Chris Ing

Co-Founder, CSO at @proteinqure, @UofT @UWaterloo. Peptide Therapeutics, Structure-based Drug Design, Computational Biophysics, AI for Drug Discovery, Cats

[Chris Ing @jsci]

@amyxlu Users will continue to leave aesthetics as an afterthought and underspecify design in prompts, the bar is just raised for a bit. Default agent UI is already tired and will eventually mark dated apps like default blog themes unless agents spend more tokens on unique design systems

[Chris Ing @jsci]

Has there ever been a sexier centerfold (technically speaking)? Thx @Nature

[Chris Ing @jsci]

@DdelAlamo Overath et al. found interface_dg and dSASA are useful metrics in combination with DL metrics (Figure 3A)
https://t.co/wNp76Ul5DL

[Chris Ing @jsci]

@charlieharris01 But if I learn how hot dogs are made, I won't want to compete in my local hot dog eating contest...

[Chris Ing @jsci]

Y'all aren't ready for what I'm gonna do with Molecular Nodes using this power...

[Claude @claudeai]

Claude now connects to the tools creative professionals already use.

With the new Blender connector, you can debug a scene, build new tools, or batch-apply changes across every object, directly from Claude.

[Chris Ing @jsci]

@kosonocky Appreciate the work consolidating these studies. I didn't see any list of tested sequences in any supplemental material?

[Chris Ing @jsci]

Slopocalypse now. Caught ChatGPT providing citations to an entirely AI generated "journal" (no editorial board, no listed authors) https://t.co/bURWHjOsqM

[Chris Ing @jsci]

@btnaughton @mgdurrant I don't think many academics can review/maintain open source software, even when bugs are reported (~50 MRs open on Boltz right now, for example). Better to make AI-maintained forks.

[Chris Ing @jsci]

Protein folding via poetry

[Alex Albert @alexalbert__]

We released Claude Opus 4.6 just two months ago. Today we're sharing some info on our new model, Claude Mythos Preview.

[Chris Ing @jsci]

@DdelAlamo I ended up following one these citation chains and it lead me in a circle! The most sinister are supplemental information references...

[Chris Ing @jsci]

@HannesStaerk Good place to shamelessly plug my Kaleidoscopic Protein Viewer (https://t.co/65XZ4m3ikO). Best experienced with Chrome.

[Chris Ing @jsci]

@MartinPacesa @DdelAlamo Professional antibody designers will tell you can't have any fun when it comes to the surface; their genius insights are like: sorry sir you can't have positive, negative, hydrophobic, or any patches of any kind, or else you won't have a chance in hell of being clinical stage😮‍💨

[Chris Ing @jsci]

@DdelAlamo "How should we adjust the design of our drug candidate to account for octopus physiology?"

[Chris Ing @jsci]

@sid_srk I acquired my refined taste through the grind

[Chris Ing @jsci]

@DdelAlamo Not the first though, ~4000 citations on the AlphaFold-Multimer preprint, https://t.co/wYEKwdT19f

[Chris Ing @jsci]

@OpenAI @Ginkgo Meanwhile the rest of us,

[Chris Ing @jsci]

@LeoCK_Wan hard mode: de novo ribosome, orthogonal translation engine that natively reads quadruplet codons, incorporates wild non-canonical monomers, and macrocyclizes on exit to print drug-like peptides on demand. Speed/fidelity/yield optimization a bonus ;)

[Chris Ing @jsci]

Expired: "One-click binder design"
Tired: "One-sentence binder design"
Wired: "One-epic-saga binder design"

[Chris Ing @jsci]

I'm not seeing much work done on human dispatch APIs. My agent seems suspiciously excited to make a programmatic TaskRabbit it can use to vicariously "touch grass"...

[Chris Ing @jsci]

@Ronalfa @owl_posting Hmm, but if the barrier to create a whole LLM competitive intelligence platform is so low, how competitive is it?🤔