PyTorchKR Papers

최신 논문

총 174편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2110.04624

Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design

Antibodies are versatile proteins that bind to pathogens like viruses and stimulate the adaptive immune system. The specificity of antibody binding is determined by complementarity-determining regions (CDRs) at the tips of these Y-shaped proteins. In this paper, we propose a generative model to automatically design the CDRs of antibodies with enhanced binding specificity or neutralization capabilities. Previous generative approaches formulate protein design as a structure-conditioned sequence generation task, assuming the desired 3D structure is given a priori. In contrast, we propose to co-design the sequence and 3D structure of CDRs as graphs. Our model unravels a sequence autoregressively while iteratively refining its predicted global structure. The inferred structure in turn guides subsequent residue choices. For efficiency, we model the conditional dependence between residues inside and outside of a CDR in a coarse-grained manner. Our method achieves superior log-likelihood on the test set and outperforms previous baselines in designing antibodies capable of neutralizing the SARS-CoV-2 virus.

arXiv:2109.13754

Deep Generative Modeling for Protein Design

Deep learning approaches have produced substantial breakthroughs in fields such as image classification and natural language processing and are making rapid inroads in the area of protein design. Many generative models of proteins have been developed that encompass all known protein sequences, model specific protein families, or extrapolate the dynamics of individual proteins. Those generative models can learn protein representations that are often more informative of protein structure and function than hand-engineered features. Furthermore, they can be used to quickly propose millions of novel proteins that resemble the native counterparts in terms of expression level, stability, or other attributes. The protein design process can further be guided by discriminative oracles to select candidates with the highest probability of having the desired properties. In this review, we discuss five classes of generative models that have been most successful at modeling proteins and provide a framework for model guided protein design.

arXiv:2109.07925

PDBench: Evaluating Computational Methods for Protein Sequence Design

Proteins perform critical processes in all living systems: converting solar energy into chemical energy, replicating DNA, as the basis of highly performant materials, sensing and much more. While an incredible range of functionality has been sampled in nature, it accounts for a tiny fraction of the possible protein universe. If we could tap into this pool of unexplored protein structures, we could search for novel proteins with useful properties that we could apply to tackle the environmental and medical challenges facing humanity. This is the purpose of protein design. Sequence design is an important aspect of protein design, and many successful methods to do this have been developed. Recently, deep-learning methods that frame it as a classification problem have emerged as a powerful approach. Beyond their reported improvement in performance, their primary advantage over physics-based methods is that the computational burden is shifted from the user to the developers, thereby increasing accessibility to the design method. Despite this trend, the tools for assessment and comparison of such models remain quite generic. The goal of this paper is to both address the timely problem of evaluation and to shine a spotlight, within the Machine Learning community, on specific assessment criteria that will accelerate impact. We present a carefully curated benchmark set of proteins and propose a number of standard tests to assess the performance of deep learning based methods. Our robust benchmark provides biological insight into the behaviour of design methods, which is essential for evaluating their performance and utility. We compare five existing models with two novel models for sequence prediction. Finally, we test the designs produced by these models with AlphaFold2, a state-of-the-art structure-prediction algorithm, to determine if they are likely to fold into the intended 3D shapes.

arXiv:2106.13058공식 코드 · Papers with Code

Fold2Seq: A Joint Sequence(1D)-Fold(3D) Embedding-based Generative Model for Protein Design

Designing novel protein sequences for a desired 3D topological fold is a fundamental yet non-trivial task in protein engineering. Challenges exist due to the complex sequence--fold relationship, as well as the difficulties to capture the diversity of the sequences (therefore structures and functions) within a fold. To overcome these challenges, we propose Fold2Seq, a novel transformer-based generative framework for designing protein sequences conditioned on a specific target fold. To model the complex sequence--structure relationship, Fold2Seq jointly learns a sequence embedding using a transformer and a fold embedding from the density of secondary structural elements in 3D voxels. On test sets with single, high-resolution and complete structure inputs for individual folds, our experiments demonstrate improved or comparable performance of Fold2Seq in terms of speed, coverage, and reliability for sequence design, when compared to existing state-of-the-art methods that include data-driven deep generative models and physics-based RosettaDesign. The unique advantages of fold-based Fold2Seq, in comparison to a structure-based deep model and RosettaDesign, become more evident on three additional real-world challenges originating from low-quality, incomplete, or ambiguous input structures. Source code and data are available at https://github.com/IBM/fold2seq.

