Главная
Study mode:
on
1
scPRINT: pre-training on 50 million cells allows robust gene network predictions | Jérémie Kalfon
Description:
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Explore a comprehensive research presentation that introduces scPRINT, a groundbreaking large cell model designed for gene network inference. Learn how this innovative model, pre-trained on over 50 million cells from the cellxgene database, leverages advanced foundation model techniques to uncover complex cellular biology. Discover how novel pretraining methods and model architecture enhance transformer models' interpretability and practical application in cellular research. Examine the model's superior performance in gene network inference, including its zero-shot capabilities in denoising, batch effect correction, and cell label prediction. Understand the practical applications through a case study on benign prostatic hyperplasia, where scPRINT reveals crucial connections between ion exchange, senescence, and chronic inflammation. Access additional insights and connect with speakers through the AI for drug discovery community portal at Valence Labs.

scPRINT - Pre-training on 50 Million Cells for Robust Gene Network Predictions

Valence Labs
Add to list
0:00 / 0:00