Steve Crossan (@stevecrossan) profile picture

Solving Biochemistry @ Dayhoff Labs. Previously: DeepMind (AlphaFold), Google Maps, Gmail, Search.

Joined April 2007
Photo by Steve Crossan (@stevecrossan)Photo by Steve Crossan (@stevecrossan)Photo by Steve Crossan (@stevecrossan)Photo by Steve Crossan (@stevecrossan)Photo by Steve Crossan (@stevecrossan)Photo by Steve Crossan (@stevecrossan)
Steve Crossan retweeted
The free links to the article keep getting used up so here's yet another fresh one: as.ft.com/r/d769da51-5cf3-43…
I've been in France since 2002 and have never seen the streets so violent nor the politics so extreme. From the sacking of the supermarket on my street to the prospect of Présidente Le Pen, here's me @FT including thoughts from 2 French Nobel laureates as.ft.com/r/addcad57-0f9d-4e…
11
20
8
92
29,009
Steve Crossan retweeted
Living systems routinely map environmental signals to responses through enzyme-catalyzed reaction networks, but what physical features actually determine their computational capacity? We developed a scalable inverse-design framework that turns chemical reaction networks into function approximators by tuning physical parameters such as transition-state energies and reaction driving forces. Here’s one example: fitting the von Neumann elephant with a chemical reaction network 🐘 We find that network size and local control over reaction driving strongly determine the expressivity of these networks. Paper: arxiv.org/abs/2609.17339 Code: github.com/dayhofflabs/chemi… with @marcotux and Jason Rocks
4
6
1
77
14,638
Understanding enzymes (proteins that act as biological catalysts) means going beyond structure to function: which enzymes enable my reaction ? Horizyn (PNAS March '26) screens and ranks millions of enzymes against your reaction of interest - for example a synthesis step in manufacturing a medicine. Now available for free via API.
Horizyn-1 is now available via public API. Give it a chemical reaction, get back a ranked list of candidate enzymes, screened against ~7M representative sequences. Free to try, no setup: 💻 github.com/dayhofflabs/horiz… 📄 Paper (@PNASNews): pnas.org/doi/abs/10.1073/pna…
2
3
478
Very much worth the time of anyone thinking about #techbio for range and insight
Weekend reading: A roadmap for TechBio disruption in drug discovery Inner thoughts on how the industry works, and what could work for start ups entering the space. A blogpost with @Michael_JRYoung.
1
3
1
10
2,380
Steve Crossan retweeted
Time for another round of binder design
8
34
1
330
15,266
Steve Crossan retweeted
Very excited to support @Caltech @OrphanANME @argonne as an industry partner! We @dayhofflabs will share our model, Horizyn-1, for enzyme-reaction annotation in microbial genomes. Read more about the model here: pnas.org/doi/10.1073/pnas.25…
Two new multi-institution collaborations led by Caltech scientists have been selected to receive first-round funding from the U.S. Department of Energy (DOE) as part of the Genesis Mission, a national effort to use AI to accelerate scientific discovery. caltech.edu/about/news/two-c…
1
6
746
Steve Crossan retweeted
The idea that Shabana Mahmood would be ‘good for bond markets’ reveals no understanding of bond markets. Again. Can we please have someone who knows what they’re doing? @CJFDillow @darioperkins @MehreenKhn @dsquareddigest @MkBlyth
5
8
1
30
3,789
Looking for a quantum chemist with experience in reaction mechanism and kinetic modeling for a well-funded project. Please RT. DM me if you’re interested in learning more!
1
3
1
2
2,990
Steve Crossan retweeted
A.I. is improving the accuracy of medical diagnoses. Today for inherited retinal diseases [IRD] (Eye2Gene vs retinal experts) nature.com/articles/s42256-0…
5
53
2
184
26,107
love this
Just published my live map of the london underground! (link below)
818
“Why did AlphaFold happen at DeepMind rather than the Broad Institute? We treated it as an engineering problem as much as a research one.” People tend to become too captivated by the thrill of new discoveries, but it’s engineering that will drive leaps in the physical world.
1
2
563
beware self-avowed world class AI researchers raising 100s of M on a deck for national AI champion startups
225
I sit in one of the dives On Fifty-second Street Uncertain and afraid As the clever hopes expire Of a low dishonest decade
1
2
255
Steve Crossan retweeted
Announcing Encode: AI for Science We're launching a fellowship powered by @ARIA_research to connect top AI talent with leading UK science labs to unlock the next wave of scientific breakthroughs.
9
80
21
291
65,353
Steve Crossan retweeted
We’re delighted to co-host @jpmorgan's flagship #NYCW #ClimateWeekNYC event: “Funding the New Industrial Revolution,” featuring @DCVC portfolio cos @twelve_co2, @fervoenergy, + @oklo!
4
1
6
1,213
It's true. A related phenomenon is senior folks in tech believing 'we are well-meaning people so what we do must be good'. ‘Enshittification’ is coming for absolutely everything ft.com/content/6fb1602d-a08b… via @ft
2
1
336