논문 ID: 1178.30712

Protein design by multiobjective optimization: evolutionary and non-evolutionary approaches

Traditional simulation-based protein design considers energy minimization of candidate conformations as a singleobjective combinatorial optimization problem. In this paper we consider a challenging protein design problem, producing twelve protein species based on collagen that uniquely assort into four groups of three: a problem defined herein as a 4-level heterotrimer. We formulate a bi-objective combinatorial minimization problem that targets both stability and specificity of the 4-level heterotrimer. In order to approximate its Pareto frontier, we utilize both evolutionary and non-evolutionary approaches, operating in either Pareto or aggregation fashions. Our practical observations suggest that the SMS-EMOA with Evolution Strategies' operators is more effective than standard heuristics deployed in computational protein design, such as Simulated Annealing, Replica Exchange or the Canonical Genetic Algorithm. We investigate the attained Pareto optimal sets using Barrier Tree analysis, aiming to provide insights into the chemical search-space, as well as to explain the observed algorithmic trends. In particular, we identify Replica Exchange as a promising non-evolutionary technique for this problem class, due to its efficient exploration capabilities. Overall, a common high-level protocol for simultaneous landscape analysis of evolutionary and non-evolutionary search methodologies is put forward for the first time.

arXiv:2103.03292

Efficient generative modeling of protein sequences using simple autoregressive models

Generative models emerge as promising candidates for novel sequence-data driven approaches to protein design, and for the extraction of structural and functional information about proteins deeply hidden in rapidly growing sequence databases. Here we propose simple autoregressive models as highly accurate but computationally efficient generative sequence models. We show that they perform similarly to existing approaches based on Boltzmann machines or deep generative models, but at a substantially lower computational cost (by a factor between $10^2$ and $10^3$). Furthermore, the simple structure of our models has distinctive mathematical advantages, which translate into an improved applicability in sequence generation and evaluation. Within these models, we can easily estimate both the probability of a given sequence, and, using the model's entropy, the size of the functional sequence space related to a specific protein family. In the example of response regulators, we find a huge number of ca. $10^{68}$ possible sequences, which nevertheless constitute only the astronomically small fraction $10^{-80}$ of all amino-acid sequences of the same length. These findings illustrate the potential and the difficulty in exploring sequence space via generative sequence models.

arXiv:2102.03881

Mimetic Neural Networks: A unified framework for Protein Design and Folding

Recent advancements in machine learning techniques for protein folding motivate better results in its inverse problem -- protein design. In this work we introduce a new graph mimetic neural network, MimNet, and show that it is possible to build a reversible architecture that solves the structure and design problems in tandem, allowing to improve protein design when the structure is better estimated. We use the ProteinNet data set and show that the state of the art results in protein design can be improved, given recent architectures for protein folding.

arXiv:2012.01736

Designing a Prospective COVID-19 Therapeutic with Reinforcement Learning

The SARS-CoV-2 pandemic has created a global race for a cure. One approach focuses on designing a novel variant of the human angiotensin-converting enzyme 2 (ACE2) that binds more tightly to the SARS-CoV-2 spike protein and diverts it from human cells. Here we formulate a novel protein design framework as a reinforcement learning problem. We generate new designs efficiently through the combination of a fast, biologically-grounded reward function and sequential action-space formulation. The use of Policy Gradients reduces the compute budget needed to reach consistent, high-quality designs by at least an order of magnitude compared to standard methods. Complexes designed by this method have been validated by molecular dynamics simulations, confirming their increased stability. This suggests that combining leading protein design methods with modern deep reinforcement learning is a viable path for discovering a Covid-19 cure and may accelerate design of peptide-based therapeutics for other diseases.

arXiv:1811.00077

Void distributions reveal structural link between jammed packings and protein cores

