Partition dataset according to amino acid type improves the prediction of deleterious non-synonymous SNPs

Many non-synonymous SNPs (nsSNPs) are associated with diseases, and numerous machine learning methods have been applied to train classifiers for sorting disease-associated nsSNPs from neutral ones. The continuously accumulated nsSNP data allows us to further explore better prediction approaches. In this work, we partitioned the training data into 20 subsets according to either original or substituted amino acid type at the nsSNP site. Using support vector machine (SVM), training classification models on each subset resulted in an overall accuracy of 76.3% or 74.9% depending on the two different partition criteria, while training on the whole dataset obtained an accuracy of only 72.6%. Moreover, the dataset was also randomly divided into 20 subsets, but the corresponding accuracy was only 73.2%. Our results demonstrated that partitioning the whole training dataset into subsets properly, i.e., according to the residue type at the nsSNP site, will improve the performance of the trained classifiers significantly, which should be valuable in developing better tools for predicting the disease-association of nsSNPs.     

hRFT2基因单核苷酸多态性的生物信息学分析

旨在筛选可能与人类疾病有关的hRFT2基因单核苷酸多态性(single nucleotide polymorphisms,nsSNPs)和突变位点,从SNP数据库中检索并筛选出395个有效的hRFT2基因SNPs,其中包括30个同义SNPs(synonymous SNPs,sSNPs)和31个非同义SNPs(non-synonymous single nucleotide polymorphisms,nsSNPs)。分别采用SIFT、SNPs3D和PolyPhen-2方法分析nsSNPs引起的氨基酸替换是否可能影响hRFT2的功能。结果表明,5个nsSNPs(rs11477762、rs146302587、rs146492942、rs76947760和rs145431028)可能严重影响hRFT2蛋白的功能,其中rs76947760和rs145431028的影响已得到临床证明,另外3个nsSNPs(rs148387972、rs140391358和rs3746802)也可能对hRFT2有较大的影响。     

DAMpred: Recognizing Disease-Associated nsSNPs through Bayes-Guided Neural-Network Model Built on Low-Resolution Structure Prediction of Proteins and Protein–Protein Interactions

Nearly one-third of non-synonymous single-nucleotide polymorphism (nsSNPs) are deleterious to human health, but recognition of the disease-associated mutations remains a significant unsolved problem. We proposed a new algorithm, DAMpred, to identify disease-causing nsSNPs through the coupling of evolutionary profiles with structure predictions of proteins and protein–protein interactions. The pipeline was trained by a novel Bayes-guided artificial neural network algorithm that incorporates posterior probabilities of distinct feature classifiers with the network training process. DAMpred was tested on a large-scale data set involving 10,635 nsSNPs from 2154 ORFs in the human genome and recognized disease-associated nsSNPs with an accuracy 0.80 and a Matthews correlation coefficient of 0.601, which is 9.1% higher than the best of other state-of-the-art methods. In the blind test on the TP53 gene, DAMpred correctly recognized the mutations causative of Li–Fraumeni-like syndrome with a Matthews correlation coefficient that is 27% higher than the control methods. The study demonstrates an efficient avenue to quantitatively model the association of nsSNPs with human diseases from low-resolution protein structure prediction, which should find important usefulness in diagnosis and treatment of genetic diseases.     

Non-synonymous and synonymous coding SNPs show similar likelihood and effect size of human disease association

Many DNA variants have been identified on more than 300 diseases and traits using Genome-Wide Association Studies (GWASs). Some have been validated using deep sequencing, but many fewer have been validated functionally, primarily focused on non-synonymous coding SNPs (nsSNPs). It is an open question whether synonymous coding SNPs (sSNPs) and other non-coding SNPs can lead to as high odds ratios as nsSNPs. We conducted a broad survey across 21,429 disease-SNP associations curated from 2,113 publications studying human genetic association, and found that nsSNPs and sSNPs shared similar likelihood and effect size for disease association. The enrichment of disease-associated SNPs around the 80(th) base in the first introns might provide an effective way to prioritize intronic SNPs for functional studies. We further found that the likelihood of disease association was positively associated with the effect size across different types of SNPs, and SNPs in the 3' untranslated regions, such as the microRNA binding sites, might be under-investigated. Our results suggest that sSNPs are just as likely to be involved in disease mechanisms, so we recommend that sSNPs discovered from GWAS should also be examined with functional studies.     

Predicting disease-associated substitution of a single amino acid by analyzing residue interactions

Background  

The rapid accumulation of data on non-synonymous single nucleotide polymorphisms (nsSNPs, also called SAPs) should allow us to further our understanding of the underlying disease-associated mechanisms. Here, we use complex networks to study the role of an amino acid in both local and global structures and determine the extent to which disease-associated and polymorphic SAPs differ in terms of their interactions to other residues.     

