I am not active on Bluesky, but wanted to share the Tweetorial about our paper about autoantibodies in cancer immunotherapy that came out yesterday in Nature in case anyone wanted a speedrun.
Paper link: www.nature.com/articles/s41...
Thread: x.com/aaronmring/s...
You're welcome! Let me know if you need any help setting up a project.
Ready to start designing some proteins? The door’s open at ariax.bio. If you have question or feedback, please drop a reply or hop in our Discord (discord.com/invite/Q7SnV...).
Cost? A typical 100-binder miniprotein design run takes 24–72 GPU-hours. That ends up being ~$100–$500 depending on the size of your target and design settings. Compare that to a mouse hybridoma campaign at a CRO, which can cost >$10,000 and drag on for months.
Here is what it looks in action. From the landing page, you can start designing in under a minute. It took just 52 seconds for this job to create miniprotein antagonists against CD122 (IL2Rβ).
Ariax removes that friction. Open the site, fill in a short form, and click start. We spin up a GPU in seconds, stream results back to a private cloud folder, and shut everything down when the job’s done. You only pay for compute—about 40 % less than AWS.
Still, for many researchers, these incredible tools are out of reach. They use powerful GPUs that most of us don’t have and complex command-line gymnastics over SSH.
BindCraft deserves special credit. @martinpacesa.bsky.social in Bruno Correia's group stitched the whole pipeline together: hallucinate, refine, score. It keeps winning Adaptyv Bio's design challenges and has become the go-to in my own lab. That’s the engine inside Ariax.
Computational protein design capabilities have exploded over the past few years. Tools like RFdiffusion, ProteinMPNN, ColabDesign, and BindCraft have changed the game. Success rates for miniprotein binder campaigns now approach 100%.
You’ve heard of vibe coding—meet vibe protein design. Click, launch, and watch the binders roll in. That’s what we're unveiling with @ariaxbio.bsky.social today.
Happy to share our new ScienceTM paper from Yingjun Cui, Erol Fikrig and @aaronmring.bsky.social et al. We built a yeast display library of 3,000+ I. scapularis proteins to profile antigens recognized by tick-resistant hosts, and developed a new anti-tick vaccine. www.science.org/doi/10.1126/...
To get started we have a lot of new discoveries coming out of the lab!!
Welcome Scott! Good to see another TST colleague from @fredhutch.bsky.social here.
Please make one!
I'm with Jony on this... I miss early Twitter when the feed was 100% cool science. It was such a cheat code! I hope bluesky can recreate that vibe. OFC if we want to get depressed/pissed off Twitter still exists...
Yes please!!!
Deleting will make it harder for people migrating here to find you. Also loss of valuable content. Why delete? Makes no sense.
Thank you Mikki