Dense packing of hydrophobic residues in the cores of globular proteins determines their stability. Recently, we have shown that protein cores possess packing fraction $\phi \approx 0.56$, which is the same as dense, random packing of amino acid-shaped particles. In this article, we compare the structural properties of protein cores and jammed packings of amino acid-shaped particles in much greater depth by measuring their local and connected void regions. We find that the distributions of surface Voronoi cell volumes and local porosities obey similar statistics in both systems. We also measure the probability that accessible, connected void regions percolate as a function of the size of a spherical probe particle and show that both systems possess the same critical probe size. By measuring the critical exponent $\tau$ that characterizes the size distribution of connected void clusters at the onset of percolation, we show that void percolation in packings of amino acid-shaped particles and protein cores belong to the same universality class, which is different from that for void percolation in jammed sphere packings. We propose that the connected void regions of proteins are a defining feature of proteins and can be used to differentiate experimentally observed proteins from decoy structures that are generated using computational protein design software. This work emphasizes that jammed packings of amino acid-shaped particles can serve as structural and mechanical analogs of protein cores, and could therefore be useful in modeling the response of protein cores to cavity-expanding and -reducing mutations.

arXiv:1809.10389

A natural upper bound to the accuracy of predicting protein stability changes upon mutations

Accurate prediction of protein stability changes upon single-site variations (DDG) is important for protein design, as well as our understanding of the mechanism of genetic diseases. The performance of high-throughput computational methods to this end is evaluated mostly based on the Pearson correlation coefficient between predicted and observed data, assuming that the upper bound would be 1 (perfect correlation). However, the performance of these predictors can be limited by the distribution and noise of the experimental data. Here we estimate, for the first time, a theoretical upper-bound to the DDG prediction performances imposed by the intrinsic structure of currently available DDG data. Given a set of measured DDG protein variations, the theoretically best predictor is estimated based on its similarity to another set of experimentally determined DDG values. We investigate the correlation between pairs of measured DDG variations, where one is used as a predictor for the other. We analytically derive an upper bound to the Pearson correlation as a function of the noise and distribution of the DDG data. We also evaluate the available datasets to highlight the effect of the noise in conjunction with DDG distribution. We conclude that the upper bound is a function of both uncertainty and spread of the DDG values, and that with current data the best performance should be between 0.7-0.8, depending on the dataset used; higher Pearson correlations might be indicative of overtraining. It also follows that comparisons of predictors using different datasets are inherently misleading.

arXiv:1809.06632

Modular decomposition of protein structure using community detection

As the number of solved protein structures increases, the opportunities for meta-analysis of this dataset increase too. Protein structures are known to be formed of domains; structural and functional subunits that are often repeated across sets of proteins. These domains generally form compact, globular regions, and are therefore often easily identifiable by inspection, yet the problem of automatically fragmenting the protein into these compact substructures remains computationally challenging. Existing domain classification methods focus on finding subregions of protein structure that are conserved, rather than finding a decomposition which spans the full protein structure. However, such a decomposition would find ready application in coarse-graining molecular dynamics, analysing the protein's topology, in de novo protein design and in fitting electron microscopy maps. Here, we present a tool for performing this modular decomposition using the Infomap community detection algorithm. The protein structure is abstracted into a network in which its amino acids are the nodes, and where the edges are generated using a simple proximity test. Infomap can then be used to identify highly intra-connected regions of the protein. We perform this decomposition systematically across 4000 distinct protein structures, taken from the Protein Data Bank. The decomposition obtained correlates well with existing PFAM sequence classifications, but has the advantage of spanning the full protein, with the potential for novel domains. The coarse-grained network formed by the communities can also be used as a proxy for protein topology at the single-chain level; we demonstrate that grouping these proteins by their coarse-grained network results in a functionally significant classification.

arXiv:1801.07130

Computational Protein Design with Deep Learning Neural Networks

Computational protein design has a wide variety of applications. Despite its remarkable success, designing a protein for a given structure and function is still a challenging task. On the other hand, the number of solved protein structures is rapidly increasing while the number of unique protein folds has reached a steady number, suggesting more structural information is being accumulated on each fold. Deep learning neural network is a powerful method to learn such big data set and has shown superior performance in many machine learning fields. In this study, we applied the deep learning neural network approach to computational protein design for predicting the probability of 20 natural amino acids on each residue in a protein. A large set of protein structures was collected and a multi-layer neural network was constructed. A number of structural properties were extracted as input features and the best network achieved an accuracy of 38.3%. Using the network output as residue type restraints was able to improve the average sequence identity in designing three natural proteins using Rosetta. Moreover, the predictions from our network show ~3% higher sequence identity than a previous method. Results from this study may benefit further development of computational protein design methods.

