Layers

Layers transforms three-dimensional protein structures into two-dimensional residue transition patterns (RTPs) by peeling the molecular surface to enable orientation-independent analysis of protein folding, structural similarity, and coarse surface modeling.

Key Features:

• Residue Transition Pattern (RTP) Generation: Peels protein atoms layer by layer from the molecular surface to convert 3D structures into orientation-independent RTPs.
• Efficient Protein Similarity Identification: Uses RTPs to identify structural similarities between proteins independent of spatial orientation.
• Coarse Model Creation: Builds a fine-tunable coarse model of the molecular surface via non-random sampling for shape comparison, protein recognition, and ligand design.
• Protein Folding Simulations: Produces biased initial configurations of molecules to support protein folding simulations.
• Integration with Machine Learning: Employs a random forest classifier to predict RTPs from polypeptide sequences.
• Scalability and Integration: Supports integration with other software to manage large datasets of protein structures and reduce computational demands.

Scientific Applications:

• Protein Structure Analysis: Converts 3D structures into RTPs to analyze folding patterns and structural organization.
• Comparative Protein Studies: Identifies structural similarity across proteins for comparative and evolutionary analyses.
• Drug Design and Development: Provides coarse molecular surface models useful for ligand design and shape-based comparisons.
• Protein Folding Research: Supplies biased initial configurations to enhance simulations investigating folding mechanisms.

Methodology:

Atom-by-atom layer peeling of the molecular surface to generate orientation-independent RTPs; non-random sampling to create fine-tunable coarse surface models; and a random forest classifier to predict RTPs from polypeptide sequences.