My quote of the day
Nobody ever thanks you for saving them from the disease they didn't know they were going to get.
William Foege
01.09.2025 11:26
👍 1144
🔁 209
💬 7
📌 9
My quote of the day
Nobody ever thanks you for saving them from the disease they didn't know they were going to get.
William Foege
Interesting piece by Novartis researchers on small molecule drug design placing more emphasis on a first principles theory based on kinetics, non-eqbm behavior rather than just interaction free energies pubs.acs.org/doi/abs/10.1...
A show of bipartisan support for NIH. Basically flat except for ARPA-H.
The President’s budget proposal basically ignored (as it should have been).
✴️ Request for proposals! ✴️
@navigation.org is funding focused gatherings to tackle key challenges in Open Science. We support meetings that define solvable problems and quickly move ideas toward implementation.
Deadline: June 15, 2025
os.nav.fund/meeting-fund
#OpenScience
ATTENTION
The Washington Post is collecting data about funding issues. Please consider contributing.
thewashingtonpost.formstack.com/forms/scient...
Please remember that emails from your work account are not confidential, private or secure in any way
ATTENTION
If you are someone who had an F31-Diversity (or similar) application submitted this cycle, please DM me here, contact me on signal (jeremymberg.78), or email me at jeremymberg@gmail.com.
I will keep all information confidential.
2/n
Best science thread on bluesky to date
Great to see journals keeping up with the times!
Up until today I had been really focused on the shutdown of the NIH from a scientist perspective and today as I sat as a patient at the cancer center receiving my infusion I thought about it from the patient perspective.
White papers from biotech that are light on method details are nothing new; neuralplexer2, AF-latest, etc. AFAIK this👇 is the first on biorxiv. Presents a generative antibody design model but gives no details on how it works. Not reproducible, not science. What gives? www.biorxiv.org/content/10.1...
Come be my boss!
We're looking for a new Program Head for the Molecular Medicine Program at the Hospital for Sick Children.
Toronto, Canada
(the deadline has been extended from Jan 31)
Please Re-post.
jobs.sciencecareers.org/job/667223/p...
There are two types of scientists. When I mention a new paper, the one type asks:
“What is it about?”
The other type asks:
“In which journal?”
#SciSky
He’s getting divorced isn’t he
My kids think the whole concept of physical media is bizarre.
When people refuse to collaborate
Also from the GitHub repo: antibody antigen docking performance comparison to Boltz-1 and AF-2.3
What a surprise (not!). Yet again ... poor evaluations of specialized medical LLMs result in overhyped claims relative to the base LLMs. #bioMLeval
At some point, the money dries up. Corporate sponsorship is inherently unstable.
Just announced: 2025 Computer-Aided Drug Design (CADD) GRC program
The topic: "Exploring the Synergy of Machine Learning and Physics-Based Computational Chemistry to Accelerate Drug Discovery"
It is shaping up to be a seminal conference. Hope to see you there!
www.grc.org/computer-aid...
People are often surprised when I tell them this: both Twitter and Facebook/insta/threads are now specifically engineered to downgrade, hide, suppress real journalism. If this platform is different, that's a huge change
Would love to know. I'm sure we'll find some wrong mechanisms in there.
Like it or not, AI is currently at the forefront of advances in technology, mathematics and statistics. Take for instance the most recent innovation, uh *checks notes* error bars
There's now a C&EN starter pack! Follow me and my colleagues @cenmag.bsky.social go.bsky.app/2ytnrK3
A first attempt to assemble GPCR aficionados here on 🦋 Let me know who I missed by replying below!
go.bsky.app/75ukb81
By popular demand @science.org is on Bluesky!! No posts until after Thanksgiving, but stand by! @karaestellepowers.bsky.social @kakape.bsky.social @jakeyeston.bsky.social
The Small Molecule Digital Chemistry group at Novo Nordisk is looking for a Senior Scientist in the computational chemistry team. Join a young, dynamic team of drug discovery scientists working with an in silico first mindset to advance pharmaceutical programs! *Closing date Nov 30th
#chemjob
Book outline
Over the past decade, embeddings — numerical representations of machine learning features used as input to deep learning models — have become a foundational data structure in industrial machine learning systems. TF-IDF, PCA, and one-hot encoding have always been key tools in machine learning systems as ways to compress and make sense of large amounts of textual data. However, traditional approaches were limited in the amount of context they could reason about with increasing amounts of data. As the volume, velocity, and variety of data captured by modern applications has exploded, creating approaches specifically tailored to scale has become increasingly important. Google’s Word2Vec paper made an important step in moving from simple statistical representations to semantic meaning of words. The subsequent rise of the Transformer architecture and transfer learning, as well as the latest surge in generative methods has enabled the growth of embeddings as a foundational machine learning data structure. This survey paper aims to provide a deep dive into what embeddings are, their history, and usage patterns in industry.
Cover image
Just realized BlueSky allows sharing valuable stuff cause it doesn't punish links. 🤩
Let's start with "What are embeddings" by @vickiboykis.com
The book is a great summary of embeddings, from history to modern approaches.
The best part: it's free.
Link: vickiboykis.com/what_are_emb...
www.msn.com/en-gb/money/...
A 20 million fund to support organizations around the world using AI to address increasingly complex problems of different disciplines of science. Such as rare and neglected disease research, experimental biology, materials science and sustainability. #ai #Science
Most #GPCR structures obtained by #cryoEM show the receptor in an *active* state, bound to the G protein.
A new protein engineering strategy out of the Kobilka Lab promises to make it easier to determine cryo-EM structures of the *inactive* state of Family A GPCRs.
www.pnas.org/doi/10.1073/...