OpenAI and Retro Biosciences collaborated to develop GPT-4b micro, a scaled-down AI model optimized for protein engineering, to advance cellular reprogramming for longevity research.

Using this AI-driven approach, the engineered factors accelerated the appearance of pluripotency markers in human fibroblasts, confirming successful reprogramming through markers like alkaline phosphatase staining.

The collaboration redesigned two key transcription factors, achieving over fifty-fold increases in stem cell marker expression and significantly improving DNA damage repair in human cells.

AI-designed variants such as RetroSOX and RetroKLF outperformed natural proteins in initial tests, enhancing the efficiency of cellular reprogramming.

These engineered factors not only promote pluripotency but also demonstrate a better ability to restore youthful function by significantly reducing DNA double-strand breaks in aged cells, indicating potential anti-aging benefits.

Retro Bio aims to extend healthy human lifespan by developing anti-aging therapeutics, building on Yamanaka's discovery that adult cells can be reverted to embryonic stem cells using specific transcription factors.

The findings have been replicated across multiple donors, cell types, and methods, confirming both pluripotency and genomic stability in reprogrammed cells.

OpenAI highlights that integrating deep domain expertise with AI tools can drastically reduce the time required for complex biological problem-solving from years to days.

OpenAI's training of GPT-4b micro involved protein sequences, structural data, and contextual annotations, enabling it to generate tailored protein sequences with enhanced properties.