ai for science - IoT Digital Twin PLM

AI Plasma Control for Tokamak Fusion: Reinforcement Learning (2026)

By MPRAUTO MPRAUTO July 30, 2026ScienceNo Comments

How reinforcement learning controls tokamak plasma: the magnetic-control problem, the RL controller architecture, sim-to-real transfer, safety interlocks and what still limits it.

Single-Cell Foundation Models: scGPT & Geneformer (2026)

By MPRAUTO MPRAUTO July 26, 2026ScienceNo Comments

Single-cell foundation models explained: how scGPT, Geneformer and scFoundation pretrain transformers on tens of millions of cells for cell typing, perturbation and gene networks - architecture, benchmarks and caveats in 2026.

How AI Weather Forecasting Models Work: GraphCast, GenCast, Aurora (2026)

By MPRAUTO MPRAUTO July 25, 2026Science1 Comment

How AI weather forecasting models work: GraphCast graph neural networks, GenCast diffusion ensembles and Microsoft Aurora - encoder-processor-decoder design, training on reanalysis, skill vs ECMWF and honest limits in 2026.

AI for Science Landscape 2026: Periodic Labs, Lila Sciences, and the Self-Driving-Lab Race

By MPRAUTO MPRAUTO July 12, 2026TechNo Comments

A 2026 map of the AI-for-science landscape: Periodic Labs, Lila Sciences, Berkeley A-Lab, Emerald Cloud Lab, and academic SDLs — capital, technical approach, moats, and honest caveats.

The Experimental-Data Moat: Why AI Labs Are Building Robots to Make Their Own Data (2026)

By MPRAUTO MPRAUTO July 12, 2026TechNo Comments

The public internet is running dry as a training source. Why frontier AI-for-science labs are manufacturing proprietary physical-world experimental data — the compounding-data thesis, explained.

Connectomics in 2026: Mapping the Brain Wire by Wire with AI

By MPRAUTO MPRAUTO July 8, 2026Science1 Comment

How connectomics maps the brain: serial electron microscopy, AI segmentation, the fly and mouse connectomes, and why wiring diagrams are reshaping neuroscience.

AI Protein Design: How RFdiffusion Generates New Proteins (2026)

By MPRAUTO MPRAUTO June 24, 2026Science1 Comment

How generative AI designs proteins from scratch: the RFdiffusion denoising pipeline, ProteinMPNN sequence design, AlphaFold validation, and wet-lab loop.

Predicting Protein Conformational States with AI (2026)

By MPRAUTO MPRAUTO June 8, 2026ScienceNo Comments

How AI is moving beyond static AlphaFold structures to predict protein conformational states and ensembles in 2026, the methods, the limits, and why it matters.

Cryo-EM Meets AI: Structure Prediction in Drug Discovery

By MPRAUTO MPRAUTO June 6, 2026ScienceNo Comments

How cryo-EM and AI structure prediction work together in 2026 drug discovery — the pipeline, what AlphaFold-class models add, and the honest limitations.

RFdiffusion 2: How AI Now Designs Functional Proteins (2026)

By MPRAUTO MPRAUTO June 2, 2026ScienceNo Comments

RFdiffusion 2 explained — what changed, how it pairs with ProteinMPNN, success rates on real binder campaigns, and the open problems.