Search

Teratogenicity is one of the main concerns in clinical medications of pregnant women. Prescription of antiseizure medications (ASMs) in women with epilepsy during pregnancy may cause teratogenic effects on the fetus. Although large scale epilepsy pregnancy registries played an important role in evaluating the teratogenic risk of ASMs, for most ASMs, especially the newly approved ones, the potential teratogenic risk cannot be effectively assessed due to the lack of evidence. In this study, the analyses are performed on any medication, with a focus on ASMs. We curated a list containing the drugs with potential teratogenicity based on the US Food and Drug Administration (FDA)-approved drug labeling, and established a support vector machine (SVM) model for detecting drugs with high teratogenic risk. The model was validated by using the post-marketing surveillance data from US FDA Spontaneous Adverse Events Reporting System (FAERS) and applied to the prediction of potential teratogenic risk of ASMs. Our results showed that our proposed model outperformed the state-of-art approaches, including logistic regression (LR), random forest (RF) and extreme gradient boosting (XGBoost), when detecting the high teratogenic risk of drugs (MCC and recall rate were 0.312 and 0.851, respectively). Among 196 drugs with teratogenic potential reported by FAERS, 136 (69.4%) drugs were correctly predicted. For the eight commonly used ASMs, 4 of them were predicted as high teratogenic risk drugs, including topiramate, phenobarbital, valproate and phenytoin (predicted probabilities of teratogenic risk were 0.69, 0.60 0.59, and 0.56, respectively), which were consistent with the statement in FDA-approved drug labeling and the high reported prevalence of teratogenicity in epilepsy pregnancy registries. In addition, the structural alerts in ASMs that related to the genotoxic carcinogenicity and mutagenicity, idiosyncratic adverse reaction, potential electrophilic agents and endocrine disruption were identified and discussed. Our findings can be a good complementary for the teratogenic risk assessment in drug development and facilitate the determination of pharmacological therapies during pregnancy. [ABSTRACT FROM AUTHOR]

Background: The aim of gene expression-based clinical modelling in tumorigenesis is not only to accurately predict the clinical endpoints, but also to reveal the genome characteristics for downstream analysis for the purpose of understanding the mechanisms of cancers. Most of the conventional machine learning methods involved a gene filtering step, in which tens of thousands of genes were firstly filtered based on the gene expression levels by a statistical method with an arbitrary cutoff. Although gene filtering procedure helps to reduce the feature dimension and avoid overfitting, there is a risk that some pathogenic genes important to the disease will be ignored. Results: In this study, we proposed a novel deep learning approach by combining a convolutional neural network with stationary wavelet transform (SWT-CNN) for stratifying cancer patients and predicting their clinical outcomes without gene filtering based on tumor genomic profiles. The proposed SWT-CNN overperformed the state-of-art algorithms, including support vector machine (SVM) and logistic regression (LR), and produced comparable prediction performance to random forest (RF). Furthermore, for all the cancer types, we firstly proposed a method to weight the genes with the scores, which took advantage of the representative features in the hidden layer of convolutional neural network, and then selected the prognostic genes for the Cox proportional-hazards regression. The results showed that risk stratifications can be effectively improved by using the identified prognostic genes as feature, indicating that the representative features generated by SWT-CNN can well correlate the genes with prognostic risk in cancers and be helpful for selecting the prognostic gene signatures. Conclusions: Our results indicated that gene expression-based SWT-CNN model can be an excellent tool for stratifying the prognostic risk for cancer patients. In addition, the representative features of SWT-CNN were validated to be useful for evaluating the importance of the genes in the risk stratification and can be further used to identify the prognostic gene signatures. [ABSTRACT FROM AUTHOR]

