Related Experiment Videos
18th Sir Hans Krebs lecture. Knowledge-based protein modelling and design
T Blundell1, D Carney, S Gardner
1Department of Crystallography, Birkbeck College, University of London.
European Journal of Biochemistry
|March 15, 1988
Summary
This study presents a knowledge-based protein modeling technique using homologous structures and a database. It enables the design of drugs, vaccines, and novel proteins by accurately predicting tertiary structures.
Area of Science:
- Computational biology
- Structural bioinformatics
- Protein engineering
Background:
- Protein structure prediction is crucial for understanding function and designing new proteins.
- Existing methods often rely on templates or de novo approaches, each with limitations.
Purpose of the Study:
- To describe a systematic, knowledge-based protein modeling technique.
- To demonstrate its applicability in drug design, vaccine development, and novel protein creation.
- To address both homology modeling and ab initio protein design.
Main Methods:
- Utilizes a relational database of known protein 3D structures.
- Employs simultaneous alignment of tertiary structures and sequence homology for fragment selection.
- Integrates loop modeling, side-chain substitution, and energy minimization.
Main Results:
- Successfully modeled homologous structures like tissue plasminogen activator.
- Modeled analogous proteins such as HIV viral proteinase.
- Addressed the inverse problem of designing amino acid sequences for specific tertiary structures.
Conclusions:
- The described protein modeling technique is versatile and applicable to various design challenges.
- It integrates multiple structural and sequence information sources for accurate model generation.
- This approach advances the fields of rational drug design and protein engineering.