与疾病相关的非同义单核苷酸多态性预测的研究进展

单核苷酸多态性(single nucleotide polymorphism,SNPs),即在基因组水平上由单个核苷酸的变异而引起的DNA序列多态性变化,具体是指在DNA序列中的单个碱基的变异,其是人类基因组变异种最常见的一种。SNP研究最主要的目的就是对人类表型变异遗传学的理解,尤其是关于人类遗传疾病的研究。而非同义单核苷酸多态性(nsSNPs)是SNPs中的一种,主要是指处于编码区会引起翻译后对应氨基酸序列变化的单核苷酸突变。因为nsSNPs可能会对蛋白质的功能造成影响,被认为是造成人类遗传病的主要原因。因此将与疾病相关的nsSNPs从中性的nsSNPs中区分出来是很重要的。本文根据国内外与疾病相关nsSNPs预测的研究,分析了预测中所涉及到的特征属性,总结了对这些特征进行优化的特征选择方法,并概述了在预测过程中使用的各种分类器。     

Applications of computational algorithm tools to identify functional SNPs in cytokine genes

Understanding the functions of single nucleotide polymorphisms (SNPs) can greatly help to understand the genetics of the human phenotype variation and especially the genetic basis of human complex diseases. However, how to identify functional SNPs from a pool containing both functional and neutral SNPs is challenging. In this study, we analyzed the genetic variations that can alter the expression and function of a group of cytokine proteins using computational tools. As a result, we extracted 4552 SNPs from 45 cytokine proteins from SNPper database. Of particular interest, 828 SNPs were in the 5'UTR region, 961 SNPs were in the 3' UTR region, and 85 SNPs were non-synonymous SNPs (nsSNPs), which cause amino acid change. Evolutionary conservation analysis using the SIFT tool suggested that 8 nsSNPs may disrupt the protein function. Protein structure analysis using the PolyPhen tool suggested that 5 nsSNPs might alter protein structure. Binding motif analysis using the UTResource tool suggested that 27 SNPs in 5' or 3'UTR might change protein expression levels. Our study demonstrates the presence of naturally occurring genetic variations in the cytokine proteins that may affect their expressions and functions with possible roles in complex human disease, such as immune diseases.     

Prediction of Disease Causing Non-Synonymous SNPs by the Artificial Neural Network Predictor NetDiseaseSNP

We have developed a sequence conservation-based artificial neural network predictor called NetDiseaseSNP which classifies nsSNPs as disease-causing or neutral. Our method uses the excellent alignment generation algorithm of SIFT to identify related sequences and a combination of 31 features assessing sequence conservation and the predicted surface accessibility to produce a single score which can be used to rank nsSNPs based on their potential to cause disease. NetDiseaseSNP classifies successfully disease-causing and neutral mutations. In addition, we show that NetDiseaseSNP discriminates cancer driver and passenger mutations satisfactorily. Our method outperforms other state-of-the-art methods on several disease/neutral datasets as well as on cancer driver/passenger mutation datasets and can thus be used to pinpoint and prioritize plausible disease candidates among nsSNPs for further investigation. NetDiseaseSNP is publicly available as an online tool as well as a web service: http://www.cbs.dtu.dk/services/NetDiseaseSNP     

Structural location of disease-associated single-nucleotide polymorphisms

Non-synonymous single-nucleotide polymorphism (nsSNP) of genes introduces amino acid changes to proteins, and plays an important role in providing genetic functional diversity. To understand the structural characteristics of disease-associated SNPs, we have mapped a set of nsSNPs derived from the online mendelian inheritance in man (OMIM) database to the structural surfaces of encoded proteins. These nsSNPs are disease-associated or have distinctive phenotypes. As a control dataset, we mapped a set of nsSNPs derived from SNP database dbSNP to the structural surfaces of those encoded proteins. Using the alpha shape method from computational geometry, we examine the geometric locations of the structural sites of these nsSNPs. We classify each nsSNP site into one of three categories of geometric locations: those in a pocket or a void (type P); those on a convex region or a shallow depressed region (type S); and those that are buried completely in the interior (type I). We find that the majority (88%) of disease-associated nsSNPs are located in voids or pockets, and they are infrequently observed in the interior of proteins (3.2% in the data set). We find that nsSNPs mapped from dbSNP are less likely to be located in pockets or voids (68%). We further introduce a novel application of hidden Markov models (HMM) for analyzing sequence homology of SNPs on various geometric sites. For SNPs on surface pocket or void, we find that there is no strong tendency for them to occur on conserved residues. For SNPs buried in the interior, we find that disease-associated mutations are more likely to be conserved. The approach of classifying nsSNPs with alpha shape and HMM developed in this study can be integrated with additional methods to improve the accuracy of predictions of whether a given nsSNP is likely to be disease-associated.     

