- OpenAI and Retro Biosciences developed a specialized AI model, GPT-4b micro, focused on protein engineering.
- The model created new versions of the Yamanaka factors, proteins that help turn adult cells into stem cells.
- Lab tests showed these new proteins made stem cell conversion 50 times more efficient than previous methods.
- Researchers say this AI-driven process could speed up work in life sciences and the development of treatments linked to aging.
- The research is in early stages, and experts warn that results may not easily move from the laboratory to real-world medical use.
OpenAI and Silicon Valley startup Retro Biosciences have collaborated to create GPT-4b micro, an Artificial Intelligence model designed for protein engineering. Announced in a company blog post, the new model specializes in redesigning proteins used in regenerative medicine.
GPT-4b micro was trained using data from protein sequences, scientific literature, and 3D structural information. The AI developed novel versions of the Yamanaka factors—proteins that can reprogram adult cells into stem cells. According to lab results, these AI-designed proteins promoted 50 times higher stem cell marker expression and improved DNA repair compared to previous variants.
Stem cells have the ability to self-renew and transform into many other cell types. This makes them central to efforts in tissue repair, disease treatment, and the study of aging. Typically, the process to turn adult cells into stem cells is slow and inefficient, with fewer than 0.1% of cells converting and the procedure taking several weeks.
Lab data from Retro Biosciences suggests that the new AI-engineered proteins greatly boost the efficiency of cell reprogramming. “When researchers bring deep domain insight to our models, problems that once took years can shift in days,” said Boris Power, who leads research partnerships at OpenAI.
The Yamanaka factors are important in regenerative medicine and have applications in treatments for conditions like blindness, diabetes, and organ failure. The possibility of using AI to find better protein variants may help cut down drug development time and expand synthetic biology into new areas researchers previously could not explore.
OpenAI has stated that this work is a proof-of-concept and the findings are still in the lab-testing phase. There are concerns about how well these new proteins will work in living organisms and the potential risks if such powerful tools are misused. OpenAI and Retro Biosciences are publishing their research openly to encourage replication and critical review.
✅ Follow BITNEWSBOT on Telegram, Facebook, LinkedIn, X.com, and Google News for instant updates.
Previous Articles:
- Canada to Lift Retaliatory Tariffs on Many US Products: Bloomberg
- $800K in Top NFTs Trapped on Defunct Parallel Finance Platform
- Bitcoin Soars to $116K After Powell Hints at September Rate Cut
- INTERPOL Cracks Down on Cybercrime: 1,209 Arrested in Africa Sweep
- Philippines Proposes Strategic Bitcoin Reserve for National Security
