Sanketh Vedula (@sankethvedula) profile picture

Postdoc at Princeton & Broad Institute. Prev: Visiting Scientist at IST Austria, PhD from Technion, Israel.

New York
Joined April 2012
Photo by Sanketh Vedula (@sankethvedula)Photo by Sanketh Vedula (@sankethvedula)Photo by Sanketh Vedula (@sankethvedula)Photo by Sanketh Vedula (@sankethvedula)
Excited to share that experiment-guided AlphaFold is now published in @NatureBiotech!
Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles nature.com/articles/s41587-0…
1
11
37
7,645
Sanketh Vedula retweeted
Thrilled to have been selected as a 2026 Google PhD Fellow in Machine Learning Foundations! Grateful to @GoogleResearch for supporting my research in AI × Causal Inference and its applications in experimental sciences.
We're excited to announce the 2026 recipients of the Google PhD Fellowship! 148 new PhD students across 38 countries, with $13M in support from @Googleorg. Their work spans quantum theory, language model reasoning, medical imaging and more: goo.gle/phd-fellows-2026
11
10
86
2,701
Sanketh Vedula retweeted
BindCraft2 v1.0.4 has been released: with significantly improved VHH, scFv, ARP, and induced fit targeting. WE also improved speed and code stability. Try your difficult campaigns again. Great work by Erik Mäeots, Leonardo Tredese and Elizaveta Chernova.
5
64
1
250
14,585
Sanketh Vedula retweeted
The cool thing about the Fold'EM paper, which directly uses images of EM particles as inputs for diffusion guidance, is that in principle it can be ported over to a codesign model like BoltzGen for modeling display/immunization-derived antibodies that have never been sequenced
1
9
85
4,426
Anyone else seen this? CryoEM models without any volume reconstruction. Straight from the micrographs, inferred structures restrained by the actual particle data. So sequence + limited particles = structure?! Bypassing all that time making a map just to still model after
2
19
3
125
7,534
Sanketh Vedula retweeted
ʙɪɴᴅᴄʀᴀꜰᴛ2 is out, and we're not waiting for the paper. The full code drops today, free for academic and industry use. We're releasing it early so you can start designing right now, and bring its full power to the current Adaptyv competition. github.com/PacesaLab/BindCra…
32
332
24
1,493
165,411
Real-world sleep phenotyping in autism: compliance, domain shift, and scalable wearable biomarkers – Sanketh Vedula – Poster Session C: Friday, August 14, 1 – 2 p.m.
1
4
156
🧬 Check out the work that our @Schmidt_Center fellows are presenting at #MLHC2026 in Baltimore this week! @mlforhc @broad_institute @JoshuaPickard_ @sankethvedula mlhc.org/accepted-research-p…
2
4
9
571
Sanketh Vedula retweeted
Go check this awesome work!!
Excited to present Inference-time Optimization (IT-Opt) for experiment-grounded protein ensemble generation at @icmlconf this week! #ICML2026 Joint work with @sankethvedula @MPegoraro42 @AnarSnowball @MartinPacesa @Alex__Bronstein + other wonderful collaborators. (1/10)
2
10
2,762
We are presenting our ITOpt framework to resolve experiment-consistent protein ensembles at ICML this week! Details in the thread below ⬇️ @Schmidt_Center @ISTAustria
Excited to present Inference-time Optimization (IT-Opt) for experiment-grounded protein ensemble generation at @icmlconf this week! #ICML2026 Joint work with @sankethvedula @MPegoraro42 @AnarSnowball @MartinPacesa @Alex__Bronstein + other wonderful collaborators. (1/10)
1
3
10
1,282
🇰🇷 Check out the work that our @Schmidt_Center fellows are presenting at #ICML2026 next week (Monday, July 6 – Saturday, July 11) in Korea, poster numbers and links in comments. @broad_institute
7
4
13
850
Inference-time optimization for experiment-grounded protein ensemble generation – @sankethvedula – Poster #613: icml.cc/virtual/2026/poster/…
1
1
60
Excited to present sampleworks: a modular platform for generating biomolecular conformational ensembles by coupling AI structure predictors to experimental data. First pub from @diffUSEproject. thestacks.org/publications/s…
1
5
1
16
1,568
Sanketh Vedula retweeted
🎨Activation steering can reliably push a text-to-image generator toward a visual concept, but at a cost: each concept needs its own estimation. ⚡HyperTransport (HT) predicts the intervention directly, matching per-concept SOTA at 3–4 orders of magnitude less cost. [1/6]
3
18
54
4,898
Sanketh Vedula retweeted
Replying to @BrahmaFi
@BrahmaFi joins @Polymarket. We started building 4 years ago and never stopped through every cycle, every narrative.The one thing that stayed constant was shipping new ideas. To our day one users, protocols that worked with us and teams that backed us. Onto the next one. Arigato
1
1
14
312
Sanketh Vedula retweeted
Wow, I got the 2025 Google PhD Fellowship! 🎉 Thanks to my advisors and friends from @ISTAustria and MPI-IS for their support. Kudos to all awardees! And of course, thanks @Googleorg for supporting our research!
17
11
272
22,065
Experiment-guided diffusion within AlphaFold3: Inject differentiable likelihood gradients from experimental data into its reverse diffusion steps. This steers sampling toward conformations that match real measurements.
1
11
56
3,840