Atomwise vs Iktos
Find the molecule or design it, which is the oldest division in computational chemistry and both sides are represented here by a company that has been at it for years. Atomwise screens: a deep convolutional network predicts small molecule bioactivity in three dimensions and searches ultra large synthesizable libraries in place of physical high throughput screening, and it published its own denominator, reporting how many selected compounds were tested and how many held up. Iktos designs, with generative medicinal chemistry since 2016 sold both as software and as a discovery collaboration, which gives a pharmaceutical team a way in that does not require handing over the programme. They are sequential rather than competitive. Ask both the same question: how many of your molecules were actually made, and how many worked.
- The method is in the peer reviewed literature, with the core screening result published and the network described and patented rather than called proprietary.
- It reports its own denominator, publishing how many model selected compounds were tested and how many held up, which is the disclosure this category almost never makes.
- Virtual screening against ultra large synthesizable libraries replaces physical high throughput screening, which is a specific and measurable substitution.
- It sells both as licensed software and as a discovery collaboration, so a pharmaceutical team can keep the work in house or hand the programme over.
- Nearly a decade of operating history in generative medicinal chemistry means the workflow has met real programmes rather than benchmark tasks.
- The dual commercial model gives a customer a route out of a services relationship into a licence, which is rare in this category.
Side by Side
| Axis | A Atomwise |
I Iktos |
|---|---|---|
| AI Centrality | ||
| Autonomy and Oversight Model | ||
| Model and Technology Transparency | ||
| Clinical and Operational Evidence | ||
| AI Safety and PHI Stewardship | ||
| HIPAA and BAA Posture | ||
| Security Certifications and Trust Center | ||
| FDA and Regulatory Status | ||
| AI Governance and Bias Disclosure | ||
| EHR and Interoperability Depth | ||
| Deployment Model and Data Residency | ||
| Commercial Transparency | ||
| Setting and Specialty Coverage |
Related comparisons
Other published head to head assessments involving these vendors or their closest peers. The full set for this category is on the Drug Discovery AI page.
These are different steps in the same pipeline rather than substitutes: one screens existing libraries for compounds likely to bind, the other designs molecules that do not exist yet, and a discovery programme may use both. Neither holds a located security attestation and neither publishes pricing.
The comparable to ask both for is what proportion of computationally selected or designed molecules were synthesised and tested, and how many survived, because a proposal that cannot be made or does not hold up is a delay rather than a discovery.