Kathy Chen

AI for science with @zhou_jian at @uchicagomed
Formerly @TroyanskayaLab (@FlatironInst, @Princeton), @hackNY '16

[Kathy Chen @kathyxchen]

Our new preprint (w/ @zhou_jian) is out! 🎉 We developed sequence classes, which allow for easily interpretable yet systematic quantification of the regulatory activities for any sequence & variant, using deep learning-based sequence modeling. (1/13) https://t.co/0xt31Ng9Yu

[Olga Troyanskaya @OlgaTroyanskaya]

Princeton Precision Health is hiring! Be part of our interdisciplinary team leveraging AI and big data to revolutionize human health and health policy! #PrecisionMedicine

[Jian Zhou @zhou_jian]

In light of the recent discussions on UMAP, I'd like to bring to attention this alternative that avoids many issues of UMAP/tSNE: GraphDR can be considered an enhanced PCA that contracts points toward each other based on high dimensional similarity, works well in 2D / 3D / more

[Frank Zijun Zhang @ZijunZhang_]

Unleash the power of #DNA foundation models with #NLP #LLMs - check out our new method LINGO!
We steer the NLP linguistic knowledge for #genome understanding, with just genomic-specific adapters & no pretraining! feedback welcome :) https://t.co/gJPgCEKXsk

[Kseniia Dudnyk @KDudnyk47866]

Excited to share our new work about the sequence basis of transcription initiation in human. We showed that simple rules can explain most human promoters, offering fresh insights into transcription initiation and gene regulation https://t.co/Q4Ex0pQUGJ

[Jian Zhou @zhou_jian]

Glad to share our new preprint on sequence basis of transcription initiation, showing that simple rules can explain the majority of human promoters! Checkout @KDudnyk47866's thread on how sequence can help understand the biology of transcription initiation:

[Frank Zijun Zhang @ZijunZhang_]

↘️First preprint from my lab! We introduce a new #GNN method that capitalizes on complementary information in *multiple* PPI networks for #cancer gene prediction. SOTA performance+ #XAI for bio insights. Fantastic work done by @MichailChatzia1. Feedback welcome!

[Jian Zhou @zhou_jian]

@kathyxchen’s paper for sequence-based global map of regulatory activity is now out (https://t.co/a9Ul7I1rHv)! Congratulations to all co-authors @wongak @OlgaTroyanskaya. Thanks to @anshulkundaje and @nameluem for the News & Views!

[Kathy Chen @kathyxchen]

@ElliotHershberg Thanks so much for checking out our paper! Always happy to chat more about it at any point :)

[Jian Zhou @zhou_jian]

This is now out! And here is an animation that shows a smooth interpolation from a linear visualization (PCA) to GraphDR (with increasing regularization parameter). Colors indicate annotated cell types. https://t.co/KwgX3XDDyQ

[Kathy Chen @kathyxchen]

& huge thanks to @zhou_jian for being such an awesome collaborator, mentor, and friend - it's been so much fun working with you on this project. (Btw, I hear he's recruiting for his new lab at UTSW 😉 https://t.co/1mEsfSZgcb) (13/13)

[Kathy Chen @kathyxchen]

Our new preprint (w/ @zhou_jian) is out! 🎉 We developed sequence classes, which allow for easily interpretable yet systematic quantification of the regulatory activities for any sequence & variant, using deep learning-based sequence modeling. (1/13) https://t.co/0xt31Ng9Yu

[Kathy Chen @kathyxchen]

We also provide code to reproduce the manuscript results: https://t.co/qTS9UgBmIU (12/13)