Riya Sinha

Excited to announce our latest work! We present ARSENAL, a short-context DNA language model specifically designed to learn important sequence features in noncoding regulatory DNA. Why is such a model important? Read on to find out! biorxiv.org/content/10.648…

@anshulkundaje @arjunmk1 @jengreitz @arjunmk1 yes! the api directly extends the Open Targets api as well. next to the enhancer gene prediction and LD sections, you can view and run the exact graphql queries used to render the pages

@arjunmk1 @jengreitz Cc @riyavsinha

Excited for a major milestone in a collab effort led by @jengreitz to map enhancers & interpret variants in the human genome: The E2G Portal e2g.stanford.edu collates predictions of enhancer-gene regulatory interactions across >1,600 cell types & tissues. Use cases 👇1/

@anshulkundaje @RichardKCollin2 @jengreitz Hi @RichardKCollin2, yes this is due to your browser being an older version that doesn't support this JS method. This is part of code from platform.opentargets.org that our portal extends. I'll add a clearer error soon, but you will need to upgrade to use the portal unfortunately!

@RichardKCollin2 @jengreitz We will see if we can fix this cc @riyavsinha

@anshulkundaje @arjunmk1 @igvteam Unfortunately not! the igv desktop application is a completely different codebase written in Java which I haven't had a chance to get familiar with yet. Would definitely like to support it in the future as well though

@arjunmk1 @riyavsinha @igvteam Not sure cc @riyavsinha

Thanks to @riyavsinha in my lab, @igvteam browser will natively support dynseq (dynamic sequence tracks) in an upcoming release. These tracks are very useful to directly visualize base-resolution scores (e.g. contribution scores from ML models, conservation etc). 1/

If you want to self-deploy #LangChain #LangGraph Runnables, check out langgraph_serve! It's an extension of #LangServe that supports modification of the LangGraph graph state, and resuming execution after graph interruptions. github.com/riyavsinha/lan…

Very excited to release tangermeme v0.2.0! This release focuses on attribution methods, seqlet calling, plotting, and variant effect scores, and tweaks the API. Docs: tangermeme.readthedocs.io/en/latest/ GitHub: github.com/jmschrei/tange… Get it with `pip install tangermeme`! 1/

You've probably seen attribution tracks where the height of each letter is its "importance" to a predictive model and motifs pop out. But the technical details behind how these are calculated can matter a lot -- and I'm worried many may be done incorrectly. 1/

@daniel_kraft @NEJM_AI @NEJM @zakkohane Where can I access the supplementary materials?

#GPT4 correctly diagnosed 57% of complex clinical cases, outperforming 99.98% of simulated human readers. Full article in the new @NEJM_AI onepub-media.nejmgroup-production.org/ai/media/ec2de… Via @NEJM-#AI Editor in Chief @zakkohane