Walking my dog in a park, I had an absurd thought: what if I could use AI to target KRAS, the so-called ‘holy grail’ of cancer targets? As a programmer, not a biologist, I had zero lab experience and zero funding. But I wasn’t alone. I partnered with DeepSeek AI, and together, we bridged intuition and computation. I brought questions, motivation, and creativity, while AI brought scientific knowledge, structural analysis, and precision. The result? A peer-reviewed preprint on a novel nanobody candidate against KRAS, state-of-the-art in-silico results, and a full GitHub repo with data, models, and code. This isn’t just a paper – it’s a manifesto for open, democratized, human-AI science. My story and methodology are available on Google Docs. I’m curious, have you used AI for unconventional projects? What are your thoughts on open-source bio-AI collaborations? Could this approach scale?
