MIT researchers compile a trained neural network into DNA inside mammalian cells
NikoMcCarty · x · 2026-09-23
An MIT-led team (Jean Disset, Georg Wachter, Ron Weiss, et al.) published a bioRxiv preprint, Generative Neuromorphic Programming of Mammalian Cells, showing how to run neural-network computation in living cells.
- Pipeline: train an artificial neural network, prune it, then "compile" the result into DNA sequences
- The compiled network was implemented in human cells, effectively executing the network's logic as a gene circuit
- The authors shared screen captures of the training and pruning process
- A cross-disciplinary neuromorphic effort bridging AI and synthetic biology, with contributors from MIT, Boston University, and the University of Maryland
More from Research
- Story Imprinting Paper Finds AI Assistants Absorb Traits From Resembling Human Characters — OwainEvans_UK · 2026-09-23
- Yann LeCun: desk-reject papers that fail LLM watermarking detection — RexDouglass · 2026-09-23
- Pedro Domingos unveils Tensor Logic, an AI language unifying deep learning and symbolic AI — pmddomingos · 2026-09-23
- Stanford's Bayesian pulse deconvolution extracts accurate heart signals from wearables — rbhar90 · 2026-09-23
- Researchers lament: no good RL environment sets exist that allow diffing behaviors across training runs — 1a3orn · 2026-09-23
- New PNAS paper: human learning is an understudied lever for boosting human–AI synergy — iyadrahwan · 2026-09-23