An official website of the United States government
Abstract Deep learning approaches like AlphaFold 2 (AF2) have revolutionized structural biology by accurately predicting the ground state structures of proteins. Recently, clustering and subsampling techniques that manipulate multiple sequence alignment (MSA) inputs into AlphaFold to generate conformational ensembles of proteins have also been proposed. Although many of these techniques have been made open source, they often require integrating multiple packages and can be challenging for researchers who have a limited programming background to employ. This is especially true when researchers are interested in subsampling to produce predictions of protein conformational ensembles, which require multiple computational steps. This manuscript introduces FastConformation, a Python-based application that integrates MSA generation, structure prediction via AF2, and interactive analysis of protein conformations and their distributions, all in one place. FastConformation is accessible through a user-friendly GUI suitable for non-programmers, allowing users to iteratively refine subsampling parameters based on their analyses to achieve diverse conformational ensembles. Starting from an amino acid sequence, users can make protein conformation predictions and analyze results in just a few hours on their local machines, which is significantly faster than traditional molecular dynamics (MD) simulations. Uniquely, by leveraging the subsampling of MSAs, our tool enables the generation of alternative protein conformations. We demonstrate the utility of FastConformation on proteins including the Abl1 kinase, LAT1 transporter, and CCR5 receptor, showcasing its ability to predict and analyze the protein conformational ensembles and effects of mutations on a variety of proteins. This tool enables a wide range of high-throughput applications in protein biochemistry, drug discovery, and protein engineering.
more »
« less
Predicting protein conformational motions using energetic frustration analysis and AlphaFold2
Proteins perform their biological functions through motion. Although high throughput prediction of the three-dimensional static structures of proteins has proved feasible using deep-learning-based methods, predicting the conformational motions remains a challenge. Purely data-driven machine learning methods encounter difficulty for addressing such motions because available laboratory data on conformational motions are still limited. In this work, we develop a method for generating protein allosteric motions by integrating physical energy landscape information into deep-learning-based methods. We show that local energetic frustration, which represents a quantification of the local features of the energy landscape governing protein allosteric dynamics, can be utilized to empower AlphaFold2 (AF2) to predict protein conformational motions. Starting from ground state static structures, this integrative method generates alternative structures as well as pathways of protein conformational motions, using a progressive enhancement of the energetic frustration features in the input multiple sequence alignment sequences. For a model protein adenylate kinase, we show that the generated conformational motions are consistent with available experimental and molecular dynamics simulation data. Applying the method to another two proteins KaiB and ribose-binding protein, which involve large-amplitude conformational changes, can also successfully generate the alternative conformations. We also show how to extract overall features of the AF2 energy landscape topography, which has been considered by many to be black box. Incorporating physical knowledge into deep-learning-based structure prediction algorithms provides a useful strategy to address the challenges of dynamic structure prediction of allosteric proteins.
more »
« less
- Award ID(s):
- 2019745
- PAR ID:
- 10595427
- Publisher / Repository:
- National Academy of Sciences
- Date Published:
- Journal Name:
- Proceedings of the National Academy of Sciences
- Volume:
- 121
- Issue:
- 35
- ISSN:
- 0027-8424
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Protein function often depends on dynamic transitions between conformations rather than just static structures. However, our current ability to characterize or predict such dynamics lags behind recent advances in protein structure prediction. Enhanced sampling methods can speed up molecular dynamics simulations to study protein conformational transitions but require prior knowledge of key collective motions involved. Here, we demonstrate for a series of proteins of varying complexity that the required information is encoded in anharmonic low-frequency vibrations. Using recently developed methods, we show that this information can be easily extracted from short dynamics simulations without requiring prior knowledge. Combined with enhanced sampling, we correctly predict conformational transitions in all test proteins and generate highly reproducible free energy landscapes. This allows for the rapid generation of accurate protein conformational ensembles, which is critical to unravel the complex relationship between protein sequence, structure, and dynamics.more » « less
-
Given proteins' fundamental importance in human health and catalysis, the relationships between protein sequence, structure, dynamics, and function have become a topic of great interest. One way to extract information from proteins is to compute the local energetic frustration of their native state. Traditionally, energetic frustration calculations require protein structures as a starting point. However, using a single protein structure to evaluate the energetic frustration for a given amino acid sequence does not always fully represent the protein's structural ensemble. Therefore, we have developed a sequence-based method to evaluate energetic frustration in proteins using direct coupling analysis and statistical potentials. Our approach exhibits significant agreement with established structure-based frustration methods in terms of their mutual agreement with crystallographic B-factor. Moreover, our sequence-based method shows elevated precision in classifying high B-factor residues, suggesting that it has some robustness to unstructured regions of proteins.more » « less
-
Abstract While significant advances have been made in predicting static protein structures, the inherent dynamics of proteins, modulated by ligands, are crucial for understanding protein function and facilitating drug discovery. Traditional docking methods, frequently used in studying protein-ligand interactions, typically treat proteins as rigid. While molecular dynamics simulations can propose appropriate protein conformations, they’re computationally demanding due to rare transitions between biologically relevant equilibrium states. In this study, we present DynamicBind, a deep learning method that employs equivariant geometric diffusion networks to construct a smooth energy landscape, promoting efficient transitions between different equilibrium states. DynamicBind accurately recovers ligand-specific conformations from unbound protein structures without the need for holo-structures or extensive sampling. Remarkably, it demonstrates state-of-the-art performance in docking and virtual screening benchmarks. Our experiments reveal that DynamicBind can accommodate a wide range of large protein conformational changes and identify cryptic pockets in unseen protein targets. As a result, DynamicBind shows potential in accelerating the development of small molecules for previously undruggable targets and expanding the horizons of computational drug discovery.more » « less
-
The goal of this paper is predicting the conformational distributions of ligand binding sites using the AlphaFold2 (AF2) protein structure prediction program with stochastic subsampling of the multiple sequence alignment (MSA). We explored the opening of cryptic ligand binding sites in 16 proteins, where the closed and open conformations define the expected extreme points of the conformational variation. Due to the many structures of these proteins in the Protein Data Bank (PDB), we were able to study whether the distribution of X-ray structures affects the distribution of AF2 models. We have found that AF2 generates both a cluster of open and a cluster of closed models for proteins that have comparable numbers of open and closed structures in the PDB and not too many other conformations. This was observed even with default MSA parameters, thus without further subsampling. In contrast, with the exception of a single protein, AF2 did not yield multiple clusters of conformations for proteins that had imbalanced numbers of open and closed structures in the PDB, or had substantial numbers of other structures. Subsampling improved the results only for a single protein, but very shallow MSA led to incorrect structures. The ability of generating both open and closed conformations for six out of the 16 proteins agrees with the success rates of similar studies reported in the literature. However, we showed that this partial success is due to AF2 “remembering” the conformational distributions in the PDB and that the approach fails to predict rarely seen conformations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1073/pnas.2410662121
