PROTHON

PROTHON compares protein conformational ensembles using local order parameters and probability distribution functions (pdfs), quantifying ensemble dissimilarity with the Jensen-Shannon distance for structural analysis.

Key Features:

• Efficient Representation: Represents ensembles as vectors of probability distribution functions (pdfs) that capture structural properties such as contact numbers between Cβ atoms.
• Quantitative Comparison: Quantifies ensemble differences by computing Jensen-Shannon distances between the probability distributions.
• Performance Optimization: Demonstrates up to 88-fold speed improvement and requires 48-fold fewer computing cores compared to ENCORE, enabling large-scale analyses.
• Validation and Application: Validated on ubiquitin and human tau protein data and applied to studies relevant to drug discovery and disease mechanisms.

Scientific Applications:

• Drug Discovery: Facilitates rapid comparison of large protein ensembles to identify conformational states relevant to ligand binding and target characterization.
• Disease Mechanism Studies: Enables analysis of conformational changes in disease-associated proteins such as human tau in Alzheimer's disease.

Methodology:

Represents ensembles as vectors of probability distribution functions of local order parameters (e.g., Cβ contact-number distributions) and computes Jensen-Shannon distances between these distributions to quantify ensemble dissimilarity.