Critical assessment of methods of protein structure prediction (CASP)—Round XIV

Abstract

Critical assessment of structure prediction (CASP) is a community experiment to advance methods of computing three-dimensional protein structure from amino acid sequence. Core components are rigorous blind testing of methods and evaluation of the results by independent assessors. In the most recent experiment (CASP14), deeplearning methods from one research group consistently delivered computed structures rivaling the corresponding experimental ones in accuracy. In this sense, the results represent a solution to the classical protein-folding problem, at least for single proteins. The models have already been shown to be capable of providing solutions for problematic crystal structures, and there are broad implications for the rest of structural biology. Other research groups also substantially improved performance.Here, we describe these results and outline some of the many implications. Other related areas of CASP, including modeling of protein complexes, structure refinement, estimation of model accuracy, and prediction of inter-residue contacts and distances, are also described.

Intrinsic Protein Disorder, Conditional Folding and AlphaFold2

“…Since 1994, the biennial Critical Assessment of techniques for protein Structure Prediction (CASP) experiment has tried to measure the state-of-the-art and progress in the field. In CASP14 1 AlphaFold has at last demonstrated a breakthrough thanks to its clever use of machine learning and multiple sequence alignments 2,3 . This is leading to a paradigm shift for structural biology due to the sudden availability of orders of magnitude more protein structures 4,5 .…”

Harnessing protein folding neural networks for peptide–protein docking

“…The advances in the field of protein structure prediction in recent years open up exciting opportunities to fully leverage such information. The development and application of deep learning (DL) neural network (NN) architectures to predict monomeric protein structures provided us with highly accurate computational models as particularly showcased by the last CASP14 experiment 27 . AlphaFold2 (AF2) developed by Google Deepmind was able to generate models of exceptional accuracy, approaching the resolution of crystallography experiments 28 .…”

“…We note that both RoseTTAFold and AF2 NNs were trained on single chain protein structural data, and both use Multiple Sequence Alignments (MSA) as a critical step in structure prediction. Prediction of protein-protein complexes was shown to be possible given an informative MSA 27 , 29 , 32 , and it has also been explored whether it is indeed necessary to provide paired sequences for successful extraction of interface information 33 , 34 . As both methods heavily rely on good quality MSA, the main challenge would be to accurately predict the peptide conformation.…”

Rethinking protein drug design with highly accurate structure prediction of anti-CRISPR proteins

“…We then used the following relative Z-error to calculate the relative difference: where Z gt is the self Dali Z-score between experimental structure and itself, and Z pd is the Dali Z-score between experimental structure and predicted structure. We obtained the Z pd and TM-score for the CASP14 set of AlphaFold from the CASP14 assessment scores 37 , whereas Z gt of CASP14 and Z gt and Z pd of Set A and Set C were calculated using DaliLite.v5 36 . Finally, a protein structure comparison and clustering tool called MaxCluster 38 was used to calculate the TM-scores of Set A and Set C. Both distance metrics have values between 0 and 1, with 1 as the best score for TM-scores and 0 as the best score for relative Z-errors.…”