Deep learning for protein complex structure prediction

Recent developments in the structure prediction of protein complexes have resulted in accuracies rivalling experimental methods in many cases. The high accuracy is mainly observed in dimeric complexes and other problems such as protein disorder and predicting the structure of host-pathogen interactions remain. This review highlights the foundation for current accurate structure prediction of protein complexes and possible ways to address the remaining limitations.     

面向蛋白质功能位点识别的机器学习平台构建

有关蛋白质功能的研究是解析生命奥秘的基础,机器学习技术在该领域已有广泛应用。利用支持向量机(support vectormachine,SVM)方法,构建一个预测蛋白质功能位点的通用平台。该平台先提取非同源蛋白质序列,再对这些序列进行特征编码(包括序列的基本信息、物化特征、结构信息及序列保守性特征等),以编码好的样本作为训练数据,利用SVM进行训练,得到敏感性、特异性、Matthew相关系数、准确率及ROC曲线等评价指标,反复测试,得到评价指标最优的SVM模型后,便可以用来预测蛋白质序列上的功能位点。该平台除了应用在预测蛋白质功能位点之外,还可以应用于疾病相关单核苷酸多态性(SNP)预测分析、预测蛋白质结构域分析、生物分子间的相互作用等。     

基于机器学习的RNA编辑位点预测方法综述

RNA编辑是一个十分重要的生物细胞分子机制。作为转录后修饰的一步,它可以增加蛋白质组学多样性,改变转录产物的稳定性,调节基因表达等。RNA编辑失调会导致各种疾病,包括神经疾病和癌症。在动物中,腺苷到肌苷(A-to-I)的编辑是最普遍的。高通量测序技术的进步大大提高了在全局范围内检测和量化RNA编辑的能力,使得RNA编辑的大规模全基因组分析变得可行,产生了一系列基于高通量测序技术的RNA编辑位点预测方法。通过对这些方法进行介绍、总结和分析,为RNA编辑的进一步研究提供一些思路。     

Deep template-based protein structure prediction

MotivationProtein structure prediction has been greatly improved by deep learning, but most efforts are devoted to template-free modeling. But very few deep learning methods are developed for TBM (template-based modeling), a popular technique for protein structure prediction. TBM has been studied extensively in the past, but its accuracy is not satisfactory when highly similar templates are not available.ResultsThis paper presents a new method NDThreader (New Deep-learning Threader) to address the challenges of TBM. NDThreader first employs DRNF (deep convolutional residual neural fields), which is an integration of deep ResNet (convolutional residue neural networks) and CRF (conditional random fields), to align a query protein to templates without using any distance information. Then NDThreader uses ADMM (alternating direction method of multipliers) and DRNF to further improve sequence-template alignments by making use of predicted distance potential. Finally, NDThreader builds 3D models from a sequence-template alignment by feeding it and sequence coevolution information into a deep ResNet to predict inter-atom distance distribution, which is then fed into PyRosetta for 3D model construction. Our experimental results show that NDThreader greatly outperforms existing methods such as CNFpred, HHpred, DeepThreader and CEthreader. NDThreader was blindly tested in CASP14 as a part of RaptorX server, which obtained the best average GDT score among all CASP14 servers on the 58 TBM targets.     

Stress-induced protein S-glutathionylation in Arabidopsis

S-Glutathionylation (thiolation) is a ubiquitous redox-sensitive and reversible modification of protein cysteinyl residues that can directly regulate their activity. While well established in animals, little is known about the formation and function of these mixed disulfides in plants. After labeling the intracellular glutathione pool with [35S]cysteine, suspension cultures of Arabidopsis (Arabidopsis thaliana ecotype Columbia) were shown to undergo a large increase in protein thiolation following treatment with the oxidant tert-butylhydroperoxide. To identify proteins undergoing thiolation, a combination of in vivo and in vitro labeling methods utilizing biotinylated, oxidized glutathione (GSSG-biotin) was developed to isolate Arabidopsis proteins/protein complexes that can be reversibly glutathionylated. Following two-dimensional polyacrylamide gel electrophoresis and matrix-assisted laser desorption/ionization time of flight mass spectrometry proteomics, a total of 79 polypeptides were identified, representing a mixture of proteins that underwent direct thiolation as well as proteins complexed with thiolated polypeptides. The mechanism of thiolation of five proteins, dehydroascorbate reductase (AtDHAR1), zeta-class glutathione transferase (AtGSTZ1), nitrilase (AtNit1), alcohol dehydrogenase (AtADH1), and methionine synthase (AtMetS), was studied using the respective purified recombinant proteins. AtDHAR1, AtGSTZ1, and to a lesser degree AtNit1 underwent spontaneous thiolation with GSSG-biotin through modification of active-site cysteines. The thiolation of AtADH1 and AtMetS required the presence of unidentified Arabidopsis proteins, with this activity being inhibited by S-modifying agents. The potential role of thiolation in regulating metabolism in Arabidopsis is discussed and compared with other known redox regulatory systems operating in plants.     

