iiinova — where your digital self keeps evolving

Simon Kozlov @sim0nsays.bsky.social · 04/05/2026

Bro resurrected NASA simulator in a weekend with Claude Code and used data from the real space launches to find the parameters to run it, and turns out it still works really damn well.

020

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

We're still far from solving biology with AI, but the progress is astonishing: median number of designs to solve a gene was around 4. Huge thanks to @sokrypton.org @harriswang.bsky.social for the mentorship and support, and all collaborators! Liyuan Liu, Charlotte Rochereau, @gogurtecho.bsky.social

020

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

No single model was able to solve every gene, it took all of them + developing novel methods. And even then, there are genes no model was able to solve - we had to use mutagenesis on last 3 stragglers. No model thinks these solutions are good, but they work in vivo. 9/

120

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

Here pLM redesigns most of the helix and introduces other mutations. What makes it work? I wish I knew. Understanding the mechanics of this solution alone requires an experimental effort comparable to this whole project, if not more. Which brings me to Feynman's famous quote. 8/

110

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

But we still haven't solved all. For the rest, we expand the design space: allowing to change not just ILE residues, but allow compensating mutations elsewhere. We masked ILE neighbors in 3D space and repeated the process. Among several new working designs, one is crazy. 7/

110

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

This solves a bunch more and produces several viable solutions not explored by nature! 6/

100

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

Or we can find designs AlphaFold considers to produce the same structure. Or better - sequences both models consider a good match! In this work, we used a new method to optimize sequences by inverting AlphaFold with a bias towards ProteinMPNN logits x.com/sim0nsays/st... 5/

x.com

100

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

But say there's a super-conserved position where nature didn't explore anything else. How can we go beyond that? Well, maybe we can use structure as a prior. We run ProteinMPNN with a known structure for still-not-solved proteins, mask ILEs, see if it makes different designs. 4/

100

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

Next, you can feed sequences into pLM like ESM, mask ILEs and take max predicted AA other than ILE! pLMs store evolutionary information so we're modeling what else evolution could place there. We do it with different methods developed by Charlotte Rochereau, solve 25% more. 3/

100

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

So, you want to redesign the ribosome to not use a specific AA, and you chose isoleucine. How would you go about it? First, baseline: just try replacing it with most similar AA, evolutionary data says it's valine. Turns out this alone works in 36% of the cases 2/

120

Simon Kozlov @sim0nsays.bsky.social · 01/05/2026

Yours truly is a proper scientist now! TL;DR: we used AI to redesign parts of essential cell machinery with only 19 canonical amino acids instead of 20. Why? Great thread by @harriswang.bsky.social provides more context and details. Let me talk a bit about the AI design part of this. 1/

1184

Reposted by Simon Kozlov

Gaurav Bhardwaj @gauravbhardwaj.bsky.social · 23/05/2025

Great to have our manuscript with @sokrypton.org 's lab describing AfCyDesign finally out in @natcomms.bsky.social . Structure prediction, sequence redesign, de novo hallucination of cyclic peptides, and some binder design examples in this version. rdcu.be/em0vA

rdcu.be

Cyclic peptide structure prediction and design using AlphaFold2

Nature Communications - AfCycDesign: Cyclic offset to the relative positional encoding in AlphaFold2 enables accurate structure prediction, sequence redesign, and de novo hallucination of cyclic...

0295

Reposted by Simon Kozlov

Eric Columbus @ericcolumbus.bsky.social · 16/04/2025

Oh wow great @haleaziz.bsky.social piece on a 2018 case in which Trump’s ICE screwed up and erroneously deported a guy. AND THEY BROUGHT HIM BACK because why wouldn’t they? We’re in a much darker place now. Free link: www.nytimes.com/2025/04/15/u...

In one court document from Aug. 23, 2018, the government owned up to its mistake, saying Mr. Subaihani had been removed to Iraq despite the court's order.
Locating Mr. Subaihani was complicated because nobody knew where he had gone, or even if he had made it to Iraq. By September 2018, several weeks after Mr. Subaihani had been deported, a federal judge demanded ICE go to great lengths to find him.