ModuleRole: A Tool for Modulization,Role Determination and Visualization in Protein-Protein Interaction Networks

Rapidly increasing amounts of (physical and genetic) protein-protein interaction (PPI) data are produced by various high-throughput techniques, and interpretation of these data remains a major challenge. In order to gain insight into the organization and structure of the resultant large complex networks formed by interacting molecules, using simulated annealing, a method based on the node connectivity, we developed ModuleRole, a user-friendly web server tool which finds modules in PPI network and defines the roles for every node, and produces files for visualization in Cytoscape and Pajek. For given proteins, it analyzes the PPI network from BioGRID database, finds and visualizes the modules these proteins form, and then defines the role every node plays in this network, based on two topological parameters Participation Coefficient and Z-score. This is the first program which provides interactive and very friendly interface for biologists to find and visualize modules and roles of proteins in PPI network. It can be tested online at the website http://www.bioinfo.org/modulerole/index.php, which is free and open to all users and there is no login requirement, with demo data provided by “User Guide” in the menu Help. Non-server application of this program is considered for high-throughput data with more than 200 nodes or user’s own interaction datasets. Users are able to bookmark the web link to the result page and access at a later time. As an interactive and highly customizable application, ModuleRole requires no expert knowledge in graph theory on the user side and can be used in both Linux and Windows system, thus a very useful tool for biologist to analyze and visualize PPI networks from databases such as BioGRID.

Availability

ModuleRole is implemented in Java and C, and is freely available at http://www.bioinfo.org/modulerole/index.php. Supplementary information (user guide, demo data) is also available at this website. API for ModuleRole used for this program can be obtained upon request.     

Functional impacts of non-synonymous single nucleotide polymorphisms: selective constraint and structural environments

In this work, we studied the correlations between selective constraint, structural environments and functional impacts of non-synonymous single nucleotide polymorphisms (nsSNPs). We found that the relation between solvent accessibility and functional impacts of nsSNPs is not as simple as generally thought. Finer structural classifications need to be taken into account to reveal the complex relations between the characteristics of a structure environment and its influence on the functional impacts of nsSNPs. We introduced two parameters for each structural environment, consensus residue percentage and residue distribution distance, to characterize the selective constraint imposed by the environment. Both parameters significantly correlate with the functional bias of nsSNPs across the structural environments. This result shows that selective constraint underlies the bias of a structural environment towards a certain type of nsSNPs (disease-associated or benign).     

Combination use of protein–protein interaction network topological features improves the predictive scores of deleterious non-synonymous single-nucleotide polymorphisms

Single-nucleotide polymorphisms (SNPs) are the most frequent form of genetic variations. Non-synonymous SNPs (nsSNPs) occurring in coding region result in single amino acid substitutions that associate with human hereditary diseases. Plenty of approaches were designed for distinguishing deleterious from neutral nsSNPs based on sequence level information. Novel in this work, combinations of protein–protein interaction (PPI) network topological features were introduced in predicting disease-related nsSNPs. Based on a dataset that was compiled from Swiss-Prot, a random forest model was constructed with an average accuracy value of 80.43 % and an MCC value of 0.60 in a rigorous tenfold crossvalidation test. For an independent dataset, our model achieved an accuracy of 88.05 % and an MCC of 0.67. Compared with previous studies, our approach presented superior prediction ability. Results showed that the incorporated PPI network topological features outperform conventional features. Our further analysis indicated that disease-related proteins are topologically different from other proteins. This study suggested that nsSNPs may share some topological information of proteins and the change of topological attributes could provide clues in illustrating functional shift due to nsSNPs.     

Predicting the phenotypic effects of non-synonymous single nucleotide polymorphisms based on support vector machines

Background  

Human genetic variations primarily result from single nucleotide polymorphisms (SNPs) that occur approximately every 1000 bases in the overall human population. The non-synonymous SNPs (nsSNPs) that lead to amino acid changes in the protein product may account for nearly half of the known genetic variations linked to inherited human diseases. One of the key problems of medical genetics today is to identify nsSNPs that underlie disease-related phenotypes in humans. As such, the development of computational tools that can identify such nsSNPs would enhance our understanding of genetic diseases and help predict the disease.     

