Answers are grounded in this paper and cite their sources.
Structural analysis is fundamental to understanding the biological function of proteins. Among structural descriptors, atomic depth plays a key role in identifying folding motifs, assessing stability, and characterising molecular interactions. A prior method for computing atomic depth, implemented in the Simple Atom Depth Index Calculator (SADIC), leveraged Discrete Geometry to capture local molecular shape. In this work, we present a novel approach that integrates Discrete Geometry and Mathematical Morphology to compute volume-based atom depth indices using a digital-space representation of macromolecular structures. A core feature of the method is a morphological strategy for detecting solvent-inaccessible cavities, enabling their inclusion in volume computation without altering the protein's overall three-dimensional structure. The approach supports solvent-specific probing of cavity accessibility and provides a balanced characterisation of deeply buried atoms while preserving the high variability of superficial atom depths, a hallmark of volume-based descriptors. Computational efficiency is further enhanced by introducing a translated reference sphere for depth calculation. We evaluate the method on a dataset of 1000 proteins with diverse sizes and shapes, demonstrating greater effectiveness, reduced estimated protein volume, and significantly faster execution times compared to existing approaches. Additionally, the depth indices agree closely with those computed by the DEPTH algorithm, but offer a finer-grained characterisation of superficial atoms.
No takes yet. Share an insight, caveat, or question.
Nunziati et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: