In recent years, the number of known protein structures has increased significantly. Predictive algorithms and experimental methods provide the positions of protein residues relative to each other with high accuracy. However, the local quality of the protein structure, including bond lengths, angles, and positions of individual atoms, often lacks the same level of precision. For this reason, protein structures are usually optimised by a force field prior to their application in further research sensitive to structural quality. Protein structure optimisation, however, is computationally challenging. In this paper, we introduce a general method Per-residue optimisation of protein structures: Rapid alternative to optimisation with constrained alpha carbons (PROPTIMUS RAPHAN). Rather than optimising the entire protein structure at once, PROPTIMUS RAPHAN divides the structure into overlapping residual substructures and optimises each substructure individually. This approach results in computational time that scales linearly with the size of the structure. Additionally, we present PROPTIMUS RAPHAN GFN-FF , a reference implementation of our method employing a generic, almost QM-accurate force field, GFN-FF. We tested PROPTIMUS RAPHAN GFN-FF on 461 AlphaFold DB structures and demonstrated that our approach achieves results comparable to the optimisation of the structure with constrained alpha carbons in significantly less time.
High-quality protein structures are essential for a wide range of computational chemistry applications. While experimental methods and predictive algorithms provide high-accuracy positions of protein residues relative to one another, the local quality of these structures, including bond lengths, angles, and individual atom positions, often lacks the same level of precision. To address this, we developed PROPTIMUS LIVE, a web application offering local constrained α-carbon optimization of protein structures. PROPTIMUS LIVE is powered by the QM-accurate, physics-based GFN-Force-Field and accelerated using a divide-and-conquer approach, allowing typical optimizations to finish within minutes. Optimized structures can be visualized and investigated directly via the integrated Mol* Viewer. The service is freely available at https://proptimus.ceitec.cz/live with no login required, including for commercial use.
Protein structural data are highly valuable for research, and many significant results have been published on their basis. A key point for their credibility and applicability is their quality. An important facet of protein structure quality is the validation of ligands. Some aspects of ligand quality have already been validated by established quality metrics. However, validation of ring conformation has yet to be comprehensively performed despite rings strongly influencing the formation of the ligand’s scaffold and shape. Most rings form several conformations that differ in their stability. The most stable ones occur frequently in nature and should, therefore, be found in Protein Data Bank (PDB) structures. In this article, we examined which conformations of rings occur in PDB structures. Our analysis focused on conformations of all cyclopentane, cyclohexane, and benzene rings in the PDB. Specifically, we examined 123 264 rings of 24 763 distinct ligands, which instances occur in 44 022 protein structures. In general, we found that most of the rings (98.32 %) are in energetically favourable conformations. Surprisingly, the existence of most of the energetically unfavourable ring conformations (2 067 samples, 1.68 %) is not supported by experimental data. Only 291 unfavourable ring conformations (0.24 %) are backed by experimental data that are accurate enough to distinguish the conformation, which shows that the existence of energetically unfavourable ring conformations is rarely supported by structural or experimental evidence. Our results suggest that each occurrence of untypical ring conformation in the PDB may indicate a potential error and should be carefully analysed.