Genetic diversity of tyrosine hydroxylase (TH) and dopamine β-hydroxylase (DBH) genes in cattle breeds

DNA from four cattle breeds was used to re-sequence all of the exons and 56% of the introns of the bovine tyrosine hydroxylase (TH) gene and 97% and 13% of the bovine dopamine β-hydroxylase (DBH) coding and non-coding sequences, respectively. Two novel single nucleotide polymorphisms (SNPs) and a microsatellite motif were found in the TH sequences. The DBH sequences contained 62 nucleotide changes, including eight non-synonymous SNPs (nsSNPs) that are of particular interest because they may alter protein function and therefore affect the phenotype. These DBH nsSNPs resulted in amino acid substitutions that were predicted to destabilize the protein structure. Six SNPs (one from TH and five from DBH non-synonymous SNPs) were genotyped in 140 animals; all of them were polymorphic and had a minor allele frequency of > 9%. There were significant differences in the intra- and inter-population haplotype distributions. The haplotype differences between Brahman cattle and the three B. t. taurus breeds (Charolais, Holstein and Lidia) were interesting from a behavioural point of view because of the differences in temperament between these breeds.     

Prediction by graph theoretic measures of structural effects in proteins arising from non-synonymous single nucleotide polymorphisms

Recent analyses of human genome sequences have given rise to impressive advances in identifying non-synonymous single nucleotide polymorphisms (nsSNPs). By contrast, the annotation of nsSNPs and their links to diseases are progressing at a much slower pace. Many of the current approaches to analysing disease-associated nsSNPs use primarily sequence and evolutionary information, while structural information is relatively less exploited. In order to explore the potential of such information, we developed a structure-based approach, Bongo (Bonds ON Graph), to predict structural effects of nsSNPs. Bongo considers protein structures as residue-residue interaction networks and applies graph theoretical measures to identify the residues that are critical for maintaining structural stability by assessing the consequences on the interaction network of single point mutations. Our results show that Bongo is able to identify mutations that cause both local and global structural effects, with a remarkably low false positive rate. Application of the Bongo method to the prediction of 506 disease-associated nsSNPs resulted in a performance (positive predictive value, PPV, 78.5%) similar to that of PolyPhen (PPV, 77.2%) and PANTHER (PPV, 72.2%). As the Bongo method is solely structure-based, our results indicate that the structural changes resulting from nsSNPs are closely associated to their pathological consequences.     

Evidence of Colorectal Cancer-Associated Mutation in MCAK: A Computational Report

Computational prediction of disease-associated non-synonymous polymorphism (nsSNP) has provided a significant platform to filter out the pathological mutations from large pool of SNP datasets at a very low cost input. Several methodologies and complementary protocols have been previously implemented and has provided significant prediction results. Although the previously implicated prediction methods were capable of investigating the most likely deleterious nsSNPs, but due to the lack of genotype–phenotype association analysis, the prediction results lacked in accuracy level. In this work we implemented the computational compilation of protein conformational changes as well as the probable disease-associated phenotypic outcomes. Our result suggested E403K mutation in mitotic centromere-associated kinesin protein as highly damaging and showed strong concordance to the previously observed colorectal cancer mutations aggregation tendency and energy value changes. Moreover, the molecular dynamics simulation results showed major loss in conformation and stability of mutant N-terminal kinesin-like domain structure. The result obtained in this study will provide future prospect of computational approaches in determining the SNPs that may affect the native conformation of protein structure and lead to cancer-associated disorders.     

Geometric De-noising of Protein-Protein Interaction Networks

Understanding complex networks of protein-protein interactions (PPIs) is one of the foremost challenges of the post-genomic era. Due to the recent advances in experimental bio-technology, including yeast-2-hybrid (Y2H), tandem affinity purification (TAP) and other high-throughput methods for protein-protein interaction (PPI) detection, huge amounts of PPI network data are becoming available. Of major concern, however, are the levels of noise and incompleteness. For example, for Y2H screens, it is thought that the false positive rate could be as high as 64%, and the false negative rate may range from 43% to 71%. TAP experiments are believed to have comparable levels of noise.We present a novel technique to assess the confidence levels of interactions in PPI networks obtained from experimental studies. We use it for predicting new interactions and thus for guiding future biological experiments. This technique is the first to utilize currently the best fitting network model for PPI networks, geometric graphs. Our approach achieves specificity of 85% and sensitivity of 90%. We use it to assign confidence scores to physical protein-protein interactions in the human PPI network downloaded from BioGRID. Using our approach, we predict 251 interactions in the human PPI network, a statistically significant fraction of which correspond to protein pairs sharing common GO terms. Moreover, we validate a statistically significant portion of our predicted interactions in the HPRD database and the newer release of BioGRID. The data and Matlab code implementing the methods are freely available from the web site: http://www.kuchaev.com/Denoising.