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.     

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.     

Improved performance in protein secondary structure prediction by combining multiple predictions

In this paper(1) we present a novel framework for protein secondary structure prediction. In this prediction framework, firstly we propose a novel parameterized semi-probability profile, which combines single sequence with evolutionary information effectively. Secondly, different semi-probability profiles are respectively applied as network input to predict protein secondary structure. Then a comparison among these different predictions is discussed in this article. Finally, na?ve Bayes approaches are used to combine these predictions in order to obtain a better prediction performance than individual prediction. The experimental results show that our proposed framework can indeed improve the prediction accuracy.     

Prediction of calcium-binding sites by combining loop-modeling with machine learning

Background  

Protein ligand-binding sites in the apo state exhibit structural flexibility. This flexibility often frustrates methods for structure-based recognition of these sites because it leads to the absence of electron density for these critical regions, particularly when they are in surface loops. Methods for recognizing functional sites in these missing loops would be useful for recovering additional functional information.     

Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14

Substantial progresses in protein structure prediction have been made by utilizing deep-learning and residue-residue distance prediction since CASP13. Inspired by the advances, we improve our CASP14 MULTICOM protein structure prediction system by incorporating three new components: (a) a new deep learning-based protein inter-residue distance predictor to improve template-free (ab initio) tertiary structure prediction, (b) an enhanced template-based tertiary structure prediction method, and (c) distance-based model quality assessment methods empowered by deep learning. In the 2020 CASP14 experiment, MULTICOM predictor was ranked seventh out of 146 predictors in tertiary structure prediction and ranked third out of 136 predictors in inter-domain structure prediction. The results demonstrate that the template-free modeling based on deep learning and residue-residue distance prediction can predict the correct topology for almost all template-based modeling targets and a majority of hard targets (template-free targets or targets whose templates cannot be recognized), which is a significant improvement over the CASP13 MULTICOM predictor. Moreover, the template-free modeling performs better than the template-based modeling on not only hard targets but also the targets that have homologous templates. The performance of the template-free modeling largely depends on the accuracy of distance prediction closely related to the quality of multiple sequence alignments. The structural model quality assessment works well on targets for which enough good models can be predicted, but it may perform poorly when only a few good models are predicted for a hard target and the distribution of model quality scores is highly skewed. MULTICOM is available at https://github.com/jianlin-cheng/MULTICOM_Human_CASP14/tree/CASP14_DeepRank3 and https://github.com/multicom-toolbox/multicom/tree/multicom_v2.0 .     

Review of machine learning methods for RNA secondary structure prediction

Secondary structure plays an important role in determining the function of noncoding RNAs. Hence, identifying RNA secondary structures is of great value to research. Computational prediction is a mainstream approach for predicting RNA secondary structure. Unfortunately, even though new methods have been proposed over the past 40 years, the performance of computational prediction methods has stagnated in the last decade. Recently, with the increasing availability of RNA structure data, new methods based on machine learning (ML) technologies, especially deep learning, have alleviated the issue. In this review, we provide a comprehensive overview of RNA secondary structure prediction methods based on ML technologies and a tabularized summary of the most important methods in this field. The current pending challenges in the field of RNA secondary structure prediction and future trends are also discussed.     

Analysis of distance-based protein structure prediction by deep learning in CASP13

This paper reports the CASP13 results of distance-based contact prediction, threading, and folding methods implemented in three RaptorX servers, which are built upon the powerful deep convolutional residual neural network (ResNet) method initiated by us for contact prediction in CASP12. On the 32 CASP13 FM (free-modeling) targets with a median multiple sequence alignment (MSA) depth of 36, RaptorX yielded the best contact prediction among 46 groups and almost the best 3D structure modeling among all server groups without time-consuming conformation sampling. In particular, RaptorX achieved top L/5, L/2, and L long-range contact precision of 70%, 58%, and 45%, respectively, and predicted correct folds (TMscore > 0.5) for 18 of 32 targets. Further, RaptorX predicted correct folds for all FM targets with >300 residues (T0950-D1, T0969-D1, and T1000-D2) and generated the best 3D models for T0950-D1 and T0969-D1 among all groups. This CASP13 test confirms our previous findings: (a) predicted distance is more useful than contacts for both template-based and free modeling; and (b) structure modeling may be improved by integrating template and coevolutionary information via deep learning. This paper will discuss progress we have made since CASP12, the strength and weakness of our methods, and why deep learning performed much better in CASP13.     