Background: Computational prediction of inhibition efficiency (IE) for inhibitor molecules is a crucial supplementary way to design novel molecules that can efficiently inhibit corrosion onto metallic surfaces. Purpose: Here we are dedicated to developing a new machine learning-based predictor for the inhibition efficiency (IE) of benzimidazole derivatives. Methods: First, a comprehensively numerical representation was given on inhibitor molecules from all aspects of energy, electronic, topological, physicochemical and spatial properties based on 3-D structures and 150 valid structural descriptors were obtained. Then, a thorough investigation of these structural descriptors was implemented. The multicollinearity-based clustering analysis was performed to remove the linear correlated feature variables, so 47 feature clusters were produced. Meanwhile, Gini importance by random forest (RF) was used to further measure the contributions of the descriptors in each cluster and 47 non-linear descriptors were selected with the highest Gini importance score in the corresponding cluster. Further, considering the limited number of available inhibitors, different feature subsets were constructed according to the Gini importance score ranking list of 47 descriptors. Results: Finally, support vector machine (SVM) models based on different feature subsets were tested by leave-one-out cross validation. Through comparisons, the optimal SVM model with the top 11 descriptors was achieved based on Poly kernel. This model yields a promising performance with the correlation coefficient (R) and root-mean-square error (RMSE) of 0.9589 and 4.45, respectively, which indicates that the method proposed by us gives the best performance for the current data. Conclusion: Based on our model, 6 new benzimidazole molecules were designed and their IE values predicted by this model indicate that two of them have high potential as outstanding corrosion inhibitors. [ABSTRACT FROM AUTHOR]

Lipid–protein interactions play a vital role in various biological processes, which are involved in cellular functions and can affect the stability, folding and the function of peptides and proteins. In this study, a sequence-based method by using support vector machine and position specific scoring matrix (PSSM) was proposed to predict lipid-binding sites. Considering the influence of surrounding residues of one amino acid, a sliding window was chosen to encode the PSSM profiles. By incorporating the evolutionary information and the local features of residues surrounding one lipid-binding site, the method yielded a high accuracy of 80.86% and the Matthew’s Correlation Coefficient of 0.58 by using fivefold cross validation test. The good result indicates the applicability of the method. [ABSTRACT FROM AUTHOR]

Protein methylation plays vital roles in many biological processes and has been implicated in various human diseases. To fully understand the mechanisms underlying methylation for use in drug design and work in methylation-related diseases, an initial but crucial step is to identify methylation sites. The use of high-throughput bioinformatics methods has become imperative to predict methylation sites. In this study, we developed a novel method that is based only on sequence conservation to predict protein methylation sites. Conservation difference profiles between methylated and non-methylated peptides were constructed by the information entropy (IE) in a wider neighbor interval around the methylation sites that fully incorporated all of the environmental information. Then, the distinctive neighbor residues were identified by the importance scores of information gain (IG). The most representative model was constructed by support vector machine (SVM) for Arginine and Lysine methylation, respectively. This model yielded a promising result on both the benchmark dataset and independent test set. The model was used to screen the entire human proteome, and many unknown substrates were identified. These results indicate that our method can serve as a useful supplement to elucidate the mechanism of protein methylation and facilitate hypothesis-driven experimental design and validation. [ABSTRACT FROM AUTHOR]

Abstract: The combinations of NIR spectroscopy and three classification algorithms, i.e., multi-class support vector machine (BSVM), k-nearest neighbor (KNN) and soft independent modeling of class analogies (SIMCA), for discriminating different brands of cigarettes, were explored. The influence of the training set size on the relative performance of each algorithm was also investigated. A NIR spectral dataset involving the classification of cigarettes of three brands was used for illustration. Three performance criteria based on “correctly classified rate (CCR)”, i.e., “Average CCR”, “95 percentile of CCR” and “S.D. of CCR”, were defined to compare different algorithms. It was revealed that BSVM is significantly better than KNN or SIMCA in the statistical sense, especially in cases where the training set is relatively small. The results suggest that NIR spectroscopy together with BSVM could be an alternative to traditional methods for discriminating different brands of cigarettes. [Copyright &y& Elsevier]

