Today for AI
HOT RADAR
llmHEAT 10.6°

AI CURATED HOT STORY · 11/14/2026

Tsinghua Unveils ProtCell Foundation Model for Single-Cell Proteomics

1 reports archived1 independent sourcesupdated 11/14/2026, 15:09:10
Synthesis & Latest Updates
1 Sources Cross-Validated

Tsinghua University's Jiang Rui research group has released ProtCell, a foundation model for single-cell proteomics pretrained on roughly 380 million cells across 98 public datasets, capable of mapping cell states and inferring unmeasured protein expression to support drug discovery and virtual-cell construction. The paper also reports a full pretraining reproduction on Moore Threads' MTT S5000 accelerator, with key quantitative metrics deviating only about 0.2% from mainstream international GPU platforms.

LATEST/Tsinghua University's Jiang Rui research group has released ProtCell, a foundation model for single-cell proteomics pretrained on roughly 380 million cells across 98 public datasets, capable of mapping cell states and inferring unmeasured protein expression to support drug discovery and virtual-cell construction. The paper also reports a full pretraining reproduction on Moore Threads' MTT S5000 accelerator, with key quantitative metrics deviating only about 0.2% from mainstream international GPU platforms.

Hourly heat curve

5 fully observed hours
Now
10.6
Peak
11.816:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. IT之家 · 智能时代T2·74 pts

    Tsinghua Unveils ProtCell Foundation Model for Single-Cell Proteomics

    • ProtCell is pretrained on about 380 million cells from 98 public datasets for cell-state mapping and unmeasured protein inference
    • Full pretraining was reproduced on Moore Threads MTT S5000 with roughly 0.2% deviation in key metrics versus mainstream GPUs
    • Virtual cells are a strategic AI4S frontier, and genomic data security drives demand for self-controlled compute