iiinova — where your digital self keeps evolving

(11/13) I guess what I'm trying to say is: Welcome to the structure-free future of drug discovery.

Mean Girls Meme: "Get in loser" "We're doing structure-fress drug discovery"

151

(10/13) My favorite story is about the recent STAT6 inhibitors disclosed by Pfizer. Ptarmigan-1 is not only able to enrich for these in a screening scenario, but also localizes them to a shallow groove on the SH2 domain.

Ptarmigan-1 finds and localizes STAT6 inhibitors.

(9/13) Now the targets we actually built this for: 👉 intrinsically disordered: Boltz-2 goes near random, Ptarmigan-1 still enriches for active ligands 👉 cryptic pockets on CryptoBench: we beat P2Rank despite this being zero-shot, never trained to find pockets.

(8/13) Localization is where it gets fun. We score every residue and get a binding site map from sequence alone: 👉 top residue within 5 Å of the ligand for 92% of drug-like PoseBusters complexes 👉 engaged cysteine in the top 1% for 8 of 9 COValid targets, held-out ones included

Ptarmigan-1 can localize compounds to where they bind.

(7/13) On LIT-PCBA (well-folded, orthosteric, the home turf of structure-based methods) we came in 2nd behind Boltz-2. Competitive, not dominant - but these are not the targets we're after. (We also can't assess data leakage for the other methods)

(6/13) A screen with Ptarmigan-1 is just a nearest-neighbor lookup instead of a simulation. This clocks in at 10 ms per compound - ~5,000x faster than when we ran Boltz-2 on the same hardware. All 20,431 human proteins against a 3.4B compound library: 20 H100-hours.

A figure showing how fast Ptarmigan-1 is.

(5/13) Protein-level labels from bioactivity can be used. Residue-level labels from chemoproteomics data can be used. We can even extract residue-level labels from the structures in the PDB. Key to it all: our massive swath of internal data on the disordered proteins of the regulome.

(4/13) The bit that makes it trainable: two contrastive losses. One pulls a compound toward the specific residues it engages. The other only needs a yes/no on whether it binds the protein at all. This unlocks a new world of training data.

(3/13) So we skipped the structure. Ptarmigan-1 embeds every residue of a protein and every compound into one shared latent space, using fine-tuned foundation models for sequence and chemistry. The residue-level resolution is the real innovation here. Engagement is just cosine similarity. 🤖

A schematic of how Ptarmigan-1 co-embeds proteins and compounds.

(2/13) Co-folding tools like AlphaFold and Boltz have been revolutionary for finding ligands against well-structured proteins. But ~40% of human proteins don't fall into that category, and structure-based training data is biased against them. That's a big chunk of the proteome. 🧬

Michael Lazear @michaellazear.bsky.social · 15/07/2026

Reposted by Will Fondrie

Do you want to work at the frontier of structural and chemical #proteomics? Interested in induced proximity and building a feedback loop for machine learning models to explore biology? I'm hiring! Apply below ⬇️ jobs.ashbyhq.com/proxima/6e39...

jobs.ashbyhq.com

Proteomics Scientist

1113

Will Fondrie @willfondrie.com · 02/07/2026

Celebrating 250 years of America the right way tonight… Watching National Treasure of course.

050

Will Fondrie @willfondrie.com · 03/06/2026

#ASMS2026 coffee pilgrimage day 4: Torque Coffee. I had a nice cortado with their lighter espresso, Yellow Diamond (yes, they have two options!) This one is a bit further out in North Park, but a nice trek for some good coffee. Thanks for the rec @michaellazear.bsky.social!

A nice cortado on a green plate with a bubbly water beside.

040

Will Fondrie @willfondrie.com · 02/06/2026

Also one of the best pour over coffee that I’ve had!

000

Will Fondrie @willfondrie.com · 02/06/2026

Also, get a cardamom bun. Trust me - you’ll thank me later.

110

Will Fondrie @willfondrie.com · 02/06/2026

#ASMS2026 coffee pilgrimage day 3: Goldchild coffee. It's a bit of a hike downtown, but 100% worth it. This one is the best I've tried so far. www.goldchildcoffee.com

110

Will Fondrie @willfondrie.com · 01/06/2026

Come out this morning and see my AI talk in the chemoproteomics section at #ASMS2026! If nothing else, you’ll learn what a ptarmigan is 😉

160

Reposted by Will Fondrie

Will Fondrie @willfondrie.com · 01/06/2026

#ASMS2026 coffee pilgrimage day 2: Achilles Coffe Roasters. Coffee was pretty good - they have a nice single origin selection. Bonus is that it’s close to the convention center! achillescoffeeroasters.com

achillescoffeeroasters.com

Buy Single Origin Coffee | Achilles Coffee Roasters San Diego

Buy Single Origin Coffee Online at Achilles Coffee Roasters. A Specialty Coffee Roaster Located in San Diego With 5 San Diego County Cafes Serving Great Coffee and Killer Breakfast.

021

Reposted by Will Fondrie

Lindsay K Pino @lindsaykpino.com · 31/05/2026

Hello, #ASMS2026!! The @talusbio.bsky.social team is on-site and ready to nerd out! We've got everything from drug discovery to metadata records, from chemoproteomics to file format converters. 1/n

1134

Will Fondrie @willfondrie.com · 30/05/2026

You bet I did and it was delicious!

000

Will Fondrie @willfondrie.com · 30/05/2026

Looking for a burrito at #ASMS2026? Definitely hit up The Taco Stand: maps.app.goo.gl/fSYTwvyUCs57...

maps.app.goo.gl

242

Reposted by Will Fondrie

Lindsay K Pino @lindsaykpino.com · 28/05/2026

Chemoproteomics is back at #ASMS2026 with a bang! After the oral session mysteriously vanished last year, now we're not only back on the schedule but kicking off the conference with Ben Cravatt?! 1/n

2123

Reposted by Will Fondrie

Lindsay K Pino @lindsaykpino.com · 31/03/2026

Getting the ol' grad school gang back together with Carafe2 -- Bo Wen (@maccoss.bsky.social and Bill Noble) and @jspaezp.bsky.social (@willfondrie.com at @talusbio.bsky.social) teamed up for essentially some late night hackathons to build out Carafe for @brukercorporation.bsky.social timsTOF data.

2166

Will Fondrie @willfondrie.com · 23/03/2026

I 100% agree with this take. Plug: If you are working with #proteomics and/or #massspec data, check out our shortcourse at #ASMS2026 to get a jump start on many of these methods! (Course 04 here: www.asms.org/conferences/...)

asms.org

Short Courses

0142

Will Fondrie @willfondrie.com · 23/02/2026

Was it the jacket that made it happen?

020

Will Fondrie @willfondrie.com · 22/02/2026

Saw a post about GPU depreciation and I could hardly believe what second-hand H100s were selling for. The GPU market is wild.

030

Will Fondrie @willfondrie.com · 20/02/2026

Landed in St Louis for #USHUPO2026 and grabbed some BBQ 😋 2 things: 1. Salt + Smoke (www.saltandsmokebbq.com) is delicious. ⭐️ ⭐️ ⭐️ ⭐️ ⭐️ 2. I’m once again reminded that Seattle doesn’t know what to do with a smoker 😔

saltandsmokebbq.com

Best St. Louis BBQ - Salt + Smoke BBQ

St. Louis BBQ that will make your mouth water. We’re known for the good stuff. The meat stuff. The melt-in-your-mouth and fall-off-the-bone stuff. We’re Salt + Smoke and we know how to BBQ real good f...

020

Reposted by Will Fondrie

Chris Ashwood @cashwood.proteaglyco.com · 18/02/2026

There's an art to representing MS-based bioinformatics, and these authors have done an amazing job.

0103

Will Fondrie @willfondrie.com · 19/02/2026

This is a work I’ve been following. Some cool ideas here - definitely check it out!

020

Reposted by Will Fondrie

Ben Neely @benneely.com · 30/01/2026

Yep, Ben @proteomicsnews.bsky.social and I finally got Bill Noble on #THEProteomicsShow. We had been wanting him on forever so he went and won an award to make it easier for us to meet. Enjoy the rambles (I hope), and mind the monkeys. Find it wherever you find fine podcasts.

174

Will Fondrie @willfondrie.com · 20/01/2026

I love that @obsidian.md requires you to type the key sequence to exit vim before enabling vim key bindings 😆