Nick Boyd

Excellent, comprehensive rundown of the state of bio lab automation by @owlposting1.bsky.social

In retrospect, it's an important topic that has had almost zero discussion over the years!

A fun surprise to see some decade-old(!) work show up in there too.

www.owlposting.com/p/heuristics...

Naturally I had to add this to mosaic. Here's a VHH designed using `protenix_base_20250630_v1.0.0`. Example notebook here: github.com/escalante-bi...

GitHub - bytedance/Protenix: Toward High-Accuracy Open-Source Biomolecular Structure Prediction. Toward High-Accuracy Open-Source Biomolecular Structure Prediction. - bytedance/Protenix

Protenix v1.0 is out with some very impressive performance numbers (exceeding AF3 performance on protein-protein complexes)

Obviously these models aren't perfect and are trained on finite data, the data generating distribution doesn't really exist, there are better ways to control generative models, etc etc etc. This is still often a surprisingly illuminating way to think about these models.

Fun fact: non-generative models instead produce (in theory!) \argmin_x E [loss(x)], where the expectation is over p(x | c). This is why AF2 produces spaghetti and AF3 hallucinates helices.

Filtering criteria for PPIFlow

A more recent example is PPIFlow, which trains on complexes that are probably strong binders rather than all pairs of proteins in PDB (which might not bind tightly!). Or, the “ubiquitin memorization” issue with BoltzGen: if you see a protein of length 76 in PDB, it’s almost certainly ubiquitin.

There are many examples of this in bio + ML: soluble proteinMPNN shifts the distribution of proteinMPNN by training only on soluble structures.

e.g., AF3 roughly answers the question “if you saw a protein with this sequence and MSA in PDB, what kinds of structures would it fold into?”

A corollary is you can control your generative model’s output by filtering the training data.

If everything goes well, these models sample from p(x | c), where the joint distribution is the data generating distribution of the training set. i.e., they answer the question: “if you saw this conditioning information in your training set, what kinds of data would it be attached to?”

Developability comes for free...? Did AI de novo antibody generation models learn developability properties without explicitly being trained on this?

Love this theory for why antibody generative models produce sequences with nice developability properties: it’s a reflection of PDB. ayusuf.substack.com/p/developabi....

This is a very nice example of one way of thinking about conditional generative models.

Thanks Martin!!

Potential fork of PPIFlow with a more permissive license is very exciting: github.com/Mingchenchen...
Highly recommend their paper: results are super impressive and it's *not* an AF3 clone

drive.google.com/file/d/1KTPO...

I’ve got a physical copy 😅

JAX projects are more modular in my experience: it's sometimes really hard to get two torch projects to install in the same environment let alone interoperate nicely

probably I did too much functional programming + Julia in my formative years

speed/JIT/parallelization are really nice but it’s mostly a style thing for me. I find most large torch projects incomprehensible: lots of OO/imperative code/manual batching etc + frameworks like lightning/omegaconf. I can't go back to life before vmap & other higher-order functions.

inspired by @delalamo.xyz

vibe translated ligand/protein/soluble-mpnn from PyTorch to JAX. not sure if this works, but it was pretty fun and took 45 minutes of my time. Claude Code is going to make my virtuous no-torch lifestyle a lot easier... github.com/nboyd/jigand...

Building antibodies blindfolded: the paradox of de novo design By Natasha Murakowska and Joseph Harman

Loved this post from A-Alpha: aalphabio.substack.com/p/building-a.... If anything I think the IPSAE (or any other post-hoc metric) picture is even worse than they show: after optimization the fraction of false positives would (probably) be even higher than in this dataset

In the Nipah binder competition, which protein designer will have proteins that bind? Nipah is one of the deadliest viruses in the world and considered one of the top future pandemic risks. We're hosting a protein design competition on Proteinbase where people can design binders agai...

TL;DR: this was a really fun exercise but now is probably a good time to bet against me on manifold.markets/Proteinbase/...

~180 lines of code to win the in silico portion of the Adaptyv Nipah binding competition Here's the script we used to get 1st place in the in silico portion of the Adaptyv Nipah competition: import modal def download_boltz2(): from mosaic.models.boltz2 import Boltz2 Boltz2() ...

finally wrote up some notes on the adaptyv competition: blog.escalante.bio/180-lines-of....

To speculate wildly though: the Boltz2 confidence module seems really, really easy to please even compared to a single AF2-multimer model. I wonder if this means hallucination is more likely to produce interfaces Boltz2 likes but AF2-SC (and likely physics 😅) does not.

IMO it’s hard to draw conclusions from these data because each method has so many hyper-parameters. There isn't much work on AF3-gen hallucination; BindCraft is the result of some really careful and brilliant HPO. I was honestly surprised to get hits with Boltz2 for the work described in that post.

GitHub - bytedance/PXDesign: Official repository of PXDesign Official repository of PXDesign. Contribute to bytedance/PXDesign development by creating an account on GitHub.

PXdesign finally open-sourced with a permissive license. Hopefully it will get more attention now, imo one of the best current binder design methods github.com/bytedance/PX...

RamaX: Fast, Selective, Sensitive Screening Platform

Phage (or maybe cDNA or something more exotic?) display as a service is an interesting offering: www.diffuse.bio/ramax-fast-s.... Turnaround time is impressively short.

I really enjoyed this paper, thank you! Did you consider AF2 initial guess or AF2 rank for AF2-ss with the native or MSA-predicted structure as input? I wonder if the improved native structure performance would be worth the likely increase in false positives

oh, this is nice: x.com/proteinbase/.... only a few lines of code…

Submission 1 This is a submission to the Nipah Binder Competition competition.

Vote for @martinpacesa.bsky.social ‘s designs in the adaptyv nipah virus competition: proteinbase.com/collections/.... because they’ll actually bind

always amazes me that you can take code written on an A10, change a single line and get power utilization like this in JAX

not surprising this works, but it seems to improve Boltz2 IPTM for minibinders. another option is to simply use your favorite helix bundle or alpha solenoid as a scaffold or SS constraint