Anthony Costa (@anthonycosta) / X

Out today! We pre-train a health system language model, NYUTron, and evaluated it on diverse prediction tasks, including 30-day all-cause readmission, in-hospital mortality, comorbidity index, LOS, and insurance denial. nature.com/articles/s4158…

288K

Anthony Costa

@anthonycosta

Nov 14, 2024

This is a foundational achievement. MMseqs2-GPU virtually *erases* the bottleneck of homology search in many workflows, including structure prediction. @sacdallago @milot_mirdita @thesteinegger, Felix, Bertil, and Alejandro too!

Boost Alphafold2 Protein Structure Prediction with GPU-Accelerated MMseqs2 | NVIDIA Technical Blog

From developer.nvidia.com

Really excited to work with the EvolutionaryScale team, @alexrives @TomSercu @salcandido and so many more to accelerate the very frontier of generative biology. Also congrats to all our awesome partners for their work with the ES team!

EvolutionaryScale Debuts With ESM3 Generative AI Model for Protein Design

From blogs.nvidia.com

Absolutely thrilled to announce the availability of cuEquivariance v0.5 and our contributions to Boltz-2! cuEquivariance v0.5 is a huge release -- now including accelerated triangle attention and multiplication kernels, fundamental to performance of next-gen geometry-aware NNs.

Congratulations to @peholderrieth @msalbergo and Tommi Jaakkola for winning the best paper award for their work entitled "LEAPS: A discrete neural sampler via locally equivariant networks" at this year's Frontiers in Probabilistic Inference workshop #ICLR2025!

I’m #hiring! Are you interested in joining us to help build the future ecosystem of generative AI in drug discovery at @NVIDIA @NVIDIAHealth? Come join us! Apply here: bit.ly/BioNeMoAllianc…

Check out new work from @NVIDIAHealth Genomics, @genentech, transcriptome-guided genomic alignment with transformer-based splice junction scoring, delivering SOTA in splice junction detection, indel identification, and resolving complex splicing patterns

biorxiv.org

DeepSAP: Improved RNA-Seq Alignment by Integrating Transcriptome Guidance with Transformer-Based...

Advancements in high-throughput sequencing have revolutionized the field of transcriptomics, pro-viding unprecedented insights into gene expression, splicing events, and fusions. Despite these...

This week, we released BioNeMo FW 2.0, a complete rewrite of our work accelerating foundation modeling in biology, focusing on a more modular accelerated architecture and a much cleaner, easier-to-program API. arxiv.org/abs/2411.10548…

GatorTron models are now available officially on Huggingface! Includes a new larger 3.9B parameter variant trained on the same corpus.

UFNLP (University of Florida NLP Group)

Incredibly excited to support and partner with @arcinstitute to accelerate our understanding of biology -- remember, if you build great biology FMs, you can use them to learn new biology!!! Critical for the future of life sciences and drug discovery.

Arc Institute launches its inaugural “virtual cell” competition using AI to solve one of biology’s...

From arcinstitute.org

Getting ready for what is sure to be an awesome panel on translation of ML into the life sciences with @mmbronstein @sacdallago Andrew Phillips and Sepp Hochreiter!

I finally did it, a picture with the great @mmbronstein!! Enjoying the talk so far at #GTC2025, “Advancing Molecular Design with Deep Learning”.

Had an amazing time talking about AI, Bio, and @NVIDIAHealth with @Dyno_Tx @ekelsic at #NYTechWeek! Thanks so much to @samsinai and the whole Dyno team for making it such a success!

Highly recommend checking out @KevinKaichuang's talk at GTC today on Multimodal Deep Learning for Protein Engineering, now available on demand. Struck a great balance for LLM experts and biologists alike. I learned a ton. register.nvidia.com/flow/nvidia/gt…

3.5K