RosettaDesign

RosettaDesign designs low-energy amino acid sequences compatible with specified protein three-dimensional structures to enable computational protein design and engineering.

Key Features:

• Novel Protein Design: Facilitates design of entirely new protein folds, exemplified by Top7, a 93-residue alpha/beta protein whose x-ray crystal structure matched the computational model with a root mean square deviation of 1.2 angstroms.
• Rational Redesign of Folding Pathways: Alters protein folding pathways by optimizing backbone conformations and amino acid sequences, exemplified by a redesigned protein G variant that folds approximately 100 times faster than wild type.
• Stabilization of Protein Structures: Stabilizes domain-swapped dimers such as protein L through strategic mutations that produce highly stable complexes with dissociation constants comparable to naturally occurring protein dimers.
• Exploration of Sequence Space: Employs a Monte Carlo optimization procedure and a free energy function based on Lennard-Jones packing interactions and the Lazaridis-Karplus implicit solvation model to characterize the sequence space compatible with a given structure.
• Synthesis and Experimental Validation: Predicts low free energy sequences for natural backbones that have been synthesized as genes, expressed, and characterized, with many redesigned proteins folding and exhibiting stability comparable or superior to wild type.

Scientific Applications:

• Protein Engineering: Enables design and redesign of proteins with desired structural and stability properties for industrial, medical, and research applications.
• Understanding Protein Folding: Provides a platform for probing and redesigning folding pathways to study folding mechanisms.
• Stabilization Studies: Supports stabilization of protein structures and protein–protein interactions relevant to therapeutic protein design.

Methodology:

Iterative sequence design coupled with structure prediction using Monte Carlo optimization and a free energy function incorporating Lennard-Jones packing interactions and the Lazaridis-Karplus implicit solvation model; low-energy sequences can be generated from random starting points.