Anthony Costa
@anthonycosta
Followers
2K
Following
4K
Media
97
Statuses
936
Digital Biology @NVIDIA @NVIDIAHealth
New York, USA
Joined December 2007
This should be fun!
new podcast filmed with someone who basically created a new subfield in bio-ml during her phd, and whose lab at Princeton continues to remain at the forefront of the field out (hopefully) next weekend to prepare, you should read the article below
0
0
0
🚀 NVIDIA unveiled 4 open model families: Nemotron for digital AI, Cosmos for physical AI, Isaac GR00T for robotics & Clara for biomedical AI. 🔥Thrilled that GenMol, ReaSyn & La-Proteina under Clara were developed by my team & BioNemo R&D jointly. https://t.co/fmW3o75CV9
blogs.nvidia.com
NVIDIA’s open model families, including NVIDIA Nemotron for digital AI, Cosmos for physical AI, Isaac GR00T for robotics and Clara for biomedical AI, provide developers with the foundation to build...
0
10
59
Big news! We are teaming up with @NVIDIAHealth to accelerate virtual cell model development to unlock news insights about human biology https://t.co/iJdqlvtMAe
#NVIDIAGTC
chanzuckerberg.com
CZI collaborates with NVIDIA to deliver tools, data, models, and benchmarks through the virtual cells platform.
3
20
90
Should say that OF3p also makes use of Nvidia’s awesome cuEquivariance library:
OpenFold3-preview (OF3p) is out: a sneak peek of our AF3-based structure prediction model. Our aim for OF3 is full AF3-parity for every modality. We now believe we have a clear path towards this goal and are releasing OF3p to enable building in the OF3 ecosystem. More👇
0
13
75
🆕 CodonFM and Clara NV-Reason New open source models added to NVIDIA Clara - a family of open-source AI foundation models, tools, and recipes advancing the frontiers of biomedical research. Details in 🧵👇 #NVIDIAGTC
1
10
21
Very happy that we can use @nvidia cuEquivariance kernels in BoltzGen and still release with MIT license! Many thanks to the team for speeding up the BoltzGen pipeline by at least 1.5x!
Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project has been collaborating with many leading biologists who tested BoltzGen at an unprecedented scale, showing success on many novel targets and pushing its limits! 🧵..
1
8
87
Gonna be in Seoul in a couple weeks. Anyone want to meet up? Have some free time over the weekend.
0
0
0
Even medicine tends to lead to dark places these days. Sigh.
0
0
1
Also who am I not following that can round this place out for me in biology, computing (probably have those down), philosophy of science, photography, aviation, literature, classical music etc.
2
0
0
I think I just need less politics in my social media life. So I find myself in the uncomfortable position of deliberately wanting to unfollow people I respect and want to read — and some I don’t but do to understand other views — to just make daily life here more tolerable. Back
1
0
2
📢 The Fundamental Generative AI Research (GenAIR) team at NVIDIA is looking for outstanding candidates to join us as summer 2026 interns. Apply via: https://t.co/DARXlEpmUg Email: genair-openings@nvidia.com Group website: https://t.co/mWY7utKK1m 👇
research.nvidia.com
4
49
337
What is AvgFlow? It is a model training and inference framework for accelerating flow-based models that generate 3D molecular conformers. Uses SO(3)-Averaged Flow to accelerate model training (fewer epochs needed and eliminating the need of rotation augmentation) and reflow +
3
16
114
Pretraining graph transformers works for ADMET, and the advantage increases with more data and tasks. 📊 In collaboration with @Merck, KERMT is statistically best in 18/30 endpoints on Merck’s internal dataset. Efficient reimplementation is 2.2x faster for fine-tuning and 2.9x
1
13
50
🧬 We’re teaming up with @NVIDIAHealth to bring #BioNeMo models directly into #BenchlingAI. Scientists will be able to access leading models like OpenFold2 for protein structure prediction — right inside their workflows. https://t.co/WYBrVMEgN9
2
11
79
Genomics team keeps cooking!
How fast is fast? âš¡ Over 14x faster runtimes with pangenome-aware DeepVariant 1.9 + Giraffe on four NVIDIA RTX PRO 6000 GPUs vs CPUs. Plus, DeepSomatic long read and WES support, quantMode GeneCounts for STAR, and Mutectcaller mitochondrial mode. What could this mean for your
1
6
58
To this day, I am still primarily a Sony camera shooter -- but I have to admit, Fuji is so, so fun.
1
0
2
Live at #NYBCI25 w/ “Big Tech & #BCI: Powering the Next Interface Revolution" exploring #AI, #HPC & next-gen #neural interfaces. Moderated by @EmilyForlini (@PCMag). Speakers: @vivnat (@DeepMind), Dr @anthonycosta (@NVIDIA & Co-Founder @SinaiBioDesign), & Dr Dan Wetmore (@Meta).
0
1
10
Virtual cell models will accelerate how we understand and treat diseases. Our new suite of benchmarking tools helps developers save time evaluating their models so they can focus on tackling big problems in biology. https://t.co/XcwrDmfNfS
1
5
8