PROTEUS2

PROTEUS2 predicts protein secondary and tertiary structure and annotates signal peptides and transmembrane helices from amino-acid sequences, supporting single-sequence and multi-sequence (proteome-scale) analyses.

Key Features:

• Signal Peptide Identification: Achieves a Q2 score of 94% for signal peptide detection.
• Transmembrane Helix Prediction: Reaches a Q2 score of 87% for transmembrane helix prediction.
• Secondary Structure Prediction: Delivers a Q3 score of 81.3% for secondary structure assignment.
• Homology Modeling: Generates homology models with RMSD within 0.2 Å, comparable to SWISS-MODEL and 3D JigSaw.

Scientific Applications:

• Protein Structure Prediction: Supports studies on individual protein sequences and whole-proteome annotation using multiple-sequence inputs.

Methodology:

Pipeline uses progressive multi-sequence alignment, structure-based mapping, hidden Markov models, multi-component neural networks, and updated secondary structure databases; predictions complete in approximately three minutes per query sequence.