ICE officials complied, explaining to the court that they were in touch with State Department staff in Iraq and in other countries he had transited through. They even called airline officials — Mr.
Subaihani had been deported on a commercial flight — for more on his whereabouts.
n late September, ICE officials notified the court that they had und evidence Mr. Subaihani had made it to Iraq. By October, Mr.
Subaihani's attorneys had tracked him down with the help of an investigator, and set about working with ICE to get him back to the United States.

11494194

Simon Kozlov @sim0nsays.bsky.social · 10/04/2025

Curios how you connected revoked visa to jaywalking - in the stories I heard there were no explanations whatsoever.

030

Reposted by Simon Kozlov

Yehlin Cho @yehlincho.bsky.social · 08/04/2025

Excited to share our preprint “BoltzDesign1: Inverting All-Atom Structure Prediction Model for Generalized Biomolecular Binder Design” — a collaboration with @martinpacesa.bsky.social, @Zhidian Zhang, @Bruno E. Correia, and @sokrypton.org 🧬 Code will be released in a couple weeks

56117

Simon Kozlov @sim0nsays.bsky.social · 08/12/2024

The best-run train line in US. Hit me up next time you’re in Boston!

000

Reposted by Simon Kozlov

Axel Visel @axelvisel.bsky.social · 04/12/2024

BOOSTER, a newly identified orphan gene, supercharges photosynthesis and, when overexpressed, boosts height of poplars by up to 200% (at least in greenhouse) To make it even more interesting, part of the gene seems derived from a poplar fungus-farming ant 🐜 Paper: www.cell.com/developmenta...

Figure from the paper, showing BOOSTER-overexpressing poplar plants that are about twice as tall as the negative controls. Googly eyes (not from original paper) look at the taller plants.

38630

Simon Kozlov @sim0nsays.bsky.social · 04/12/2024

Yes, as long as all the code is in the Jupyter Notebook :(

000

Reposted by Simon Kozlov

Harris Wang @harriswang.bsky.social · 30/11/2024

💡💥Here’s a modest proposal I’ve been pondering for a while on how to change the PEER REVIEW process and start more community-driven SCIENCE FUNDING. Lots of feedback/discussion welcomed! Other variants of the idea might exist already—not claiming originality. (a long 🧵) (1/9)

84414

Simon Kozlov @sim0nsays.bsky.social · 29/11/2024

What would be an economic model for this? Are there economic incentives for farmers to reduce nitrous oxide emissions already?

100

Simon Kozlov @sim0nsays.bsky.social · 29/11/2024

Fully accepting nothing may work out is a surprisingly effective way to work something out. Here’s the catch: knowing this fights against acceptance.

010

Simon Kozlov @sim0nsays.bsky.social · 27/11/2024

A venture creation fund, where all companies stay small lol

010

Simon Kozlov @sim0nsays.bsky.social · 25/11/2024

Holy shit, @akosnyerges.bsky.social got a meaningful reply to a highly technical question. We are so back! bsky.app/profile/geno...

040

Simon Kozlov @sim0nsays.bsky.social · 25/11/2024

I would love for a social network to exist where questions like this get a meaningful discussion. I don’t think Twitter had it even if it’s best days, with a few exceptions when tweet randomly goes viral. Can we make it work here? That would take the whole community to a next level.

140

Reposted by Simon Kozlov

Jason Kelly @jasonrkelly.bsky.social · 20/11/2024

Made a biotech founders starter pack, DM me to be added! go.bsky.app/LWm8uPE

2311639

Simon Kozlov @sim0nsays.bsky.social · 20/11/2024

Noticed that whenever I see a tweet I like, the first thing I do is check if author is on BlueSky. Glad to be here!

030

Simon Kozlov @sim0nsays.bsky.social · 20/11/2024

Also struggling with looking silly when exercising - sometimes when I walk back from work, I’m thinking I could jog for a few minutes, but feel I’ll look silly doing it in my regular clothes when people around me are jogging all geared up

010

Simon Kozlov @sim0nsays.bsky.social · 17/11/2024

Wanted to ask for a while - what is a good resource describing the main drivers of how we were able to reduce emissions and bend the curve so far? Like, who are the main players and what are the incentives that worked? (Fully agree we should bend it more)

000