Artificial intelligence-based protein structure prediction approaches have had a transformative effect on biomolecular sciences. The predicted protein models in the AlphaFold protein structure database, however, all lack coordinates for small molecules, essential for molecular structure or function: hemoglobin lacks bound heme; zinc-finger motifs lack zinc ions essential for structural integrity and metalloproteases lack metal ions needed for catalysis. Ligands important for biological function are absent too; no ADP or ATP is bound to any of the ATPases or kinases. Here we present AlphaFill, an algorithm that uses sequence and structure similarity to ‘transplant’ such ‘missing’ small molecules and ions from experimentally determined structures to predicted protein models. The algorithm was successfully validated against experimental structures. A total of 12,029,789 transplants were performed on 995,411 AlphaFold models and are available together with associated validation metrics in the alphafill.eu databank, a resource to help scientists make new hypotheses and design targeted experiments.
Artificial intelligence (AI) methods for constructing structural models of proteins on the basis of their sequence are having a transformative effect in biomolecular sciences. The AlphaFold protein structure database makes available hundreds of thousands of protein structures. However, all these structures lack cofactors essential for their structural integrity and molecular function (e.g. hemoglobin lacks a bound heme), key ions essential for structural integrity (e.g. zinc-finger motifs) or catalysis (e.g. Ca2+ or Zn2+ in metalloproteases), and ligands that are important for biological function (e.g. kinase structures lack ADP or ATP). Here, we present AlphaFill, an algorithm based on sequence and structure similarity, to “transplant” such “missing” small molecules and ions from experimentally determined structures to predicted protein models. These publicly available structural annotations are mapped to predicted protein models, to help scientists interpret biological function and design experiments.
The quality of macromolecular structure models crucially depends on refinement and validation targets, which optimally describe the expected chemistry. Commonly used software for these two procedures has been designed and developed in a protein-centric manner, resulting in relatively few established features for the refinement and validation of nucleic acid-containing structure models. Here, new nucleic acid-specific approaches implemented in PDB-REDO are described, including a new restraint model using noncovalent geometries (base-pair hydrogen bonding and base-pair stacking) as refinement targets. New validation routines are also presented, including a metric for Watson–Crick base-pair geometry normality (Z bpG). Applying the PDB-REDO pipeline with the new restraint model to the whole Protein Data Bank (PDB) demonstrates an overall positive effect on the quality of nucleic acid-containing structure models. Finally, we discuss examples of improvements in the geometry of specific nucleic acid structures in the PDB. The new PDB-REDO models and pipeline are available at https://pdb-redo.eu/.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.