Protein Structure Prediction

Abstract Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous experimental effort 1–4 , the structures of around 100,000 unique proteins have been determined 5 , but this represents a small fraction of the billions of known protein sequences 6,7 . Structural coverage is bottlenecked by the months to years of painstaking effort required to determine a single protein structure. Accurate computational approaches are needed to address this gap and to enable large-scale structural bioinformatics. Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence—the structure prediction component of the ‘protein folding problem’ 8 —has been an important open research problem for more than 50 years 9 . Despite recent progress 10–14 , existing methods fall far short of atomic accuracy, especially when no homologous structure is available. Here we provide the first computational method that can regularly predict protein structures with atomic accuracy even in cases in which no similar structure is known. We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Critical Assessment of protein Structure Prediction (CASP14) 15 , demonstrating accuracy competitive with experimental structures in a majority of cases and greatly outperforming other methods. Underpinning the latest version of AlphaFold is a novel machine learning approach that incorporates physical and biological knowledge about protein structure, leveraging multi-sequence alignments, into the design of the deep learning algorithm.

Abstract AlphaFold-Multimer has emerged as the state-of-the-art tool for predicting the quaternary structure of protein complexes (assemblies or multimers) since its release in 2021. To further enhance the AlphaFold-Multimer-based complex structure prediction, we developed a new quaternary structure prediction system (MULTICOM) to improve the input fed to AlphaFold-Multimer and evaluate and refine the outputs generated by AlphaFold2-Multimer. Specifically, MULTICOM samples diverse multiple sequence alignments (MSAs) and templates for AlphaFold-Multimer to generate structural models by using both traditional sequence alignments and new Foldseek-based structure alignments, ranks structural models through multiple complementary metrics, and refines the structural models via a Foldseek structure alignment-based refinement method. The MULTICOM system with different implementations was blindly tested in the assembly structure prediction in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022 as both server and human predictors. Our server (MULTICOM_qa) ranked 3rd among 26 CASP15 server predictors and our human predictor (MULTICOM_human) ranked 7th among 87 CASP15 server and human predictors. The average TM-score of the first models predicted by MULTICOM_qa for CASP15 assembly targets is ~0.76, 5.3% higher than ~0.72 of the standard AlphaFold-Multimer. The average TM-score of the best of top 5 models predicted by MULTICOM_qa is ~0.80, about 8% higher than ~0.74 of the standard AlphaFold-Multimer. Moreover, the novel Foldseek Structure Alignment-based Model Generation (FSAMG) method based on AlphaFold-Multimer outperforms the widely used sequence alignment-based model generation.

In recent years, advances in artificial intelligence (AI) have transformed structural biology, particularly protein structure prediction. Though AI-based methods, such as AlphaFold (AF), often predict single conformations of proteins with high accuracy and confidence, predictions of alternative folds are often inaccurate, low-confidence, or simply not predicted at all. Here, we review three blind spots that alternative conformations reveal about AF-based protein structure prediction. First, proteins that assume conformations distinct from their training-set homologs can be mispredicted. Second, AF overrelies on its training set to predict alternative conformations. Third, degeneracies in pairwise representations can lead to high-confidence predictions inconsistent with experiment. These weaknesses suggest approaches to predict alternative folds more reliably. • AlphaFold-based methods often predict single protein structures with high accuracy. • However, these methods sometimes fail to predict alternative conformations. • Three explanations for these failures are discussed.

The structure of a protein is of great importance in determining its functionality, and this characteristic can be leveraged to train data-driven prediction models. However, the limited number of available protein structures severely limits the performance of these models. AlphaFold2 and its open-source data set of predicted protein structures have provided a promising solution to this problem, and these predicted structures are expected to benefit the model performance by increasing the number of training samples. In this work, we constructed a new data set that acted as a benchmark and implemented a state-of-the-art structure-based approach for determining whether the performance of the function prediction model can be improved by putting additional AlphaFold-predicted structures into the training set and further compared the performance differences between two models separately trained with real structures only and AlphaFold-predicted structures only. Experimental results indicated that structure-based protein function prediction models could benefit from virtual training data consisting of AlphaFold-predicted structures. First, model performances were improved in all three categories of Gene Ontology terms (GO terms) after adding predicted structures as training samples. Second, the model trained only on AlphaFold-predicted virtual samples achieved comparable performances to the model based on experimentally solved real structures, suggesting that predicted structures were almost equally effective in predicting protein functionality.

