Characterizing and explaining the impact of disease-associated mutations in proteins without known structures or structural homologs

Abstract

Mutations in human proteins lead to diseases. The structure of these proteins can help understand the mechanism of such diseases and develop therapeutics against them. With improved deep learning techniques, such as RoseTTAFold and AlphaFold, we can predict the structure of proteins even in the absence of structural homologs. We modeled and extracted the domains from 553 disease-associated human proteins without known protein structures or close homologs in the Protein Databank. We noticed that the model quality was higher and the Root mean square deviation (RMSD) lower between AlphaFold and RoseTTAFold models for domains that could be assigned to CATH families as compared to those which could only be assigned to Pfam families of unknown structure or could not be assigned to either. We predicted ligand-binding sites, protein–protein interfaces and conserved residues in these predicted structures. We then explored whether the disease-associated missense mutations were in the proximity of these predicted functional sites, whether they destabilized the protein structure based on ddG calculations or whether they were predicted to be pathogenic. We could explain 80% of these disease-associated mutations based on proximity to functional sites, structural destabilization or pathogenicity. When compared to polymorphisms, a larger percentage of disease-associated missense mutations were buried, closer to predicted functional sites, predicted as destabilizing and pathogenic. Usage of models from the two state-of-the-art techniques provide better confidence in our predictions, and we explain 93 additional mutations based on RoseTTAFold models which could not be explained based solely on AlphaFold models.

Digging into the 3D Structure Predictions of AlphaFold2 with Low Confidence: Disorder and Beyond

“…First, the proposed 3D structures should be adopted but are not yet validated by AF2 due to either: original folds/structures, the lack of representation in the databases used for learning, or an insufficient amount of homologous sequences to validate the predicted contacts. This hypothesis was recently supported in particular by Sen and colleagues [ 62 ], showing lower AF2 pLDDT values for models of sequences corresponding to unassigned domains, compared to those corresponding to CATH or Pfam entries.…”

Conformational Stability and Denaturation Processes of Proteins Investigated by Electrophoresis under Extreme Conditions

“…The recent release of the AlphaFold2 AI program for protein structure prediction [ 58 , 156 ] is widely seen as a breakthrough in the field of protein structure prediction [ 157 , 158 ]. Though its ability to predict the structural and functional effects of mutations is being debated [ 159 , 160 , 161 ], AlphaFold2-based approaches are already being developed [ 162 ].…”

AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor

“…13 Although the protein models predicted by AlphaFold have variable qualities from good, bad to ugly, 10 the predicted local distance difference test score is provided as a confidence metric to guide the usage of 3D structures produced by AlphaFold. AlphaFold models have been used to aid the determination of experimental structures by crystallography 14 and cryo-EM, 15 to guide the functional study of PINK1, 16 to help identify pathogenic mutations, 17,18 and to explore the protein–protein interaction. 19 Public databases include AlphaFold models as references, e.g.…”