Creating auxotrophic mutants in Methylophilus methylotrophus AS1 by combining electroporation and chemical mutagenesis

Stable auxotrophic mutants of the methylotroph Methylophilus methylotrophus AS1 were obtained by a novel mutagenesis technique in which electroporation is used to transport the chemical mutagen N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) across the cell membrane. By combining chemical mutagenesis with electroporation and screening single colonies for auxotrophy in 36 different amino acids and growth factors, 3 auxotrophs per 156 colonies screened were obtained, whereas no auxotrophs were found with chemical mutagenesis alone. MNNG mutagen toxicity was also increased in the methylotroph with this novel mutagenesis technique (death rate 96% compared to 79%). This technique did not increase the mutation rate for strain Escherichia coli BK6 which responds well to simple exposure to the mutagen. Received: 9 December 1996 / Received revision: 31 March 1997 / Accepted: 13 April 1997     

Accurate prediction of protein secondary structure and solvent accessibility by consensus combiners of sequence and structure information

Background  

Structural properties of proteins such as secondary structure and solvent accessibility contribute to three-dimensional structure prediction, not only in the ab initio case but also when homology information to known structures is available. Structural properties are also routinely used in protein analysis even when homology is available, largely because homology modelling is lower throughput than, say, secondary structure prediction. Nonetheless, predictors of secondary structure and solvent accessibility are virtually always ab initio.     

Nucleotide sequence changes in polyoma ts-a mutants: correlation with protein structure.

The mutations in three polyoma ts-a mutants have been determined. Two mutants, ts-25 and ts-52, have different single-base changes at the same position (2883) in the early region corresponding to a conserved glycine residue very near the C-terminus of the polyoma large T antigen. Mutant ts-48 has a single-base change at position 2341, as well as a second change at position 1228, in the region of large T antigen shared with medium T antigen.     

Glioma stages prediction based on machine learning algorithm combined with protein-protein interaction networks

BackgroundGlioma is the most lethal nervous system cancer. Recent studies have made great efforts to study the occurrence and development of glioma, but the molecular mechanisms are still unclear. This study was designed to reveal the molecular mechanisms of glioma based on protein-protein interaction network combined with machine learning methods. Key differentially expressed genes (DEGs) were screened and selected by using the protein-protein interaction (PPI) networks.ResultsAs a result, 19 genes between grade I and grade II, 21 genes between grade II and grade III, and 20 genes between grade III and grade IV. Then, five machine learning methods were employed to predict the gliomas stages based on the selected key genes. After comparison, Complement Naive Bayes classifier was employed to build the prediction model for grade II-III with accuracy 72.8%. And Random forest was employed to build the prediction model for grade I-II and grade III-VI with accuracy 97.1% and 83.2%, respectively. Finally, the selected genes were analyzed by PPI networks, Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, and the results improve our understanding of the biological functions of select DEGs involved in glioma growth. We expect that the key genes expressed have a guiding significance for the occurrence of gliomas or, at the very least, that they are useful for tumor researchers.ConclusionMachine learning combined with PPI networks, GO and KEGG analyses of selected DEGs improve our understanding of the biological functions involved in glioma growth.     

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。     

基于机器学习的蛋白质编码区识别

针对DNA序列编码区的识别问题,本研究提出一个特征向量和逻辑回归的组合模型。首先对DNA序列进行数值处理转化为特征向量,并结合k字符相对频率技术提取特征向量的元素特征,之后利用二分类逻辑回归算法,对编码区和非编码区进行准确区分。选取了HMR195和BG570两个基准数据集进行五折交叉验证,结果表明,平均AUC(Area Under Curve)值分别为0.981 3和0.987 4,明显优于传统的贝叶斯判别法和VOSSDFT等方法。此外,本文提出的特征向量的维度很低,提高了运算效率。因此,本文组合模型能够较为高效准确地识别蛋白质编码区。     

A machine learning approach for the prediction of protein surface loop flexibility

