scientific Publishing with Thomas Lemberger

🤖✍️ How is #AI transforming scientific publishing? From trust to truth, @tlemberger.bsky.social @embopress.org explains what’s changing and why it matters for all of us.

youtu.be/k-_MuFEwFx0 @salmaamin.bsky.social @mkiwanski.bsky.social

Correction: 40,000 spectators.

A message of hope and optimism in these crazy times: Gewandhaus Orchestra open air concert, in Leipzig, Germany, right now: 20,000 people peacefully listening to the music of Max Bruch. Beautiful event.

An important, moving, and courageous piece from @juliosaezrod.bsky.social Respect!

"Becoming a patient has given me a new perspective on biomedical research. Despite my background in data analytics, I have found that I do not necessarily have access to the data or information I need..."

Could these unlucky applications be transferred and "re-used" to other funding agencies? A @reviewcommons.org for funding?

#TCTeAC - the Woodstock of Biology2 & #nightscience - was such an amazing event!! We met great scientists who creatively presented their high quality work. We dreamt and discussed in a very broad, open-minded setting. Talking science outside of our nutshell can be soo valuable - let's do this again!

Let's not be #TCTeAC "The Conference To END All Conferences", but a new format of many more such events. Depating science outside of our typical nutshell (scientifically and mentally) has great potential - we could have come up with 50+ project proposals together. Let's bring fun back to science :)

Those who confessed received my absolution for only a modest APC (Absolution Participation Charge)

Was a cool experience! I was very touched by the conversations. Did not expect this…

@paveltomancak.bsky.social and @itaiyanai.bsky.social fighting /debating about the hourglass idea
#theconferencetoendallconferences #TCTeAC

#fightforsciemce at #TCTeAC with #nightscience :) cell & gene people gather together.

(no scientist got hurt)

Our publishing process needs to be revolutionised @odedrechavi.bsky.social #TCTeAC high-quality AI generated peer review

Introverts at the Woodstock of Biology
#theconferencetoendallconferences #TCTeAC

I'm late to review the "Illusion of Thinking" paper, so let me collect some of the best threads by and critical takes by @scaling01 in one place and sprinkle some of my own thoughts in as well.

The paper is rather critical of reasoning LLMs (LRMs):

x.com/MFarajtabar...

In the meantime, Claude 4 is out, and the progress continues, albeit clearly remaining subthreshold in many benchmarks. Probably a matter of time and/or dedicated (rogue) training efforts...

Claude 4 system card:
www-cdn.anthropic.com/6be99a52cb68...

Indeed, working on it as we speak!

Massive experimental quantification allows interpretable deep learning of protein aggregation xAI trained on >100,000 random peptides predicts and elucidates primary sequence determinants of aggregation.

We quantify the aggregation of >100,000 random protein sequences to train CANYA, a convolution-attention hybrid neural network to predict aggregation from sequence. With @bennibolo.bsky.social www.science.org/doi/10.1126/...

A kind of cat

A kind of cat https://www.wikiart.org/en/paul-klee/a-kind-of-cat-1937

Many of the moons orbiting Jupiter and Saturn appear to host subsurface oceans. Mysteries abound as to what fills these deep sea worlds. But the hints that we’ve already gathered make them tantalizing targets in the search for life beyond Earth.

🧵

Very sad news!

A light blue diagram depicting the sea, with the water's surface at the top, and just below it is an outline of the Mary Rose, with the depth (12m, 40ft or 6.7 fathoms). There is a line showing the seabed, which takes a rapid dive into the vertical...

On the anniversary of the sinking of the Titanic, we thought we'd answer a question that's often asked,

"If they raised the Mary Rose, why not raise the Titanic?"

Allow our scaled diagram to explain...

New Binet cover

Woohoo it’s arrived!!! If it’s even 25% as good as Civilizations it’ll be amazing…

We then used a physics-based model to predict DNA linkers between the nucleosomes obtained by cryo-ET. This allowed us to visualize kbp-long chromatin fragments from inside human cells.

New preprint on 3D heterochromatin architecture in human cells! Great collab with @sergiocruzleon.bsky.social & @johannesbetz.bsky.social from @hummerlab.bsky.social, @marinalusic.bsky.social & the Turoňová lab. Many thanks to my supervisor @becklab.bsky.social. bioRxiv: tinyurl.com/3a74uanv 🧵👇

The Courtyard of the Hospital in Arles

The Courtyard of the Hospital in Arles, 1889 #artbots #vangogh
https://botfrens.com/collections/46/contents/14663

OK, self-replicating AI agents are perhaps not yet quite ready. Models still struggle to debug, a new Microsoft study reports.

Claude 3.7 Sonnet dominates. Will there be a Claude 3.7 Opus?

Preprint: arxiv.org/abs/2503.21557

Blog: www.microsoft.com/en-us/resear...

Group Leader | Generative and Synthetic Genomics at Wellcome Sanger Institute in Hinxton (UK)
Closing date: Sunday 01-JUN-25
#sciencejobs
sanger.wd103.myworkdayjobs.com/en-US/Wellco...

Tissue-like multicellular development triggered by mechanical compression in archaea The advent of clonal multicellularity is a critical evolutionary milestone, seen often in eukaryotes, rarely in bacteria, and only once in archaea. We show that uniaxial compression induces clonal mul...

Pretty much everything is fucked, but here is some beautiful and exciting science from @archaeon-alex.bsky.social and colleagues.

www.science.org/doi/10.1126/...

The Tariff rate is calculated the same way as the Impact Factor