OWKIN ads

Your biotech can hire an AI Scientist tomorrow. But which one? Join our webinar with Roy Granit (Compugen) and Jonas Béal (Owkin)

AI Scientists are everywhere. What does it mean for drug discovery? Join our webinar with Roy Granit (Compugen) and Jonas Béal (Owkin)

Competitors’ reports give you bulk. Owkin’s reports gives you spatial. Better resolution, better quality, better insights - all for the same price.

Explore deep multimodal patient data in NSCLC and beyond, from spatial to single-cell. Book a meeting to stay ahead of your research.

Claude Science launched, and reading between the lines, it says something important about where AI for biology is heading. In some places, it’s a big step forward. Give researchers better ways to run analyses, reproduce results and move through complex workflows, and you remove real day-to-day friction from science. 𝗕𝘂𝘁 𝗯𝗶𝗼𝗹𝗼𝗴𝘆 𝗶𝘁𝘀𝗲𝗹𝗳 𝗵𝗮𝘀 𝗻𝗼𝘁 𝗯𝗲𝗰𝗼𝗺𝗲 𝗮𝗻𝘆 𝗲𝗮𝘀𝗶𝗲𝗿. The harder question is not only which experiment to run next, but which question matters, in which biological context, and which data can be trusted, so the result actually move closer to clinic. That still comes down to messy, contextual, dynamic judgement calls that live in someone's head, institutional knowledge and proprietary data, not in a curated library of tools! 𝗧𝗵𝗲 𝗼𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝘆 𝗡𝗢𝗪 𝗶𝘀𝗻'𝘁 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 "𝗮𝗵𝗮" 𝗺𝗼𝗱𝗲𝗹. It's parallel science: many hypotheses tested at once, with the strongest moving fastest toward the bench, and, ultimately, to patients. 𝗔𝗻𝗱 𝘁𝗵𝗲𝗻 𝘁𝗵𝗲𝗿𝗲'𝘀 𝗱𝗮𝘁𝗮. Plugging into public tools and standard formats is one thing. Getting your own structured, clean, well-annotated, ontology-consistent data in order is an entirely different and much less glamorous project. Nobody ships a keynote about ontology decisions, but they're the thing that decides whether any of this compounds. We talked through exactly this at our AI for Biology dinner in Zurich last month, alongside brilliant minds from AstraZeneca and Boehringer Ingelheim. Better tools were not the thing anyone in that room was worried about. Trust, context and iteration were. Read the full blog below 👇 https://lnkd.in/e49u4jdH I’d be interested to hear how others are thinking about this next phase of AI for biology. DM me and let’s compare notes! #AI #Ecosystems #PrecisionMedicine #Owkin

Automatically detect cell types and tissue structures from histology data. See how K-Pro delivers pan-cancer cell detection, segmentation and classification in H&E-stained pathology images — outperforming SOTA with 5% better detection and 24% higher F1 classification, using 5x fewer parameters.