iPBA

iPBA performs fast and accurate pairwise protein structure comparisons by translating 3D structures into Protein Blocks (PBs) and aligning the resulting PB sequences for structural analysis.

Key Features:

• Protein Blocks (PBs): Uses a structural alphabet of 16 pentapeptide conformations to convert three-dimensional protein structures into one-dimensional PB sequences.
• Needleman-Wunsch alignment: Employs the Needleman-Wunsch algorithm to align PB sequences for global structural comparison.
• PB Substitution Matrices: Utilizes specialized PB substitution matrices (SM) tailored to improve structural alignment accuracy over raw PB substitution matrices.
• Anchor-based alignment: Implements an anchor-based methodology to identify anchor points within structures and refine alignments.
• Local alignment option: Provides local alignment capability to detect sub-structural similarities.
• Output formats: Produces sequence alignments and superposed 3D structures with visualization via PyMol and Jmol.
• Performance metrics: Demonstrates improved alignment quality in approximately 88% of cases and outperforms DALI, MUSTANG, and GANGSTA(+) in over 80% of instances.
• Applications supported: Supports both pairwise comparisons and database searches for structural analysis.

Scientific Applications:

• Pairwise structural comparison: Comparative analysis of two protein structures using PB-based alignments.
• Database searching: Identification of structurally similar proteins through PB-sequence-based searches.
• Sub-structural similarity detection: Detection of local or partial structural similarities via local alignment.
• Method benchmarking: Empirical comparison and benchmarking against methods such as DALI, MUSTANG, and GANGSTA(+).

Methodology:

Conversion of 3D structures into a 1D sequence of 16 PBs, alignment of PB sequences using the Needleman-Wunsch algorithm with specialized PB substitution matrices, and refinement via an anchor-based alignment approach with an optional local alignment mode.