So excited to share that we won the best poster award!! @icml_compbio #ICML @omarnmahmood, @RichBonneauNYC, @kchonyc

The Skok Lab is recruiting an experienced computational scientist to lead single-molecule epigenomic and 3D genome analysis. We integrate long-read sequencing with Hi-C/Hi-ChIP, scmulti-omics, and machine-learning to explore chromatin topology, and gene regulation.

GLM-Prior reframes expression as a dynamic annotation over a latent regulatory scaffold—moving the challenge from structure discovery to interpreting regulatory activity across conditions.

We propose a shift in GRN modeling: rather than using expression to discover regulatory structure, it should be used to contextualize or modulate a fixed, sequence-derived network.

When combined with expression data for GRN inference, we find that expression adds little when the prior is strong — revealing that prior quality, not inference, has been the main performance bottleneck.

This sequence-derived prior outperforms motif-based and curated databases, generalizes across yeast, mouse, and human, and enables zero-shot transfer between species.

In this work, we fine-tune the Nucleotide Transformer, pretrained on 850 species, to predict TF-gene regulatory interactions from nucleotide sequence alone.

Excited to share our new pre-print! GLM-Prior: a nucleotide transformer model reveals prior knowledge as the driver of GRN inference performance https://t.co/9WhYH9PmtO @_angie_chen @RichBonneauNYC @kchonyc

Thrilled to share our new preprint, 'A genome wide code to define cell-type specific CTCF binding and chromatin organization'. https://t.co/tzzEfysBHi

Exciting research from CDS PhD student Claudia Skok Gibbs (@cskokgibbs)! Her new paper on PMF-GRN improves gene regulatory network inference, potentially opening doors to better disease understanding and treatments. Learn more:

it's rare for me to use it for my research, but matrix factorization is my favourite framework in ML. thanks to @cskokgibbs, @omarnmahmood and @RichBonneauNYC, i had a pleasure of using it for gene regulatory network inference:

CDS PhD student Claudia Skok-Gibbs @cskokgibbs, CDS PhD alumnus Omar Mahmood @omarnmahmood, former CDS Associated Faculty Richard Bonneau @RichBonneauNYC, and CDS Associate Professor of Computer Science and Data Science Kyunghyun Cho @kchonyc received a Best Poster Award at ICML!

@cskokgibbs (@nyuniversity) et al won the best paper award at @icml_compbio with their stellar study on variational inference for single-cell gene network inference - providing state-of-the-art accuracy as well as well-calibrated uncertainty estimates. https://t.co/XV0ynt2Guq

Thank you to all organizers and sponsors @cziscience @GSK for another outstanding @icml_compbio workshop @icmlconf #ICML2023 photos courtesy of @shi_lingting

My intern is so good she does a summer internship AND still presents her PhD work at a conference.

Excited to share I will be presenting our work at the #ICML Computational Biology workshop today. Come stop by to hear a spotlight talk at 10am (room 314) or see our poster from 11:50am to 1:30pm. @omarnmahmood @RichBonneauNYC @kchonyc

NYU AI School 2023 is going strong 🔥

CDS PhD student Claudia N. Skok-Gibbs (@cskokgibbs) recently received a National Science Foundation (NSF) Graduate Research Fellowship (GRFP).

Only one week left to apply for the *FREE* NYU AI School *in-person* @NYUDataScience. We encourage undergrads with non-computer science backgrounds from underrepresented communities to apply!! Don’t miss out on this exciting opportunity to learn about AI/ML/DS

Beyond excited to share that I have been selected to receive the NSF Graduate Research Fellowship! 🥳🥳🥳