Atomwise
Deep learning for structure-based drug discovery — AtomNet screens 16B+ compounds to find hits for any disease target
Gallery
4 items
About Atomwise
Atomwise is the pioneer of deep learning for structure-based drug discovery, having invented and validated the use of convolutional neural networks in molecular screening more than a decade ago. Founded in 2012 and headquartered in San Francisco with $226M+ in funding from DCVC, Khosla Ventures, Y Combinator, Tencent, Baidu, and B Capital Group, Atomwise's AtomNet technology remains the most validated AI platform for structure-based hit discovery in the industry.
AtomNet — the world's first deep convolutional neural network built specifically for drug discovery — can screen over 16 billion compounds against a disease target in under two days, predicting binding affinity and identifying novel small molecule candidates with therapeutic potential. This virtual screening capability has powered 775+ collaborations across 250+ partners including Merck, four of the top 10 US pharmaceutical companies, and 40+ major research universities (Harvard, Stanford, Duke, Baylor). Atomwise has addressed 600+ unique disease targets, with demonstrated success across protein types and hard-to-drug targets that previously resisted conventional approaches.
Under its new CEO Steve Worland, Atomwise is now pivoting from a pure platform collaboration model to building its own proprietary pipeline of preclinical drug candidates, while also launching its next-generation Foundation Model for molecular discovery. Atomwise is ideally suited for pharma and biotech R&D teams seeking AI-powered hit discovery and lead optimization against challenging targets at industrial scale.
Key Features
- AtomNet — world's first deep CNN for structure-based drug discovery, screening 16B+ compounds in under 2 days
- 600+ unique disease targets tackled across oncology, CNS, infectious disease, and rare diseases
- 185+ completed projects validated across diverse protein types including undruggable targets
- 775+ collaborations with 250+ partners worldwide — pharma, biotech, agrochemical, and academic
- Foundation Model for molecular discovery — next-generation AI announced at TRANSONC 2026
- Proprietary pipeline of small-molecule drug candidates advancing into preclinical studies
- AIDDISON platform for structure-based hit discovery and lead optimization
- Co-development agreements with major pharma — Merck, and 4 of top 10 US pharmaceutical companies
- Academic ATOM program: collaborations with 40+ universities including Harvard, Stanford, and Duke
- Founded 2012 — pioneered deep learning in drug discovery; $226M+ total raised
Pros
- Invented deep learning for structure-based drug discovery in 2012 — true pioneer in the field
- AtomNet screens 16B+ compounds in under 2 days — fastest virtual screening at this scale
- 775+ collaborations and 250+ partners — most validated partnership network in AI drug discovery
- Tackles undruggable targets that traditional methods can't address
- Strong academic program: 40+ universities including Harvard, Stanford, and Duke
- New Foundation Model announced 2026 — next-generation AI for molecular discovery
- $226M+ raised from DCVC, Khosla, Y Combinator, Tencent, B Capital, and Baidu
Cons
- Early-stage company — no FDA-approved drugs from the platform yet
- Shifting focus from platform licensing to proprietary pipeline adds execution risk
- Smaller team (~100-250 employees) compared to larger AI drug discovery peers
- Revenue still modest at $5-25M range — pre-commercial stage
- Deep tech focus means long timelines before patient impact is realized
Pricing
Atomwise operates through enterprise partnership and co-development agreements — there is no self-serve access or published pricing.
Collaboration types include: sponsored research agreements (pharma/biotech pays Atomwise to run AI-powered hit discovery campaigns), co-development partnerships (shared IP and milestone/royalty structures), and academic program access for universities through the ATOM initiative.
New CEO Steve Worland is also steering Atomwise toward an internal proprietary drug pipeline, which will generate revenue through future licensing and partnering deals as candidates mature into preclinical and clinical stages.
Contact Atomwise through atomwise.com for partnership and collaboration inquiries.
Claim Verified Creator Badge
Are you the founder of Atomwise? Display this listing's verified badge on your website to show your customers that your product has been vetted and listed on AI Central Resources.
<a href="https://www.aicentralresources.com/tool/atomwise" target="_blank" rel="noopener"> <img src="https://www.aicentralresources.com/badges/featured-badge-dark.svg" alt="Featured on AICentralResources" width="200" height="54" style="border: none;" /> </a>
* Place this HTML snippet in your website's footer, landing page, or press section. This creates a search-friendly backlink directly to your verification page.
Connect with Atomwise
Frequently Asked Questions
Atomwise uses its AtomNet deep learning model to virtually screen billions of molecular compounds against disease protein targets — identifying which molecules are most likely to bind effectively. This replaces expensive and time-consuming physical wet-lab screening with AI-powered virtual screening that is orders of magnitude faster.
AtomNet is the world's first deep convolutional neural network designed specifically for structure-based drug discovery. It can screen over 16 billion compounds in under 2 days, predicting binding affinity to disease proteins far faster than traditional high-throughput screening.
Atomwise works with 250+ partners across pharma, biotech, and agrochemicals — including Merck and four of the top 10 US pharma companies — as well as 40+ major research universities including Harvard, Stanford, Duke, and Baylor College of Medicine.
Atomwise has tackled over 600 unique disease targets including cancer, infectious disease, CNS disorders, and rare diseases. It has completed 185+ projects across a wide range of protein types, including many historically considered undruggable.
Atomwise raised $45M in its Series C in early 2025, appointing Steve Worland Ph.D. as new CEO. Total funding stands at $226M+. The company is shifting from a pure platform/collaboration model to building its own proprietary pipeline of small-molecule drug candidates advancing into preclinical studies.
Similar AI Tools to Atomwise
View all alternatives of Atomwise
BenevolentAI
Clinical-stage AI drug discovery with integrated wet labs and AstraZeneca and Merck collaborations
Insilico Medicine
Generative AI drug discovery platform — from target identification to Phase 3 clinical trials with Pharma.AI
Recursion Pharmaceuticals
AI-native drug discovery OS with industrial-scale automated labs and $500M+ in pharma partnerships
Schrödinger
Physics-based drug discovery with Bunsen AI co-scientist — trusted by 500+ pharma companies worldwide
Infermedica
AI clinical triage and symptom assessment platform — MDR Class IIb certified, 94% accuracy, 23M+ health checks
Tempus AI
AI precision medicine platform — genomic profiling, clinical insights, and trial matching for personalized cancer care
Compared with Atomwise
Direct head-to-head feature matrices, pricing, and category breakdowns