Avian species-specific tandem repeats contain nuclear protein binding sites.

Three avian highly repetitive tandem repeats were identified and examined. These repeats had similar unit lengths (about 42 bp long) but completely different sequences each containing particular protein binding sites. Each of these repeats was found within only one of the five closely related genera studied.     

Deep learning for DNase I hypersensitive sites identification

Background

The DNase I hypersensitive sites (DHSs) are associated with the cis-regulatory DNA elements. An efficient method of identifying DHSs can enhance the understanding on the accessibility of chromatin. Despite a multitude of resources available on line including experimental datasets and computational tools, the complex language of DHSs remains incompletely understood.

Methods

Here, we address this challenge using an approach based on a state-of-the-art machine learning method. We present a novel convolutional neural network (CNN) which combined Inception like networks with a gating mechanism for the response of multiple patterns and longterm association in DNA sequences to predict multi-scale DHSs in Arabidopsis, rice and Homo sapiens.

Results

Our method obtains 0.961 area under curve (AUC) on Arabidopsis, 0.969 AUC on rice and 0.918 AUC on Homo sapiens.

Conclusions

Our method provides an efficient and accurate way to identify multi-scale DHSs sequences by deep learning.

     

S-glutathionylation in protein redox regulation

Protein S-glutathionylation, the reversible formation of mixed disulfides between glutathione and low-pKa cysteinyl residues, not only is a cellular response to mild oxidative/nitrosative stress, but also occurs under basal (physiological) conditions. S-glutathionylation has now emerged as a potential mechanism for dynamic, posttranslational regulation of a variety of regulatory, structural, and metabolic proteins. Moreover, substantial recent studies have implicated S-glutathionylation in the regulation of signaling and metabolic pathways in intact cellular systems. The growing list of S-glutathionylated proteins, in both animal and plant cells, attests to the occurrence of S-glutathionylation in cellular response pathways. The existence of antioxidant enzymes that specifically regulate S-glutathionylation would emphasize its importance in modulating protein function, suggesting that this protein modification too might have a role in cell signaling. The continued development of proteomic and analytical methods for disulfide analysis will help us better understand the full extent of the roles these modifications play in the regulation of cell function. In this review, we describe recent breakthroughs in our understanding of the potential role of protein S-glutathionylation in the redox regulation of signal transduction.     

Fluorescein-labeled glutathione to study protein S-glutathionylation

Numerous studies of S-glutathionylation of cysteine thiols indicate that this protein modification plays a key role in redox regulation of proteins. To facilitate the study of protein S-glutathionylation, we developed a synthesis and purification to produce milligram quantities of fluorescein-labeled glutathione. The amino terminus of the glutathione tripeptide reacted with fluorescein isothiocyanate readily in ammonium bicarbonate. Purification by solid phase extraction on C8 and C18 columns separated excess reactants from desired products. Both oxidized and reduced fluorescein-labeled glutathione reacted with a variety of thiol-containing proteins to yield fluorescent proteins.     

