DCC!

Very cool. For large document processing, does the library handle chunking or other content extraction? Or should users do that before the upload step?

Been excited about this one for a while! What would you do with a new alphabet and the wealth of protein sequence bioinformatics at your disposal? We're also around at #EMBOComp3D Heidelberg and MLSB Copenhagen this week to discuss

Out of curiosity, what is it running on today?

Legend says the ancient Babylonians once tried to sequence and annotate God's own genome, and for their ambition and hubris they were forever cursed to have different annotation formats and standards so they could never do genomics with ease again.

Folks. Check your feeds. It's done

OpenAI just updated ChatGPT to be able to use RDKit, a cheminformatics Python package.

OpenAI's president says this makes ChatGPT "useful for scientific work across health, biology, and chemistry," but it is hilariously still not good at chemistry (🧵)

#chemsky #AI ⚗️🧪🖥️

Imagine starting a car that hadn't run in 21 years, that's 15 billion miles away in interstellar space. That's what the NASA team just did with Voyager's thrusters. People are amazing. jpl.nasa.gov/news/nasas-v...

Introducing Strands Agents, an Open Source AI Agents SDK | Amazon Web Services Today I am happy to announce we are releasing Strands Agents. Strands Agents is an open source SDK that takes a model-driven approach to building and running AI agents in just a few lines of code. Str...

Excited! aws.amazon.com/blogs/openso...

A. maz. ing.

felixrieseberg.github.io/clippy/

How I Made Google’s “Web” View My Default Search Forget AI. Google just created a version of its search engine free of the extra junk it has added over the past decade-plus. You just need one URL parameter.

tedium.co/2024/05/17/g...

I love uv and this is a good intro: hclimente.github.io/blog/python-...

I like this perspective a lot, but I’ll add a caveat that “institutional research” is an imperfect system for information retrieval and there will always be examples of lost knowledge/candidates/research paths that AI/ML tools can help explore

“Vaccines cause adults” ❤️

Backpropagation 101 #machinelearning with #cats #caturday

AI 2027 A research-backed AI scenario forecast.

ai-2027.com

Emu War - Wikipedia

TIL en.wikipedia.org/wiki/Emu_War...

The quantum revolution, 100 years on. - Claremont Review of Books The quantum revolution, 100 years on.

claremontreviewofbooks.com/where-mind-m...

Learning the language of protein-protein interactions www.biorxiv.org/content/10.1... 🧬🖥️🧪 github.com/VarunUllanat...

🤨

Screen shot of the article

Legend

Composite image illustrating enzyme inhibition concepts with a black dog in different poses: Top-left: The dog wears a pink flying disc on its head, labeled “Non-competitive (allosteric) inhibition (E–Inc).” Center: The dog in a tub, tongue out, labeled “E,” connected to “E–S” (enzyme–substrate). Top-right: The dog holding a yellow tennis ball, labeled “Competitive inhibition (E–Ic).” Bottom-center: The dog chewing a blue ball, labeled “Substrate binding.” Arrows and labels show the relationships between each form of the enzyme.

Composite image demonstrating “good” versus “bad” enzyme–substrate binding with the same black dog: Left: The dog with a blue ball in its mouth, labeled “E–Sᵍ” (“bad” binding mode). Center-left: The dog in a tub with tongue out, labeled “E.” Center-right: The dog holding a blue ball in a more controlled way, labeled “E–S” (“good” binding mode). Far right: The dog beside shredded toy pieces, labeled “E + P” (product formation). Text labels and arrows illustrate the transition from enzyme–substrate complexes to products.

It's important to communicate to the public the importance of basic science. Coco is happy to help explain drug binding mechanisms and enzyme kinetics basics:
#scicomm

Lab-in-the-loop therapeutic antibody design with deep learning Therapeutic antibody design is a complex multi-property optimization problem that traditionally relies on expensive search through sequence space. Here, we introduce "Lab-in-the-loop," a paradigm shift for antibody design that orchestrates generative machine learning models, multi-task property predictors, active learning ranking and selection, and in vitro experimentation in a semi-autonomous, iterative optimization loop. By automating the design of antibody variants, property prediction, ranking and selection of designs to assay in the lab, and ingestion of in vitro data, we enable a holistic, end-to-end approach to antibody optimization. We apply lab-in-the-loop to four clinically relevant antigen targets: EGFR, IL-6, HER2, and OSM. Over 1,800 unique antibody variants are designed and tested, derived from lead molecule candidates obtained via animal immunization and state-of-the-art immune repertoire mining techniques. Four lead candidate and four design crystal structures are solved to reveal mechanistic insights into the effects of mutations. We perform four rounds of iterative optimization and report 3-100x better binding variants for every target and ten candidate lead molecules, with the best binders in a therapeutically relevant 100 pM range. ### Competing Interest Statement All authors are or were employees of Genentech Inc. (a member of the Roche Group) or Roche, and may hold Roche stock or related interests.

www.biorxiv.org/content/10.1...

As others have noted, the life of Carl Bosch is worth a look in these times. Fritz Haber demonstrated that nitrogen could be reduced to ammonia, but Bosch’s work turned that into a technology that changed the world with sudden new supplies of fertilizer (and of explosives). (1/8)

Pharma CEOS Speaking Up, Damn It

Pharma CEOS Speaking Up, Damn It | Science | AAAS www.science.org/content/blog...

A message from Andy Jassy about our teammate, Sasha Troufanov Andy Jassy shared the following message with Amazon employees today.

Absolutely fantastic news
www.aboutamazon.com/news/company...

Every time I hear someone suggest, “track process X on a blockchain” my first thought is, “Why not git?