🗞️ Techie uses ChatGPT and AlphaFold to build DIY mRNA cancer vaccine, saves dog

2025-08-21 · Source: Rohan's Bytes · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation, Data Science & Analytics · Depth: Fundamental Awareness, medium

Summary

An Australian engineer, Paul Conyngham, with no medical background, successfully used ChatGPT and Google's AlphaFold to develop a custom mRNA cancer vaccine for his dog, which reduced a tumor by 50%. Conyngham spent $3000 to obtain his dog's healthy and cancerous DNA sequences. He then leveraged ChatGPT as a biological consultant to interpret gigabytes of raw genetic data, identify cancer-causing mutations, and receive step-by-step instructions for data pipelines, including using AlphaFold for protein mapping. This process enabled him to formulate a chemical recipe for an mRNA vaccine, which university researchers subsequently manufactured. This case highlights a significant advancement in "citizen science" and the potential for AI tools to democratize complex scientific research. Separately, Microsoft released GigaTIME, an AI model that converts inexpensive medical images into detailed protein maps of cancer cells, reducing the cost of tumor immune response analysis from thousands to approximately $5 per patient. GigaTIME was trained on 40 million cells and processed 14,256 patients, uncovering 1,234 new connections between cell proteins and patient survival.

Key takeaway

For AI Scientists and Research Scientists exploring novel applications of AI in biotechnology, consider how large language models and specialized AI tools like AlphaFold can accelerate research and development, even without deep domain expertise. This approach can significantly reduce costs and time in areas like personalized medicine and drug discovery, enabling rapid prototyping of solutions. Your focus should be on integrating these AI capabilities to translate complex biological data into actionable insights and treatments, potentially opening new avenues for "citizen science" initiatives.

Key insights

AI tools empower individuals without specialized training to conduct complex scientific research and develop novel solutions.

Principles

• AI can democratize access to complex scientific fields.
• Software can bypass hardware limits in medical diagnostics.

Method

An engineer used ChatGPT to interpret genetic data, identify mutations, and design an mRNA vaccine, then utilized AlphaFold for protein structure mapping.

In practice

• Use LLMs as consultants for complex data interpretation.
• Apply AI for cost-effective medical image analysis.

Topics

• AI in Healthcare
• AI Infrastructure
• AI Societal Impact
• Large Language Models
• AI Agent Development

Code references

• mariodian/jobs

Best for: AI Scientist, Research Scientist, AI Researcher, Policy Maker, General Interest

Related on AIssential

• The Dog That Hacked Cancer — AI tools can accelerate personalized medicine development, even for non-experts.
• 😸 ChatGPT saved this dog's life — AI tools enable non-experts to achieve significant scientific breakthroughs, exemplified by a personalized cancer vaccine for a dog.
• Understanding cancer at a genetic level with AI — AI tools like AlphaFold are democratizing advanced biomedical research in resource-limited settings.
• He Raised $70M to Cure Every Disease With AI — AI agents, particularly specialized models, significantly accelerate scientific discovery by removing human talent bottlenecks and enabling high-throughput reasoning.
• He won a Nobel here for AlphaFold. Then he left. - John Jumper — AlphaFold demonstrates that highly specialized, empirically-driven AI can solve long-standing scientific "holy grail" problems.
• 3 Questions: Using AI to accelerate the discovery and design of therapeutic drugs — AI, particularly deep learning and generative models, significantly accelerates novel antibiotic discovery and design.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Rohan's Bytes.