FTAlign
FTAlign performs topology-independent and global protein structure alignments using a Fast Fourier Transform (FFT)-based exhaustive search to enable accurate comparisons of protein structures, including non-sequential alignments.
Key Features:
- Topology Independence: Aligns protein structures without requiring preservation of topological order, enabling non-sequential alignments.
- Global Alignment Capability: Produces global structure alignments covering entire protein structures.
- FFT-Based Algorithm: Uses a Fast Fourier Transform (FFT)-based exhaustive search algorithm to identify optimal alignments.
- Performance Metrics: Evaluated on six datasets and reported to outperform TMalign, DeepAlign, Kpax, 3DCOMB, MICAN, SPalignNS, and CLICK, achieving 96.7% on the MALIDUP benchmark and 90.0% on the MALISAM benchmark.
- Biologically Meaningful Overlap: Produces biologically meaningful structure overlaps (SO) and TM-scores, with particular strength in non-sequential alignment scenarios.
- Computational Efficiency: Despite exhaustive search, typically completes pairwise alignments within one second.
Scientific Applications:
- Protein structure comparison: Enables accurate structural comparisons for detecting similarity between protein folds, including non-sequential relationships.
- Functional and evolutionary inference: Supports analyses of protein function and evolutionary relationships based on structural alignment.
- Interaction and atypical structure analysis: Facilitates study of protein interactions and analysis of complex or atypical protein structures that require topology-independent alignment.
Methodology:
Performs an exhaustive Fast Fourier Transform (FFT)-based search to compute topology-independent and global structure alignments and reports structure overlap (SO) and TM-score metrics.
Topics
Details
- Tool Type:
- web application
- Added:
- 11/14/2019
- Last Updated:
- 12/29/2020
Operations
Publications
Wen Z, He J, Huang S. Topology-independent and global protein structure alignment through an FFT-based algorithm. Bioinformatics. 2019;36(2):478-486. doi:10.1093/bioinformatics/btz609. PMID:31384919.
PMID: 31384919
Funding: - National Natural Science Foundation of China: No. 31670724
- National Key R&D Program of China: 2016YFC1305805, Nos. 2016YFC1305800