>be postdoc
>2012 in Cambridge
>join Warp Drive Bio
>plan to mine bacterial genomes
>find macrocycles
>what is a macrocycle
>RAS is undruggable
>everyone says so
>surface is a Death Star
>smooth no pockets
>picomolar grip on GTP
>whole field races the OFF state
>bossman crazy
>says hit the ON state
>the active one
>soil bug makes sanglifehrin
>CypA just sitting in the cell
>free chaperone?
>what if we glue it?
>macrocycle binds CypA
>new surface grabs RAS
>only the GTP bound state
>it clamps
>signal dies
>pure three way complex
>years of wet lab work
>first compound is garbage
>micromolar
>aim only at conserved residues
>swap phenyl for thiazole
>better CypA contact
>potency jumps
>no bioavailability
>unfixable half life
>fix it anyway
>chop the right side to a cyclopropyl
>stick a methylpiperazine on the left
>affinity explodes
>811 daltons
>big chungus
>breaks all rules
>should not be oral
>it is oral
>2018 buyout
>molecule becomes RMC-6236
>early data looks promising
>vp writes the blog post
>doesn't get it
>2023 need cash for phase 3
>EQRx dying with 1.1B in the bank
>buy the company for the cash
>sent CRO our structures to make
>they like
>they steal
>joyo to erasca
>where compound 4a
>comes back as ERAS-0015
>RASolute 302 reads out
>13.2 months
>double chemo SoC
>ASCO plenary
>1 min standing ovation
>presenter angry face
>senator on expanded access
>calls it a miracle drug
>press goes wild
>rash nbd
>dying is worse
>approval August 2026 easy
>198 million in five weeks
>mfw a macrocycle, from a dead startup, funded by a dead company, generational drug

the transcriptional state of a cell is not generally sufficient to predict its past or future. in our new preprint we show that conditioning on lineage context (sisters, cousins, etc) can help a lot.

this one was v fun :)

arxiv.org/abs/2610.40447

This is a Zebrafish

It has the same amount of Synapses as Qwen 3.8 27B but lacks the ability to one-shot Minecraft in Three.js

Disappointing

@proteinrosh- where are LLMs close to SOTA for small molecules in your experience?

There's this which I was very impressed by - rowansci.com/blog/binder-op…. Inspired by that post, I did a similar but simpler experiment using docking scores with similar results. And the Leash Bio example of using DEL screens for search was great. leashbio.substack.com/p/making-reali…

But there's also a lot of closely adjacent tasks where LLMs are still pretty bad. And even on these successful tasks, they're pretty slow.

Got fable to hand-write weights of a 100k param model to recognize real versus fake legal citations.

Basically one-shotted a full NN that’s better than training it

We ran a pipeline in a person team at Recursion in 2021 to start the oncology work. Pre-LLM we could execute quickly outsourcing at scale - the models will make this much more efficient. But the game changer was scaled biological insights from the platform, this is still the gap.

i've reproduced the results from the MIT blog post today on weight-space programming. but one additional result i have to add: this is a new, step-change emerging capability that's appeared in just the latest set of models.

Can we help discover new, unknown small molecules using modern generative AI and scientific agents?

Excited to share our work, led by superstar @zyxiong123, on ChefNMR2🧑‍🍳⚛️, a new approach for structure elucidation from NMR spectra!

Paper: chefnmr.cs.princeton.edu

🧵 1/

Carrion crows make decisions faster than monkeys and humans.

Learn more in a new #SciencePerspective: scim.ag/4zmOF1i

The most remarkable thing about Alnylam is how it turned breakthrough science into an engineered platform into a predictable machine for new medicines

This chart showing Alnylam’s “probability of success by phase” remains my favorite chart in biotech:

Epigenetic age is not a one-way street for the ovary. Sharif Iqbal and Vadim Gladyshev use pan-mammalian DNA methylation clocks to show ovarian epigenetic age runs in both directions: an old ovary placed in a young mouse becomes epigenetically younger. Conversely, a young ovary lder. biorxiv.org/content/10.648…

This is only for a single-celled bacterium…

The hype regarding VCMs is, frankly, quite disillusioning.

The complexity of the model (if we're to build one to 'simulate human biology’) in terms of the number of modules, parameters, data modalities, and scalable dataset size needed for it to be "useful" makes it presently intractable and impragmatic, especially as a commercial opportunity, which is what most of this hype is about.

Indeed, unmet patients exist, and we need more shots on goal to reach the market. But I'm highly skeptical virtual cell models will fix this. However, if we were to try, which cell would we even model?

--

The “sexy” disease areas to work on commercially and academically aren’t cell-autonomous; their pathology isn't restricted to a single cell.

In type 2 diabetes, pathophysiology emerges from heterogeneous, interconnected dysfunction across multiple cell types, tissues, and organ systems.

Alzheimer’s has a similarly complex, multicellular pathology, and timescales to consider too: affected neurons originate during prenatal neurogenesis (beginning ~3-4 wks after conception and until ~24 gestational wks), become terminally differentiated and remain in the G0 phase of the cell cycle for decades, evolving molecular pathologies long before clinical symptoms emerge.

An effective model will need to incorporate all these variables and data into its parameters. But given the scope and capabilities of current architectures, we're just not there yet… (they seem a lot more applicable to narrower and more constricted use cases)

--

That all said, many institutions want to work with target biology that hasn’t been drugged before. Either because the target's existence was unknown or it’s generally considered “undruggable.” But how can a model extrapolate beyond training data for genuinely novel targets and mechanisms of action?

