Over Half a Century of Burial of ρ, θ and ϕ in PDB

Since its inception in 1971, the Protein Data Bank (PDB) has archived experimentally measured biomolecular structures in Cartesian coordinates \(\lbrack x,y,z\rbrack\). While this convention aligns with the output of experimental techniques such as X-ray crystallography and NMR spectroscopy, it does not explicitly encode chemical connectivity or atomic bonding networks within protein structures. In light of the geometric equivalence of Cartesian (CCS) and spherical (SCS) coordinate systems, this manuscript proposes an alternative geometric framework—the Atomic Bonding Network-based Spherical Coordinate System (ABN-SCS)—in which atomic positions are expressed using spherical coordinates \(\lbrack\rho,\theta,\phi\rbrack\) anchored to the covalent atomic bonding network within a protein structure. In ABN-SCS, \(\rho\) corresponds to the equilibrium covalent bond length, while \(\theta\) and \(\phi\) capture angular distributions, providing chemically meaningful descriptors that complement Cartesian-based representations. With Caenopore-5 as an example, this manuscript demonstrates that ABN-SCS enables the characterization of spherical bond-level geometries, expanding the feature space available for computational pipelines such as AlphaFold2. Finally, this work argues that integrating ABN-SCS features into protein structure prediction pipelines can enhance geometric fidelity and that the time is now ripe for the community to consider the other side of the coin (Ref: Graphical Abstract) and for the trapped spherical features [\(\rho\), \(\theta\), \(\phi\)] to be released from the PDB and integrated into algorithms such as AF2 to improve protein structure prediction performance.