Sanjukta Bhattacharya
Founder of Atlas Discovery, Predicting human response to drugs in clinical trials
Company
Sanjukta Bhattacharya is listed as a founder of Atlas Discovery. Predicting human response to drugs in clinical trials
Founder traction evidence
Public posts and activity attributed directly to this founder.
- X
We’re building foundation models that predict patient response in clinical trials. Excited for the future at @atlasdiscovery0 with @ShaamilKarim1 and @Cgensbigler ! Quote Atlas Discovery @atlasdiscovery0 · Jun 23 Nine out of ten drugs fail in clinical trials. The reason isn't a shortage of data. It's that none of it connects. We started Atlas Discovery to fix that with foundation models of patient drug response. We're backed by @ycombinator, @pearvc, and more.
We’re building foundation models that predict patient response in clinical trials. Excited for the future at @atlasdiscovery0 with @ShaamilKarim1 and @Cgensbigler ! Quote Atlas Discovery @atlasdiscovery0 · Jun 23 Nine out of ten drugs fail in clinical trials. The reason isn't a...
- X
We were curious whether the autoregressive next token prediction recipe that has worked for language can be applied to single-cell biolog...
We were curious whether the autoregressive next token prediction recipe that has worked for language can be applied to single-cell biology. We first tried to understand if a discrete representation is the right substrate for modelling in this domain? Tldr: Yes! We studied a...
- X
I spend most of my days working with millions of samples across tissues, diseases, and perturbations. Here’s how it goes: months go into defining an objective - it’s pretty much defined over the latent space a model has learned - so you’re just trying to get the model to Show more Quote Shaamil @ShaamilKarim1 · Jul 1 We trained a foundation model of patient biology that predicts drug response from a single biopsy before treatment. As a case study, we tested on a phase 3 trial of ustekinumab in IBD and get a 0.76 AUROC, enough to run it with ~450 fewer patients at the same statistical power.
I spend most of my days working with millions of samples across tissues, diseases, and perturbations. Here’s how it goes: months go into defining an objective - it’s pretty much defined over the latent space a model has learned - so you’re just trying to get the model to Show...
- X
New ways of quantifying cellular states are coming up everyday, with different granularities & modalities of information. AI cell models that we know of today use RNA-seq as the primary modality of learning - simply due to the ease and abundance of capturing this. I've been Show more
New ways of quantifying cellular states are coming up everyday, with different granularities & modalities of information. AI cell models that we know of today use RNA-seq as the primary modality of learning - simply due to the ease and abundance of capturing this. I've been...
- X
Super exciting! We embedded baseline colon biopsies from the UNIFI phase 3 trial using a foundational model of patient biology, trained a small supervised head on top, and predicted week-8 clinical response at AUROC 0.76 from a 358-patient cohort. Blog in the thread. Quote Christian Gensbigler @Cgensbigler · Jul 1 Excited to share a new blog for @atlasdiscovery0 outlining foundational models of patient biology and the role they can play in clinical trials by predicting patient response.
Super exciting! We embedded baseline colon biopsies from the UNIFI phase 3 trial using a foundational model of patient biology, trained a small supervised head on top, and predicted week-8 clinical response at AUROC 0.76 from a 358-patient cohort. Blog in the thread. Quote...
- X
Now for part (b), Papers disagree on which model, and which decoder head, is best, largely because they report different metrics. We ran ...
Now for part (b), Papers disagree on which model, and which decoder head, is best, largely because they report different metrics. We ran a controlled head ablation on one frozen backbone to find out why. Every standard metric encodes a hidden position on ONE axis: how rich the...
- X
Love this release. Their open weight model is apt for the overall tinker ecosystem AND it’s comparable to the frontier in model capabilit...
Love this release. Their open weight model is apt for the overall tinker ecosystem AND it’s comparable to the frontier in model capabilities!
- X
Also checkout our more recent progress at @atlasdiscovery0 !
Also checkout our more recent progress at @atlasdiscovery0 !