Deep Origin
banner
deeporigin.bsky.social
Deep Origin
@deeporigin.bsky.social
Better molecular modeling more scientists can use. We let scientists skip the sciOps.

https://deeporigin.com
Big news at Deep Origin: we’re leading a $31.7M ARPA-H award to build in-silico models that could replace animal testing.
Animal studies are slow, costly, and often fail to predict human biology.

Read more here 👉 www.deeporigin.com/blog/arpa-h-...
We’re Leading an Up to $31.7M ARPA-H Contract Award to Build In-Silico Models To Replace Animal Testing | Deep Origin
We’re thrilled to share that Deep Origin and our partners in the Pharmacological Research and Evaluation through Digital Integration and Clinical Trial Simulation (PREDICTS) consortium have been award...
www.deeporigin.com
December 4, 2025 at 9:49 PM
GenAI is getting better, but science needs more than text. Claude shows AI must use tools and real data—but chemistry is different: structures, pockets, ADMET, simulations. We dug into what GenAI can/can’t do for chemists + Balto’s 86% LitQA2 score below 👇

www.deeporigin.com/blog/out-of-...
Out-of-the-Box GenAI for Medicinal Chemistry | Deep Origin
What’s out there, what’s useful, what’s game changing
www.deeporigin.com
November 18, 2025 at 3:49 PM
Shout out to Deep Origin team member Aleks Grigoryan who wasn’t satisfied with just being proud of his team on the ground, but took it to the sky and made it loud! 🚀 🪂 🧬
October 18, 2025 at 9:57 AM
One of our much loved team members (and an all around great human being) Kate was in a horrible accident. If you have the means or time to share or contribute to her recovery fund, you will be helping to make the world a brighter place:

gofund.me/9553c9134
Donate to Support Kate's Recovery and Journey Back Home, organized by Matt Shlosberg
Help Bring Kate Home to Heal Dear friends, We are reaching out wi… Matt Shlosberg needs your support for Support Kate's Recovery and Journey Back Home
gofund.me
September 25, 2025 at 9:06 PM
In our new blog post, we dive into:
🔺 Limits of global docking models
🔺 False positives & cryptic pockets
🔺 How Deep Origin PocketFinder (in Balto) boosts accuracy
🔺 >80% success in recovering real binding sites

www.deeporigin.com/blog/a-blind...
A Blind Spot in Drug Discovery: Pocket Finding | Deep Origin
It’s been said a million times before but it’s worth re-emphasizing. Creating. a. new. drug. is. hard.
www.deeporigin.com
September 25, 2025 at 2:58 PM
We're excited to be at #ARDD2025 — the premier conference focused on advancing therapies for aging and age-related diseases. As the field moves from breakthrough research to real-world impact, Deep Origin is here to explore the future of collab with biotech and pharma leaders shaping what comes next
August 4, 2025 at 3:08 PM
Using our OncoSim toolkit—combining ML + mechanistic models across 1,000+ cancer cell lines—we’ve predicted 100+ novel synthetic lethal pairs. Some are validated and may guide therapy & resistance prediction.
Poster by Oliver Purcell @ GRC👇

www.deeporigin.com/academic-pos...
August 1, 2025 at 1:39 PM
Thrilled to share our co-founder and CSO Garegin Papoian in Nature!

Actin cytoskeleton stores stress, unleashing sudden “cytoquakes”. Murrell @ Yale and Papoian @ UMD show orderly filaments diffuse force; tangled networks give power-law bursts.

Read more 👉 nature.com/articles/s41567-025-02919-4
Feedback between F-actin organization and active stress governs criticality and energy localization in the cell cytoskeleton - Nature Physics
Self-organized criticality can occur in cellular systems, but its origins remain unclear. Now it is shown that cytoskeletal criticality is influenced by the F-actin architecture and myosin active stre...
nature.com
June 18, 2025 at 4:58 PM
This summer, we're is opening its doors in Yerevan to curious, driven, and technically skilled students for a 10–12 week paid internship in biosimulations and protein modeling.