A global machine learning based scoring function for protein structure prediction

We present a knowledge‐based function to score protein decoys based on their similarity to native structure. A set of features is constructed to describe the structure and sequence of the entire protein chain. Furthermore, a qualitative relationship is established between the calculated features and the underlying electromagnetic interaction that dominates this scale. The features we use are associated with residue–residue distances, residue–solvent distances, pairwise knowledge‐based potentials and a four‐body potential. In addition, we introduce a new target to be predicted, the fitness score, which measures the similarity of a model to the native structure. This new approach enables us to obtain information both from decoys and from native structures. It is also devoid of previous problems associated with knowledge‐based potentials. These features were obtained for a large set of native and decoy structures and a back‐propagating neural network was trained to predict the fitness score. Overall this new scoring potential proved to be superior to the knowledge‐based scoring functions used as its inputs. In particular, in the latest CASP (CASP10) experiment our method was ranked third for all targets, and second for freely modeled hard targets among about 200 groups for top model prediction. Ours was the only method ranked in the top three for all targets and for hard targets. This shows that initial results from the novel approach are able to capture details that were missed by a broad spectrum of protein structure prediction approaches. Source codes and executable from this work are freely available at http://mathmed.org /#Software and http://mamiris.com/ . Proteins 2014; 82:752–759. © 2013 Wiley Periodicals, Inc.     

Catalase and estradiol inhibit mitochondrial protein S-glutathionylation

Regulation and downstream effects of mitochondrial protein S-glutathionylation in response to oxidative stress are poorly understood. The study aim was to determine whether anti-oxidants such as catalase and estradiol alter mitochondrial protein S-glutathionylation and in turn affect apoptosis following ultraviolet B (UV-B) light irradiation. HeLa cells were transduced with increasing amounts of adenovirus encoding catalase (Ad-Cat) and β-galactosidase (Ad-Lac Z) or pre-incubated with estradiol before induction of apoptosis by UV-B light exposure. Inhibition of mitochondrial protein S-glutathionylation was assessed using autoantibodies specific for the non-S-glutathionylated form of PDC-E2. The percentage of apoptotic cells following UV-B irradiation were not significantly different between mock cells (cells with no virus infection) and Ad-Cat and Ad-Lac Z infected cells at all viral doses (all p > 0.050). Autoantibody staining of non-S-glutathionylated PDC-E2 in apoptotic cells was three times greater in only Ad-Cat infected cells compared to only Ad-Lac Z infected cells (81.3 ± 16.7 vs 26 ± 7.2 %, respectively, p = 0.030). Similarly estradiol treatment (33 and 100 nM) also significantly increased PDC-E2 staining in apoptotic cells compared to non-treated cells (both p < 0.010). The percentage of apoptotic cells was not significantly different with any of the estradiol concentrations (all p > 0.100). The observed procaspase 12 cleavage following UV-B irradiation suggests that a mitochondrial-independent apoptotic pathway was activated. In conclusion, following an apoptotic stimulus, estradiol may inhibit mitochondrial protein S-glutathionylation without inhibiting apoptosis. This effect may play a role in ninefold greater prevalence of autoantibodies against PDC-E2 in women with primary biliary cirrhosis.     

Machine learning in protein structure prediction

Prediction of protein structure from sequence has been intensely studied for many decades, owing to the problem's importance and its uniquely well-defined physical and computational bases. While progress has historically ebbed and flowed, the past two years saw dramatic advances driven by the increasing “neuralization” of structure prediction pipelines, whereby computations previously based on energy models and sampling procedures are replaced by neural networks. The extraction of physical contacts from the evolutionary record; the distillation of sequence–structure patterns from known structures; the incorporation of templates from homologs in the Protein Databank; and the refinement of coarsely predicted structures into finely resolved ones have all been reformulated using neural networks. Cumulatively, this transformation has resulted in algorithms that can now predict single protein domains with a median accuracy of 2.1 Å, setting the stage for a foundational reconfiguration of the role of biomolecular modeling within the life sciences.     

