Genome Biology study introduces Pep2Net peptide prediction platform

The wire

1. 01Pep2Net combines a knowledge base assembled from 52 peptide databases with a cross-attention deep-learning model the authors report achieves state-of-the-art prediction of cell-penetrating, anticancer and multi-functional peptides.
2. 02Consolidated peptide databases and prediction tools feed early-stage discovery pipelines that eventually produce candidate therapeutic peptides.
3. 03This is a computational benchmark; no wet-lab or clinical validation of predicted peptides is described.

From the source

Pep2Net Unifies Fragmented Peptide Databases With a Cross-Attention AI Model

A new Genome Biology study introduces Pep2Net, a platform combining a knowledge base built from 52 peptide databases with a cross-attention deep-learning model that achieves state-of-the-art multi-task prediction of cell-penetrating, anticancer and multi-functional peptides.