Related Experiment Video
Updated: Jul 4, 2026

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
Published on: October 28, 2018
Structural Outlier Detection and Zernike-Canterakis Moments for Molecular Surface Meshes-Fast Implementation in
1Department of Bioinformatics and Telemedicine, Faculty of Medicine, Jagiellonian University Medical College, Medyczna 7, 30-688 Kraków, Poland.
A new Python library implements the Pozo-Koehl algorithm for fast 3D protein shape analysis. This enables efficient protein structure retrieval by encoding molecular shapes, improving upon existing methods.
Area of Science:
- Structural Bioinformatics
- Computational Biology
- Geometric Deep Learning
Background:
- Protein Data Bank (PDB) growth necessitates efficient 3D model retrieval.
- Current methods for protein shape comparison face scalability challenges.
- Zernike-Canterakis moments are effective for shape encoding but lack accessible Python libraries for local computation.
Purpose of the Study:
- To present a fast, well-documented Python implementation of the Pozo-Koehl (PK) algorithm for Zernike-Canterakis moment calculation.
- To introduce a novel protein structure retrieval pipeline utilizing the PK algorithm.
- To enhance the efficiency and accuracy of 3D molecular shape analysis and retrieval.
Main Methods:
- Developed a Python library for the Pozo-Koehl algorithm, directly processing triangular surface meshes.
- Implemented Numba's just-in-time compilation for accelerated moment calculations.
- Integrated a Principal Component Analysis (PCA)-based subroutine for eliminating outlying chain fragments.
Main Results:
- The PK-Zernike library processes 50,000 facets per second at moment order 20.
- The novel retrieval pipeline achieved an Area Under the ROC Curve of 0.961 (0.997 for assemblies) in the BioZernike validation suite.
- High correlation (up to 0.99) was observed between the proposed method and the 3D Surfer program.
Conclusions:
- The presented PK-Zernike library offers a significant improvement for local Zernike-Canterakis moment computation in Python.
- The novel retrieval pipeline demonstrates high efficiency and discrimination ability for protein structure similarity searches.
- This work provides a valuable tool for structural bioinformatics, facilitating large-scale analysis of the Protein Data Bank.
Related Concept Videos
Molecular Geometry and Dipole Moments
Molecules with Multiple Chiral Centers
Ziegler–Natta Chain-Growth Polymerization: Overview
Mesh Analysis
A fundamental concept in mesh analysis is the definition of meshes and mesh currents. A mesh is a closed...
Mesh Analysis with Current Sources
Current Source in One Mesh: The analysis process is straightforward when a current source is found in only one mesh within the circuit. Mesh currents are assigned as usual, with the mesh containing the current source excluded from the analysis. Kirchhoff's voltage law (KVL)...
Three-Dimensional Analysis of Strain