Abstract: Hot spots residues in protein–protein interface play crucial roles in protein binding. In the present study, complex network method was applied to uncover influence of neighboring residues on hot spots and then several network and microenvironment features were designed to describe the diversity of environment of hot spots. After feature analysis by permutation importance in Random Forest (RF), an optimal 58-dimensional feature set including ten network and microenvironment features was selected and then applied to construct a Support Vector Machine (SVM) prediction model for hot spots. A satisfactory accuracy (ACC) value of 79.0% and a Mathew's correlation coefficient (MCC) value of 0.470 were obtained for independent test set. The novel network features and microenvironment features were proved to be promising in discovering hot spots in interfaces. A further microenvironment analysis was also performed. Amino acid residues directly contacting with hot spots in residue–residue interaction network exhibit significant importance for the microenvironment of hot spots. Amino acid alanine (A), aspartic acid (D), glycine (G), histidine (H), isoleucine (I), asparagine (N), serine (S) and tyrosine (Y) are more likely to occur in the vicinity of hot spots than in the vicinity of non-hot spots. These amino acid residues probably cluster together to construct a proper microenvironment for hot spots. [Copyright &y& Elsevier]

Abstract: Flavin mono-nucleotide (FMN) closely evolves in many biological processes. In this study, a computational method was proposed to identify FMN binding sites based on amino acid sequences of proteins only. A modified Position Specific Score Matrix was used to characterize the local environmental sequence information, and a visible improvement of performance was obtained. Also, the ensemble SVM was applied to solve the imbalanced data problem. Additionally, an independent dataset was built to evaluate the practical performance of the method, and a satisfactory accuracy of 87.87% was achieved. It demonstrates that the method is effective in predicting FMN-binding sites. [Copyright &y& Elsevier]

Abstract: In proteins, the number of interacting pairs is usually much smaller than the number of non-interacting ones. So the imbalanced data problem will arise in the field of protein–protein interactions (PPIs) prediction. In this article, we introduce two ensemble methods to solve the imbalanced data problem. These ensemble methods combine the based-cluster under-sampling technique and the fusion classifiers. And then we evaluate the ensemble methods using a dataset from Database of Interacting Proteins (DIP) with 10-fold cross validation. All the prediction models achieve area under the receiver operating characteristic curve (AUC) value about 95%. Our results show that the ensemble classifiers are quite effective in predicting PPIs; we also gain some valuable conclusions on the performance of ensemble methods for PPIs in imbalanced data. The prediction software and all dataset employed in the work can be obtained for free at http://cic.scu.edu.cn/bioinformatics/Ensemble_PPIs/index.html. [Copyright &y& Elsevier]

Abstract: Secreted proteins play pivotal biological regulatory roles in eukaryotic cells with different functions, and have the potential for disease biomarkers and protein therapeutics. Furthermore, the comprehension of their functions is not only indispensable for helping genome annotation, but also a supplement to the existing methods. As the rapid increase of protein sequences generated in the post-genomic age, it is urgent to develop a computational method to effectively annotate the function types of numerous newly discovered secreted proteins. In view of this, a support vector machine (SVM)-based predictor is proposed to classify secreted proteins with different functions in this paper, including cytokine, hormone, immune system protein, protease, and protease inhibitor. Here, proteins are represented by position specific scoring matrix (PSSM) and auto covariance (AC), which incorporates the evolution and the sequence-order information of proteins. When distinguishing the five types of secreted proteins, an accuracy of 83.2%, 88.9%, 86.1%, 90.9%, and 90.6% is achieved for cytokine, hormone, immune system protein, protease and protease inhibitor, respectively. Particularly, when performed on an independent test set of 325 proteins, the method also yielded a satisfactory accuracy of 91.2%. It shows that this method can be a complementary tool for identifying different functions of secreted proteins. The code and all datasets used in this article are freely available at http://cic.scu.edu.cn/bioinformatics/fcSecretP.zip. [Copyright &y& Elsevier]

