Related Experiment Videos
CATH--a hierarchic classification of protein domain structures
C A Orengo1, A D Michie, S Jones
1Department of Biochemistry and Molecular Biology, University College London, UK.
Structure (London, England : 1993)
|August 15, 1997
Summary
Protein evolution creates diverse protein families with low sequence identity. A new hierarchical classification (CATH) aids in automatically identifying structural relationships and assigning protein functions.
Area of Science:
- Structural bioinformatics
- Computational biology
- Protein science
Background:
- Protein evolution generates families with low sequence identity, necessitating structure-based classification.
- Manual classification of the growing protein structure database is infeasible, requiring automated methods.
- Structure-based classifications can reveal hidden relationships and aid in function assignment.
Purpose of the Study:
- To develop a semi-automatic method for classifying protein domain structures.
- To establish a hierarchical classification system for protein structures.
- To facilitate the assignment of structure-function/evolution relationships.
Main Methods:
- Developed a semi-automatic procedure for hierarchical classification of protein domain structures.
- Defined four classification levels: Class (C), Architecture (A), Topology (T), and Homologous Superfamily (H).
- Grouped proteins into superfamilies based on structural similarity and functional evidence.
Main Results:
- Introduced the CATH classification system with four hierarchical levels.
- Identified superfolds (major T-levels) and simple architectures.
- Found that a third of superfamilies belong to ten superfolds, and two-thirds cluster into nine architectures.
Conclusions:
- The CATH classification provides insights into protein structure space.
- It aids in assigning structure-function/evolution relationships to known and new protein structures.
- A well-characterized database like CATH is crucial for advancing protein science.