In essence, there's a long way to go. We're only at the very beginning of biology's golden age!

Great little review. The lipid membrane is probably the most underrated and poorly studied part of the cell.

When people make "synthetic" cells, they usually do so using a single type of lipid. And yet, for reasons we don't yet understand, cells have dozens of different lipids e of which have remarkable properties.

"It was recently shown," Sáenz writes, "that in deep-sea ctenophores, plasmalogens, a type of curvature-promoting lipid, keep membranes functional under extreme pressure."

Oddly enough, when the J. Craig Venter Institute cut down a Mycoplasma cell's genome size to its near-minimum, and then studied the resulting organism, they "found around 90 different lipids, a number comparable to that of the naturally occurring mycoplasma from which its synthetic genome was derived. A minimal genome, it turns out, does not produce a minimal lipidome." Cells seem to *want* to have a high variance of lipids in their membranes, for reasons that are not clear.

Furthermore, when the researchers found ways to finally cut down the number of lipids in the Mycoplasma membrane, the cells still found clever ways to get around their constraints. "The cell...resists lipidome minimization," in other words. In one case, the researchers gave the cells "a conventional phospholipid" in their growth medium, and the cells quickly "scavenge[d] the acyl chains through the activity of lipases and rebuil[t]
a diverse lipidome from the pieces."

They finally cut down the number of lipids in these cells to just two, but this caused the cells to grow slowly and become artificially enlarged. So one argument for why cells try to maintain such a panoply of lipids in their membranes might be because it "allows them to control otherwise coupled properties independently of one another." From the review:

"Chain length, unsaturation, head-group chemistry and molecular shape
provide chemical degrees of freedom for tuning membrane properties. Because changing any one feature (for example, a double bond or a longer or shorter chain) generally shifts several properties at once, a minimal or fixed composition offers only coupled control. But not all lipidomic degrees of freedom are equivalent. Some properties can be produced through many lipidomic solutions, whereas others require a specific molecular structure."

Interesting throughout.

More preprints! More public datasets!

@leashbio is about a dozen people and we dropped 3 papers in the last year. And the year's not over!

Barton Springs Pool, Austin, Texas, 1926

Introducing Mol-1: Open Frontier Biomolecular Intelligence 🚀🚀🚀
📊 75.0% on FoldBench-AB, near IsoDDE's reported 75.58%!!!
🧬 60k-token assemblies on one single H20, outperforming Anthropic 8/10.
🎯 VHHs binding hit rate as high as 63% with best SPR KD at 43 pM.

GitHub: github.com/molence/Mol-1
Try it: lab.molence.com

Today we can finally talk about @enveda's partnership with @AnthropicAI . “Partnership” is a word that covers a lot of sins, so here is what it actually bought us:

1. a Phase 2 program pulled forward by three weeks
2. an IND submission accelerated for another soon-to-be-announced clinical program
3. an Investigator’s Brochure, drafted by @claudeai and signed off by our team after one minor edit, now in front of the FDA

We chose to work with Anthropic on development, not discovery. Enveda already finds medicines faster than we can develop them, and I think the industry is heading the same way. The slowest part of development is paperwork, directly or indirectly: batch records, study reports, the Investigator’s Brochure, and the IND that packages it all for the FDA. Discovery gets the headlines. Document review is where the months go.

Test 1: manufacturing batch release for a Phase 2 program. Before a batch can go to a trial, someone checks every page of its manufacturing record by hand. Claude read 1,061 scanned, largely handwritten pages from the campaign. Four of our reviewers and outside consultants independently reviewed the same pages, and a Quality adjudicator scored every comment against one question: would an FDA auditor raise it? Claude matched all 11 actionable issues the experts found and added 7 they had missed. Review time fell from 26 hours (typically spread over several weeks) to under 3, an 89% reduction, for about $37 in model cost. On another program, it flagged a climbing impurity that routine review had cleared; it later went out of specification.

Test 2: the Investigator’s Brochure for an IND, drafted in four hours instead of weeks. For a new Enveda medicine about to enter human trials, Claude drafted the brochure every trial physician relies on, from a structured layer of facts pulled from our tox, bioanalytical, and PK reports. Our reviewers found one minor unit-conversion error. We fixed it and included that draft in the IND. From the same source layer, Claude drafted the core of the IND itself in four hours, work that takes the best teams a month or more.

None of this works unless the model stays factful, so we built a custom harness to force it. It reads every page as an image, keeps a code ledger of which pages it read, and ties every finding to its page and governing document. Across the evaluation, no finding cited anything that was not in the record.

The full method is public. Our technical report on the CMC Document Processor covers the architecture, GMP controls, study design, and results, and we open-sourced a companion implementation so you can build your own!

Thank you to the incredible @enveda and @AnthropicAI teams for making this possible.

By far the coolest part of the Texas Election Code is the chapter allowing astronauts, many of whom live around Houston, to vote from space. 🚀🌕👨‍🚀

“wet lab biologists are anti AI” I am like 100% scheduling my daily life around astra usage limits

Lots of talk on simulating biology over the weekend.

I want to bring up this paper from 2012 to give a sense of what that might actually look like.

Take a look at those different things we'd need to model. We have scalable measurements for maybe 5 of these.

Paper: pmc.ncbi.nlm.nih.gov/articles/PMC34…

Of course, these are cells by themselves. Not even gonna get into what this might take inside a tissue!