Proteins often undergo conformational changes when binding to each other. A major fraction of backbone conformational changes involves motion on the protein surface, particularly in loops. Accounting for the motion of protein surface loops represents a challenge for protein-protein docking algorithms. A first step in addressing this challenge is to distinguish protein surface loops that are likely to undergo backbone conformational changes upon protein-protein binding (mobile loops) from those that are not (stationary loops). In this study, we developed a machine learning strategy based on support vector machines (SVMs). Our SVM uses three features of loop residues in the unbound protein structures-Ramachandran angles, crystallographic B-factors, and relative accessible surface area-to distinguish mobile loops from stationary ones. This method yields an average prediction accuracy of 75.3% compared with a random prediction accuracy of 50%, and an average of 0.79 area under the receiver operating characteristic (ROC) curve using cross-validation. Testing the method on an independent dataset, we obtained a prediction accuracy of 70.5%. Finally, we applied the method to 11 complexes that involve members from the Ras superfamily and achieved prediction accuracy of 92.8% for the Ras superfamily proteins and 74.4% for their binding partners.     

Automated use of mutagenesis data in structure prediction

In the absence of experimental structural determination, numerous methods are available to indirectly predict or probe the structure of a target molecule. Genetic modification of a protein sequence is a powerful tool for identifying key residues involved in binding reactions or protein stability. Mutagenesis data is usually incorporated into the modeling process either through manual inspection of model compatibility with empirical data, or through the generation of geometric constraints linking sensitive residues to a binding interface. We present an approach derived from statistical studies of lattice models for introducing mutation information directly into the fitness score. The approach takes into account the phenotype of mutation (neutral or disruptive) and calculates the energy for a given structure over an ensemble of sequences. The structure prediction procedure searches for the optimal conformation where neutral sequences either have no impact or improve stability and disruptive sequences reduce stability relative to wild type. We examine three types of sequence ensembles: information from saturation mutagenesis, scanning mutagenesis, and homologous proteins. Incorporating multiple sequences into a statistical ensemble serves to energetically separate the native state and misfolded structures. As a result, the prediction of structure with a poor force field is sufficiently enhanced by mutational information to improve accuracy. Furthermore, by separating misfolded conformations from the target score, the ensemble energy serves to speed up conformational search algorithms such as Monte Carlo-based methods.     

Machine learning approach for the prediction of protein secondary structure

PROMIS (protein machine induction system), a program for machine learning, was used to generalize rules that characterize the relationship between primary and secondary structure in globular proteins. These rules can be used to predict an unknown secondary structure from a known primary structure. The symbolic induction method used by PROMIS was specifically designed to produce rules that are meaningful in terms of chemical properties of the residues. The rules found were compared with existing knowledge of protein structure: some features of the rules were already recognized (e.g. amphipathic nature of alpha-helices). Other features are not understood, and are under investigation. The rules produced a prediction accuracy for three states (alpha-helix, beta-strand and coil) of 60% for all proteins, 73% for proteins of known alpha domain type, 62% for proteins of known beta domain type and 59% for proteins of known alpha/beta domain type. We conclude that machine learning is a useful tool in the examination of the large databases generated in molecular biology.     

Deep learning based prediction of species-specific protein S-glutathionylation sites

As a widespread and reversible post-translational modification of proteins, S-glutathionylation specifically generates the mixed disulfides between cysteine residues and glutathione, which regulates various biological processes including oxidative stress, nitrosative stress and signal transduction. The identification of proteins and specific sites that undergo S-glutathionylation is crucial for understanding the underlying mechanisms and regulatory effects of S-glutathionylation. Experimental identification of S-glutathionylation sites is laborious and time-consuming, whereas computational predictions are more attractive due to their high speed and convenience. Here, we developed a novel computational framework DeepGSH (http://deepgsh.cancerbio.info/) for species-specific S-glutathionylation sites prediction, based on deep learning and particle swarm optimization algorithms. 5-fold cross validation indicated that DeepGSH was able to achieve an AUC of 0.8393 and 0.8458 for Homo sapiens and Mus musculus. According to critical evaluation and comparison, DeepGSH showed excellent robustness and better performance than existing tools in both species, demonstrating DeepGSH was suitable for S-glutathionylation prediction. The prediction results of DeepGSH might provide guidance for experimental validation of S-glutathionylation sites and helpful information to understand the intrinsic mechanisms.