DeepRHP: A Hybrid Variational Autoencoder for Designing Random Heteropolymers as Protein Mimics
Researchers have developed DeepRHP, a hybrid variational autoencoder designed to aid in the creation of synthetic random heteropolymers that can mimic protein functions. This model uses a semi-supervised framework, incorporating both chemical features and sequence patterns within its latent space. DeepRHP's effectiveness was demonstrated by successfully predicting monomer compositions that stabilize membrane proteins, with predictions validated against existing research. AI
IMPACT This AI model could accelerate the design of novel biomaterials and protein-like structures for various applications.