Text mining improves prediction of protein functional sites

We present an approach that integrates protein structure analysis and text mining for protein functional site prediction, called LEAP-FS (Literature Enhanced Automated Prediction of Functional Sites). The structure analysis was carried out using Dynamics Perturbation Analysis (DPA), which predicts functional sites at control points where interactions greatly perturb protein vibrations. The text mining extracts mentions of residues in the literature, and predicts that residues mentioned are functionally important. We assessed the significance of each of these methods by analyzing their performance in finding known functional sites (specifically, small-molecule binding sites and catalytic sites) in about 100,000 publicly available protein structures. The DPA predictions recapitulated many of the functional site annotations and preferentially recovered binding sites annotated as biologically relevant vs. those annotated as potentially spurious. The text-based predictions were also substantially supported by the functional site annotations: compared to other residues, residues mentioned in text were roughly six times more likely to be found in a functional site. The overlap of predictions with annotations improved when the text-based and structure-based methods agreed. Our analysis also yielded new high-quality predictions of many functional site residues that were not catalogued in the curated data sources we inspected. We conclude that both DPA and text mining independently provide valuable high-throughput protein functional site predictions, and that integrating the two methods using LEAP-FS further improves the quality of these predictions.     

Machine learning techniques for protein function prediction

Proteins play important roles in living organisms, and their function is directly linked with their structure. Due to the growing gap between the number of proteins being discovered and their functional characterization (in particular as a result of experimental limitations), reliable prediction of protein function through computational means has become crucial. This paper reviews the machine learning techniques used in the literature, following their evolution from simple algorithms such as logistic regression to more advanced methods like support vector machines and modern deep neural networks. Hyperparameter optimization methods adopted to boost prediction performance are presented. In parallel, the metamorphosis in the features used by these algorithms from classical physicochemical properties and amino acid composition, up to text-derived features from biomedical literature and learned feature representations using autoencoders, together with feature selection and dimensionality reduction techniques, are also reviewed. The success stories in the application of these techniques to both general and specific protein function prediction are discussed.     

Improved prediction of protein ligand-binding sites using random forests

This article describes a novel method for predicting ligand-binding sites of proteins. This method uses only 8 structural properties as input vector to train 9 random forest classifiers which are combined to predict binding residues. These predicted binding residues are then clustered into some predicted ligand-binding sites. According to our measurement criterion, this method achieved a success rate of 0.914 in the bound state dataset and 0.800 in the unbound state dataset, which are better than three other methods: Q-SiteFinder, SCREEN and Morita's method. It indicates that the proposed method here is successful for predicting ligand-binding sites.     

Deep learning allows genome-scale prediction of Michaelis constants from structural features

The Michaelis constant K_M describes the affinity of an enzyme for a specific substrate and is a central parameter in studies of enzyme kinetics and cellular physiology. As measurements of K_M are often difficult and time-consuming, experimental estimates exist for only a minority of enzyme–substrate combinations even in model organisms. Here, we build and train an organism-independent model that successfully predicts K_M values for natural enzyme–substrate combinations using machine and deep learning methods. Predictions are based on a task-specific molecular fingerprint of the substrate, generated using a graph neural network, and on a deep numerical representation of the enzyme’s amino acid sequence. We provide genome-scale K_M predictions for 47 model organisms, which can be used to approximately relate metabolite concentrations to cellular physiology and to aid in the parameterization of kinetic models of cellular metabolism.

To understand the action of an enzyme, we need to know its affinity for its substrates, quantified by Michaelis constants, but these are difficult to measure experimentally. This study shows that a deep learning model that can predict them from structural features of the enzyme and substrate, providing KM predictions for all enzymes across 47 model organisms.     

Deep learning architectures for prediction of nucleosome positioning from sequences data

Background

Nucleosomes are DNA-histone complex, each wrapping about 150 pairs of double-stranded DNA. Their function is fundamental for one of the primary functions of Chromatin i.e. packing the DNA into the nucleus of the Eukaryote cells. Several biological studies have shown that the nucleosome positioning influences the regulation of cell type-specific gene activities. Moreover, computational studies have shown evidence of sequence specificity concerning the DNA fragment wrapped into nucleosomes, clearly underlined by the organization of particular DNA substrings. As the main consequence, the identification of nucleosomes on a genomic scale has been successfully performed by computational methods using a sequence features representation.

Results

In this work, we propose a deep learning model for nucleosome identification. Our model stacks convolutional layers and Long Short-term Memories to automatically extract features from short- and long-range dependencies in a sequence. Using this model we are able to avoid the feature extraction and selection steps while improving the classification performances.

Conclusions

Results computed on eleven data sets of five different organisms, from Yeast to Human, show the superiority of the proposed method with respect to the state of the art recently presented in the literature.