arXiv:1704.05883

Rigidity strengthening is a vital mechanism for protein-ligand binding

Protein-ligand binding is essential to almost all life processes. The understanding of protein-ligand interactions is fundamentally important to rational drug design and protein design. Based on large scale data sets, we show that protein rigidity strengthening or flexibility reduction is a pivoting mechanism in protein-ligand binding. Our approach based solely on rigidity is able to unveil a surprisingly long range contribution of four residue layers to protein-ligand binding, which has a ramification for drug and protein design. Additionally, the present work reveals that among various pairwise interactions, the short range ones within the distance of the van der Waals diameter are most important. It is found that the present approach outperforms all the other state-of-the-art scoring functions for protein-ligand binding affinity predictions of two benchmark data sets

arXiv:1606.05373

Folding and Stabilization of Native-Sequence-Reversed Proteins

Though the problem of sequence-reversed protein folding is largely unexplored, one might speculate that reversed native protein sequences should be significantly more foldable than purely random heteropolymer sequences. In this article, we investigate how the reverse-sequences of native proteins might fold by examining a series of small proteins of increasing structural complexity ({\alpha}-helix, \b{eta}-hairpin, {\alpha}-helix bundle, and {\alpha}/\b{eta}-protein). Employing a tandem protein structure prediction algorithmic and molecular dynamics simulation approach, we find that the ability of reverse sequences to adopt native-like folds is strongly in influenced by protein size and the flexibility of the native hydrophobic core. For \b{eta}-hairpins with reverse-sequences that fail to fold, we employ a simple mutational strategy for guiding stable hairpin formation that involves the insertion of amino acids into the \b{eta}-turn region. This systematic look at reverse sequence duality sheds new light on the problem of protein sequence-structure mapping and may serve to inspire new protein design and protein structure prediction protocols.

arXiv:1408.5803

Ideal gas behavior of rotamerically defined conformers in native globular proteins

Protein conformational transitions, which are essential for function, may be driven either by entropy or enthalpy when molecular systems comprising solute and solvent molecules are the focus. Revealing thermodynamic origin of a given molecular process is an important but difficult task, and general principles governing protein conformational distributions remain elusive. Here we demonstrate that when protein molecules are taken as thermodynamic systems and solvents being treated as the environment, conformational entropy is an excellent proxy for free energy and is sufficient to explain protein conformational distributions. Specifically, by defining each unique combination of side chain torsional state as a conformer, the population distribution (or free energy) on an arbitrarily given order parameter is approximately a linear function of conformational entropy. Additionally, span of various microscopic potential energy terms is observed to be highly correlated with both conformational entropy and free energy. Presently widely utilized free energy proxies, including minimum potential energy, average potential energy terms by themselves or in combination with vibrational entropy\cite, are found to correlate with free energy rather poorly. Therefore, our findings provide a fundamentally new theoretical base for development of significantly more reliable and efficient next generation computational tools, where the number of available conformers,rather than poential energy of microscopic configurations, is the central focus. We anticipate that many related research fields, including structure based drug design and discovery, protein design, docking and prediction of general intermolecular interactions involving proteins, are expected to benefit greatly.

arXiv:2009.01411공식 코드 · Papers with Code

Learning from Protein Structure with Geometric Vector Perceptrons

Learning on 3D structures of large biomolecules is emerging as a distinct area in machine learning, but there has yet to emerge a unifying network architecture that simultaneously leverages the graph-structured and geometric aspects of the problem domain. To address this gap, we introduce geometric vector perceptrons, which extend standard dense layers to operate on collections of Euclidean vectors. Graph neural networks equipped with such layers are able to perform both geometric and relational reasoning on efficient and natural representations of macromolecular structure. We demonstrate our approach on two important problems in learning from protein structure: model quality assessment and computational protein design. Our approach improves over existing classes of architectures, including state-of-the-art graph-based and voxel-based methods. We release our code at https://github.com/drorlab/gvp.

