Crysalis

Crysalis predicts and redesigns protein sites to improve crystallization propensity for X-ray crystallography by identifying site non-optimality and proposing single-point mutations.

Key Features:

• Proteome-Level Selection: Enables identification and prioritization of proteins with higher potential for crystallization across entire proteomes.
• Identification of Site Non-Optimality: Systematically detects specific sites within protein structures where non-optimal conditions may hinder crystallization.
• Systematic Mutation Analysis: Evaluates and ranks potential single-point mutations aimed at enhancing protein crystallizability.
• Structural Property Annotation: Annotates targets with predicted structural properties and physicochemical features that influence crystallization outcomes.

Scientific Applications:

• Proteome-scale analysis of non-crystallizable proteins: Applied to proteome-scale datasets to reveal that site non-optimality is influenced by biases in residues, predicted structures, physicochemical properties, and sequence loci.
• Design of crystallization strategies and mutant selection: Provides insights that aid the design of targeted mutations and prioritization strategies to improve the likelihood of obtaining diffraction-quality crystals for X-ray crystallography.

Methodology:

Crysalis integrates support-vector regression (SVR) models with bioinformatics approaches to predict crystallization propensity, identify site non-optimality, and assess potential single-point mutation effects.