Abstract: A three-stage support vector machine (SVM) was constructed to predict the state of cysteines by fusing sequence information, evolution information and annotation information of protein sequences. The first and second stages were for predicting whether the protein sequences contain disulfide bonds and whether all of the cysteines are involved in disulfide bonds. In the last stage, one SVM was constructed for predicting which cysteines are involved in disulfide bonds, among all these cysteines in proteins. The three SVMs give a good performance and the overall prediction accuracy are 90.05%, 96.36% and 80.00%, respectively, which indicates that the features selected in this work are effective for predicting the state of cysteines. In addition, current methods only paid too much attention to the prediction performance and never showed us how much important the roles of these features played in the prediction. As a result a feature importance measurement designated as F-score function was used to evaluate these features. The result shows that among these protein descriptors; evolution information is the most important feature for representing the disulfide-containing proteins. The prediction software and data sets used in this article are freely available at http://cic.scu.edu.cn/bioinformatics/Predict_Cys.zip. [ABSTRACT FROM AUTHOR]

Abstract: Protein secretion plays an important role in bacterial lifestyles. Secreted proteins are crucial for bacterial pathogenesis by making bacteria interact with their environments, particularly delivering pathogenic and symbiotic bacteria into their eukaryotic hosts. Therefore, identification of bacterial secreted proteins becomes an important process for the study of various diseases and the corresponding drugs. In this paper, fusing several new features into Chou’s pseudo-amino acid composition (PseAAC), two support vector machine (SVM)-based ternary classifiers are developed to predict secreted proteins of Gram-negative and Gram-positive bacteria. For the two types of bacteria, the high accuracy of 94.03% and 94.36% are obtained in distinguishing classically secreted, non-classically secreted and non-secreted proteins by our method. In order to compare the practical ability of our method in identifying bacterial secreted proteins with those of six published methods, proteins in Escherichia coli and Bacillus subtilis are collected to construct the test sets of Gram-negative and Gram-positive bacteria, and the prediction results of our method are comparable to those of existing methods. When performed on two public independent data sets for predicting NCSPs, it also yields satisfactory results for Gram-negative bacterial proteins. The prediction server SecretP can be accessed at http://cic.scu.edu.cn/bioinformatics/secretPV2/index.htm. [ABSTRACT FROM AUTHOR]

Abstract: In contrast to a large number of classically secreted proteins (CSPs) and non-secreted proteins (NSPs), only a few proteins have been experimentally proved to enter non-classical secretory pathways. So it is difficult to identify non-classically secreted proteins (NCSPs), and no methods are available for distinguishing the three types of proteins simultaneously. In order to solve this problem, a data mining has been taken firstly, and mammalian proteins exported via ER-Golgi-independent pathways are collected through extensive literature searches. In this paper, a support vector machine (SVM)-based ternary classifier named SecretP is proposed to predict mammalian secreted proteins by using pseudo-amino acid composition (PseAA) and five additional features. When distinguishing the three types of proteins, SecretP yielded an accuracy of 88.79%. Evaluating the performance of our method by an independent test set of 92 human proteins, 76 of them are correctly predicted as NCSPs. When performed on another public independent data set, the prediction result of SecretP is comparable to those of other existing computational methods. Therefore, SecretP can be a useful supplementary tool for future secretome studies. The web server SecretP and all supplementary tables listed in this paper are freely available at http://cic.scu.edu.cn/bioinformatics/secretp/index.htm. [Copyright &y& Elsevier]

• Applied the machine learning methods for mining structure information of infrared spectra. • Compared the algorithm perforamance on functional group inference task. • Constructed accurate models to infer the presence/absence of 16 functional groups. The support vector machine was used as a tool for mining the structure information of infrared spectra. A total set of 823 compounds from the OMNIC Fourier transform infrared database was used for training and testing with respect to the presence or absence of 16 functional groups. The results were evaluated by a variety of indices and compared favorably with those obtained by using artificial neural networks method. The trained support vector machines can identify the 16 functional groups with an average prediction accuracy of 93.3% for the presence and 99.0% for the absence of functionalities. [ABSTRACT FROM AUTHOR]