This paper aims to provide a comprehensive review of the trends and challenges of deep neural networks for protein secondary structure prediction (PSSP). In recent years, deep neural networks have become the primary method for protein secondary structure prediction. Previous studies showed that deep neural networks had uplifted the accuracy of three-state secondary structure prediction to more than 80%. Favored deep learning methods, such as convolutional neural networks, recurrent neural networks, inception networks, and graph neural networks, have been implemented in protein secondary structure prediction. Methods adapted from natural language processing (NLP) and computer vision are also employed, including attention mechanism, ResNet, and U-shape networks. In the post-AlphaFold era, PSSP studies focus on different objectives, such as enhancing the quality of evolutionary information and exploiting protein language models as the PSSP input. The recent trend to utilize pre-trained language models as input features for secondary structure prediction provides a new direction for PSSP studies. Moreover, the state-of-the-art accuracy achieved by previous PSSP models is still below its theoretical limit. There are still rooms for improvement to be made in the field.

AlphaFold can predict the structures of monomeric and multimeric proteins with high accuracy but has a limit on the number of chains and residues it can fold. Here we show that a combination of AlphaFold and all-atom symmetric docking simulations enables highly accurate prediction of the structure of complex symmetrical assemblies. We present a method to predict the structure of complexes with cubic - tetrahedral, octahedral and icosahedral - symmetry from sequence. Focusing on proteins where AlphaFold can make confident predictions on the subunit structure, 27 cubic systems were assembled with a median TM-score of 0.99 and a DockQ score of 0.72. 21 had TM-scores of above 0.9 and were categorized as acceptable- to high-quality according to DockQ. The resulting models are energetically optimized and can be used for detailed studies of intermolecular interactions in higher-order symmetrical assemblies. The results demonstrate how explicit treatment of structural symmetry can significantly expand the size and complexity of AlphaFold predictions.

Abstract Predicting structures accurately for natural protein sequences by DeepMind’s AlphaFold is certainly one of the greatest breakthroughs in biology in the twenty-first century. For designed or engineered sequences, which can be unstable, predicting the stabilities together with their structures is essential since unstable structures will not function properly. We found that experimentally measured stability changes of point mutations correlate poorly with the confidence scores produced by AlphaFold. However, the stability changes can be accurately predicted using features extracted from the representations learned by AlphaFold, indicating greater generalizability of AlphaFold to designed or engineered sequences than previously thought. We then used AlphaFold to validate our previously developed protein design method, ProDCoNN, that designs sequences to fold to target protein structures given only the backbone structure information of the target proteins. We showed that ProDCoNN was able to design sequences that fold to structures very close to target structures. By combining a modified ProDCoNN, AlphaFold, and sequential Monte Carlo, we designed a novel framework to estimate the designability of protein structures. The designability of a protein structure is defined as the number of sequences, which encode the protein structure, and is an indicator of the functional robustness of proteins. For the first time, we estimated the designability of a real protein structure, chain A of FLT3 ligand (PDB ID: 1ETE) with 134 residues, as 3.12±2.14E85.

Models of protein structures enable molecular understanding of biological processes. Current protein structure prediction tools lie at the interface of biology, chemistry and computer science. Millions of protein structure models have been generated in a very short space of time through a revolution in protein structure prediction driven by deep learning, led by AlphaFold. This has provided a wealth of new structural information. Interpreting these predictions is critical to determining where and when this information is useful. But proteins are not static nor do they act alone, and structures of proteins interacting with other proteins and other biomolecules are critical to a complete understanding of their biological function at the molecular level. This review focuses on the application of state-of-the-art protein structure prediction to these advanced applications. We also suggest a set of guidelines for reporting AlphaFold predictions.

The introduction of AlphaFold2 1 , its subsequent AlphaFold3 version 2 , and the public release of hundreds of millions of protein structures predicted by this landmark method 3,4 have fundamentally changed the fields of structural biology and protein bioinformatics.Structural biology, which once relied on painstaking and relatively slow experimental determination of protein structure, has become one of abundance, with structural models now available for nearly every known protein sequence.Protein bioinformatics has evolved from developing structure prediction tools to analyzing and applying structural models.This special issue of the Computational and Structural Biotechnology Journal commemorates these transitions by exploring the many diverse ways in which AlphaFold is used.We initiated this special issue in July of 2024 and were flooded with submissions.We extended the original submission deadline from October 31, 2024, to January 31, 2025, to accommodate this strong interest.The resulting special issue includes 32 articles that collectively define the current state of structural protein bioinformatics.We are deeply grateful to the authors who entrusted us with their studies, the reviewers who provided their timely and insightful advice, and the editorial team for their help with this process and encouragement to organize this special issue.Following, we provide a brief overview of our special issue.

This protocol describes the protein structure prediction with AlphaFold-Multimer.