Starts: End of June
Location: Yerevan, Armenia
Duration: 10–12 weeks
Paid internship

apply.workable.com/deeporigin/j...
May 23, 2025 at 10:23 AM
We’ve developed a hybrid MM/NNP framework. It delivers high-accuracy solvation and binding ΔG predictions at an average throughput of ~600 ns/day on CPU—10× faster than GPU-based approaches.

Check out our scientific poster below 👇

www.deeporigin.com/academic-pos...
May 12, 2025 at 7:00 PM
DO Challenge — is a new benchmark designed to test autonomous agentic systems in a resource-constrained, simulated drug discovery environment.

We applied it to our own agentic system, Deep Thought, comparing its performance against human teams.

See what we found 👇

arxiv.org/abs/2504.19912
Can AI Agents Design and Implement Drug Discovery Pipelines?
The rapid advancement of artificial intelligence, particularly autonomous agentic systems based on Large Language Models (LLMs), presents new opportunities to accelerate drug discovery by improving in...
arxiv.org
April 29, 2025 at 12:42 PM
We had a great time at #ACSSpring2025! From presenting posters, to hosting a great happy hour, to seeing what's new in chemistry, the energy in-person is great!

We'll share our posters in the coming days. For now, stay in the loop by joining our newsletter! www.deeporigin.com/newsletter
March 31, 2025 at 7:00 PM
Deep Origin Challenge has wrapped up! This year we had over 40 teams made up of over 110 talented programmers, students, data scientists, and life scientists in a 10-hour-long prediction challenge at our Yerevan office 🌄

Want to be involved next year? Stay informed 👉 www.deeporigin.com/newsletter
March 28, 2025 at 1:37 PM
Headed to American Chemical Society Spring '25 expo? Join us Monday night for tacos, drinks, swag and a chance to try out Balto - our AI assistant for drug discovery - with some of the most fun scientists around!

Register below 👇

lu.ma/3sy7pqtb
Democratizing Drug Discovery - Deep Origin Happy Hour · Luma
Are you attending ACS Spring and looking for cutting-edge, AI-powered solutions to advance your research? Join the Deep Origin team for an evening of tacos,…
lu.ma
March 17, 2025 at 7:53 PM
Last chance to sign up for the Deep Origin challenge!

🌄 A day-long AI/ML for drug discovery hackathon
🤝 Bring a team of up to 4 (or sign up yourself)
🤖 Meet and collaborate with leading ML and drug discovery experts
💵 Compete for cash prizes

www.deeporigin.com/events/deep-...
Deep Origin Challenge 2025
Join us for the 2025 Deep Origin challenge
www.deeporigin.com
March 11, 2025 at 3:12 PM
New pub from our team - check out our Smart Distributed Data Factory (think SETI@Home but for molecules) below 👇

www.nature.com/articles/s41...

Want to download the DFT dataset or volunteer compute to create new DFT data points for the community? Find out more here: sddfactory.cloud
Smart distributed data factory volunteer computing platform for active learning-driven molecular data acquisition - Scientific Reports
Scientific Reports - Smart distributed data factory volunteer computing platform for active learning-driven molecular data acquisition
www.nature.com
March 10, 2025 at 4:30 PM
The #DeepOriginChallenge is back! Join us in our Yerevan office for a day of competition and collaboration building efficient #machineLearning models in small teams. Prizes up to $3k and the promise to meet leaders in AI and drug discovery.
Register below!
Deep Origin Challenge 2025
Join us for the 2025 Deep Origin challenge
buff.ly
February 19, 2025 at 2:40 PM
Aram Davtyan giving an awesome poster presentation about our work modeling protein-small molecule-protein ternary complexes at @keystonesymposia.bsky.social Proximity-Based Therapeutics! #biotech #drugDev #degraders #molecularGlues #PROTACs

Find the poster here: www.deeporigin.com/academic-pos...
February 18, 2025 at 3:03 PM
Miss #JPM? Not enough eyes and ears to take it all in? Here's our download of what's coming up in #biotech for 2025 👇

www.deeporigin.com/blog/jpm-25-...
JPM ‘25 From the Trenches: AI in Biotech Gets Tactical | Deep Origin
Our take on the biggest trends and best observations at JPM '25
www.deeporigin.com
January 30, 2025 at 9:35 PM