arXiv:2008.12473

Pre-training of Graph Neural Network for Modeling Effects of Mutations on Protein-Protein Binding Affinity

Modeling the effects of mutations on the binding affinity plays a crucial role in protein engineering and drug design. In this study, we develop a novel deep learning based framework, named GraphPPI, to predict the binding affinity changes upon mutations based on the features provided by a graph neural network (GNN). In particular, GraphPPI first employs a well-designed pre-training scheme to enforce the GNN to capture the features that are predictive of the effects of mutations on binding affinity in an unsupervised manner and then integrates these graphical features with gradient-boosting trees to perform the prediction. Experiments showed that, without any annotated signals, GraphPPI can capture meaningful patterns of the protein structures. Also, GraphPPI achieved new state-of-the-art performance in predicting the binding affinity changes upon both single- and multi-point mutations on five benchmark datasets. In-depth analyses also showed GraphPPI can accurately estimate the effects of mutations on the binding affinity between SARS-CoV-2 and its neutralizing antibodies. These results have established GraphPPI as a powerful and useful computational tool in the studies of protein design.

arXiv:2004.13167공식 코드 · Papers with Code

Energy-based models for atomic-resolution protein conformations

We propose an energy-based model (EBM) of protein conformations that operates at atomic scale. The model is trained solely on crystallized protein data. By contrast, existing approaches for scoring conformations use energy functions that incorporate knowledge of physical principles and features that are the complex product of several decades of research and tuning. To evaluate the model, we benchmark on the rotamer recovery task, the problem of predicting the conformation of a side chain from its context within a protein structure, which has been used to evaluate energy functions for protein design. The model achieves performance close to that of the Rosetta energy function, a state-of-the-art method widely used in protein structure prediction and design. An investigation of the model's outputs and hidden representations finds that it captures physicochemical properties relevant to protein energy.

arXiv:2003.06601

Lattice protein design using Bayesian learning

Protein design is the inverse approach of the three-dimensional (3D) structure prediction for elucidating the relationship between the 3D structures and amino acid sequences. In general, the computation of the protein design involves a double loop: a loop for amino acid sequence changes and a loop for an exhaustive conformational search for each amino acid sequence. Herein, we propose a novel statistical mechanical design method using Bayesian learning, which can design lattice proteins without the exhaustive conformational search. We consider a thermodynamic hypothesis of the evolution of proteins and apply it to the prior distribution of amino acid sequences. Furthermore, we take the water effect into account in view of the grand canonical picture. As a result, on applying the 2D lattice hydrophobic-polar (HP) model, our design method successfully finds an amino acid sequence for which the target conformation has a unique ground state. However, the performance was not as good for the 3D lattice HP models compared to the 2D models. The performance of the 3D model improves on using a 20-letter lattice proteins. Furthermore, we find a strong linearity between the chemical potential of water and the number of surface residues, thereby revealing the relationship between protein structure and the effect of water molecules. The advantage of our method is that it greatly reduces computation time, because it does not require long calculations for the partition function corresponding to an exhaustive conformational search. As our method uses a general form of Bayesian learning and statistical mechanics and is not limited to lattice proteins, the results presented here elucidate some heuristics used successfully in previous protein design methods.

arXiv:1901.10060

Conditioning by adaptive sampling for robust design

We present a new method for design problems wherein the goal is to maximize or specify the value of one or more properties of interest. For example, in protein design, one may wish to find the protein sequence that maximizes fluorescence. We assume access to one or more, potentially black box, stochastic "oracle" predictive functions, each of which maps from input (e.g., protein sequences) design space to a distribution over a property of interest (e.g. protein fluorescence). At first glance, this problem can be framed as one of optimizing the oracle(s) with respect to the input. However, many state-of-the-art predictive models, such as neural networks, are known to suffer from pathologies, especially for data far from the training distribution. Thus we need to modulate the optimization of the oracle inputs with prior knowledge about what makes `realistic' inputs (e.g., proteins that stably fold). Herein, we propose a new method to solve this problem, Conditioning by Adaptive Sampling, which yields state-of-the-art results on a protein fluorescence problem, as compared to other recently published approaches. Formally, our method achieves its success by using model-based adaptive sampling to estimate the conditional distribution of the input sequences given the desired properties.