Ayush Noori
@ayushnoori
Rhodes Scholar researching AI for neuroscience at @HarvardDBMI, @WyssInstitute, @MGHNeurology to advance human health. Tweets/RT = my own.
Thrilled to share "Empowering biomedical discovery with AI agents" in @CellCellPress, from our team (led by @GaoShanghua) in @marinkazitnik's lab. We envision agentic "AI scientists" that coordinate: - LLMs 🤖 - ML tools 🛠️ - experimental platforms 🧪 - human input 🥼 1/9
Excited to share our perspective in @CellCellPress, where we discuss “AI scientists” as collaborative AI agents designed to empower biomedical research cell.com/cell/fulltext/… While the concept of an “AI scientist” is aspirational, advances in agent-based AI are paving the way…
This is hugely welcome from @kulesatony and team at @pillar_vc. If you're a U.S. PhD student or postdoc who has lost funding to attend ICML 2025, check out the grants below. 👇🏽 ⏰ Timeline: Deadline: June 20th Decisions: June 25th Attend ICML: July 13-19, Vancouver
University budgets everywhere are getting slashed, and we hear many PhD students with accepted ICML papers can no longer afford to attend. We are stepping in to offer travel grants for researchers who lost funding. Sponsored by Pillar VC, Ormoni Bio, Latent Labs, and Cimulate
What awful news… 💔 Condolences to his family, colleagues, mentees, and friends in the community.
Atul Butte died yesterday. The world lost a giant. A big bear of a man. With a huge smile. With love for everyone. With energy that could power a room. I loved everything about Atul. I loved how he was always happy. I loved how excited he was about science and helping people.
Data is everything. We need more data to unlock the next scientific breakthroughs with AI. Here’s how we’re addressing this at NeurIPS. With a $10,000 prize pool to get it started. 🧵 @NeurIPSConf @AI_for_Science @RenPhil21 #NeurIPS2025
Beyond excited for @grx_xce, @jaydenpersonnat, and @KamrynOhly who have just launched @designarena_ai, the first crowdsourced benchmark for AI-generated design with 35K+ users in 3 weeks. It’s incredible to watch your college friends change the world! 🌎📈
Three weeks ago, we started building an AI game engine. But some models kept making things look... sloppy. So we turned finding the best one into a game. In three weeks, that game grew to 35K+ users across 135 countries. Introducing @designarena_ai, the fastest-growing…
Taking ATOMICA to the wetlab has been a lot of fun! Excited to announce in our updated preprint we validate predicted dark proteome binders to heme 🧪🧬 Thanks to Michael Desgagné, Bradley L. Pentelute, and Joseph Loscalzo!
Trained on 2M+ interaction complexes, ATOMICA learns atomic-to-molecular embeddings of interaction interfaces @AdaFang_ In the latest preprint: 🎯 Wetlab tests of AI predictions: Experimental validation confirms heme binding for 5 ATOMICA-predicted sites in uncharacterized…
1/4 🚀 Announcing the 2025 Protein Engineering Tournament. This year’s challenge: design PETase enzymes, which degrade the type of plastic in bottles. Can AI-guided protein design help solve the climate crisis? Let’s find out! ⬇️ #AIforBiology #ClimateTech #ProteinEngineering…
Congratulations to @biogerontology @InSilicoMeds on a successful Phase 2A trial of rentosertib! clinicaltrials.gov/study/NCT05938…
📢 AI-enabled drug discovery reaches clinical milestone rdcu.be/eugUu Few AI-designed drug candidates have gone beyond in silico benchmarks. Now, a study in @NatureMedicine @biogerontology reports a successful phase 2a trial of rentosertib, an AI-discovered drug and…
Grateful to @BiswasFamilyFdn for highlighting our biomedical AI research in @Forbes. Excited to see this work reach a broader audience!
The Biswas Family Foundation is backing AI research that bridges the gap between the lab and the clinic. From drug repurposing to personalized treatment, they’re focused on translational breakthroughs that improve how we diagnose and treat disease. Read more in the latest from…
Atul Butte in memoriam ucsf.edu/news/2025/06/4… @UCSF @StanfordMed @Harvard @BostonChildrens @BrownUniversity
Just staring at the genome is not enough. As for human experts, having the AI "know" about clustering of clinical signs and symptoms and molecular pathways can signficantly increase accuracy of diagnosis of rare and even first-of-its-kind diseases.
There are over 7000 rare diseases. Due to clinicians’ limited experience with such diseases and the heterogeneity of clinical presentations, 70% of individuals remain undiagnosed. Can deep learning help close the gap? nature.com/articles/s4174…
There are over 7000 rare diseases. Due to clinicians’ limited experience with such diseases and the heterogeneity of clinical presentations, 70% of individuals remain undiagnosed. Can deep learning help close the gap? nature.com/articles/s4174…
Grateful to have contributed to this work on AI for rare disease diagnosis. Learn more about SHEPHERD at nature.com/articles/s4174… in @npjDigitalMed or below! 👇🏽 Spearheaded by @Emily_Alsentzer and @_michellemli, with team @shilpanads, @UDNconnect, @zakkohane, @marinkazitnik.
The challenge: Many rare diseases affect <50 per 100,000 people. Clinicians see so few cases that ~70% remain undiagnosed, long, costly “diagnostic odysseys” with no answers 💡 Innovation: Few‑shot + knowledge graph: SHEPHERD marries simulated patient data (~40k synthetic…
Very grateful to be a part of this work, mentored by @sachinjshah. Here, we identify a subgroup of patients for which AF screening is effective, and show that effect-based screening approaches should be used with risk-based approaches when screening for atrial fibrillation.
Identifying a Heterogeneous Effect of Atrial Fibrillation Screening in Older Adults: A Secondary Analysis of the VITAL-AF Trial @steven_lubitz @patrick_ellinor @JeffAshburner @jaymiyer @sachinjshah #AHAJournals ahajrnls.org/45v3eUI
Learning from @JohnJumperSci about the importance of social proof for AlphaFold's adoption, as demonstrated by posts on this app.
Rhodes Scholar contingent at @ycombinator AI Startup School, thanks @garrytan for the shoutout!
📢 One patient, many contexts, yet our AIs are still refreshing outdated prompts We envision context-switching AI will adapt to infinitely many medical contexts, new medical specialties, healthcare roles, diseases, and populations @_michellemli arxiv.org/abs/2506.10157 ♾…
Excited to see this: the UK has so much potential and so much talent to be catalysed. Thank you @tylercowen @FoundersPledge @RenPhil21 🙏
Emergent Ventures UK Tranche marginalrevolution.com/marginalrevolu…
Heartbroken to share my mentor & friend Dr. Atul J. Butte @atulbutte has passed. He saw potential in me and turned it into opportunity-- for me, and so many others. We bonded over Star Trek and he was our Captain. Picard’s parting toast in his honor, final episode:…
Excited to share results of the SHEPHERD program "Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases...we demonstrate SHEPHERD's effectiveness ...diverse diagnostic tasks, performing causal gene discovery, retrieving "patients-like-me", and…
Excited to present “Thinking in Space: How MLLMs See, Remember, and Recall Spaces” at #CVPR2025 as an Oral paper alongside my amazing co-authors @shushengyang @jihanyang13 (supervised by @sainingxie)! We’ll be speaking at 10:45am on Saturday, June 14, in the Davidson Ballroom!