SophontAI launches Holon-S1 to predict gene activity from routine pathology slides
iScienceLuvr · x · 2026-10-22
AI startup SophontAI has launched Holon-S1, a platform that predicts gene expression and biomarkers directly from routine pathology slides, bypassing slow and expensive molecular testing.
- Motivation: cancer patients with the same diagnosis often have very different tumor biology, and molecular testing is slow and costly.
- In a controlled comparison, Holon-S1 beat two other models at predicting spatial gene activity.
- It has identified potentially novel biomarkers in breast and colon cancer.
- Evaluated across 16 organ types and 15+ scanners, using both FFPE and fresh-frozen tissue slides.
The team believes researchers can use it to find new biomarkers in slides they already have and discover treatments.
Related event: Sophont Launches Holon-S1 to Predict Gene Expression from Pathology Slides(24 posts)→
More from Apps
- Endless Powder launches free on Spawn: endless procedural mountain with a synth-soundtrack snowboard — sterlingcrispin · 2026-10-22
- Nest Hub users report Gemini replacing web browsing with AI-only answers — JesseBell935 · 2026-10-22
- Endless Powder dev details snowboard footprints: a shape, not a point stamp — sterlingcrispin · 2026-10-22
- Notion tip: select any block to see if it was last edited by a teammate or a bot — ivanhzhao · 2026-10-22
- rabbit ships OS3 update with new Fun mode, smarter background workers and cleaner settings — SimonBalmain · 2026-10-22
- ChatGPT Dots Rollout Rules: Business Premium Worldwide, Pro Tiers Skip EEA — py-net · 2026-10-22