Graph Neural Network Predicts Protein Topology Using All Atom Embeddings

Researchers have developed a new method for predicting transmembrane protein topology using the SchNet graph neural network (GNN). This approach utilizes all atom-level embeddings, a departure from traditional methods that rely solely on protein sequences or alpha-carbons. Trained on the same dataset as the DeepTMHMM model with 5-fold cross-validation, the GNN shows significant potential for topological predictions without the need for pre-trained weights. AI

IMPACT This research demonstrates the potential of graph neural networks for complex biological predictions, potentially advancing drug discovery and protein engineering.

RANK_REASON The cluster contains a research paper detailing a new methodology for protein topology prediction using graph neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• DeepTMHMM
• Gotit.pub
• Hugging Face
• Influence Flower
• ScienceCast

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →