Search

Identifying essential genes in comparison states (EGS) is vital to understanding cell differentiation, performing drug discovery, and identifying disease causes. Here, we present a machine learning method termed Prediction of Essential Genes in Comparison States (PreEGS). To capture the alteration of the network in comparison states, PreEGS extracts topological and gene expression features of each gene in a five-dimensional vector. PreEGS also recruits a positive sample expansion method to address the problem of unbalanced positive and negative samples, which is often encountered in practical applications. Different classifiers are applied to the simulated datasets, and the PreEGS based on the random forests model (PreEGSRF) was chosen for optimal performance. PreEGSRF was then compared with six other methods, including three machine learning methods, to predict EGS in a specific state. On real datasets with four gene regulatory networks, PreEGSRF predicted five essential genes related to leukemia and five enriched KEGG pathways. Four of the predicted essential genes and all predicted pathways were consistent with previous studies and highly correlated with leukemia. With high prediction accuracy and generalization ability, PreEGSRF is broadly applicable for the discovery of disease-causing genes, driver genes for cell fate decisions, and complex biomarkers of biological systems.

Every performance, in an officially sanctioned meet, by a registered USA swimmer is recorded into an online database with times dating back to 1980. For the first time, statistical analysis and machine learning methods are systematically applied to 4,022,631 swim records. In this study, we investigate performance features for all strokes as a function of age and gender. The variances in performance of males and females for different ages and strokes were studied, and the correlations of performances for different ages were estimated using the Pearson correlation. Regression analysis show the performance trends for both males and females at different ages and suggest critical ages for peak training. Moreover, we assess twelve popular machine learning methods to predict or classify swimmer performance. Each method exhibited different strengths or weaknesses in different cases, indicating no one method could predict well for all strokes. To address this problem, we propose a new method by combining multiple inference methods to derive Wisdom of Crowd Classifier (WoCC). Our simulation experiments demonstrate that the WoCC is a consistent method with better overall prediction accuracy. Our study reveals several new age-dependent trends in swimming and provides an accurate method for classifying and predicting swimming times.

A protein contact map is a simplified, two-dimensional version of the three-dimensional protein structure. Protein contact map is proved to be crucial in forming the three-dimensional structure. Contact map prediction has now become an indispensable and promising intermediate step towards final three-dimensional structure prediction, while directed sequence-structure prediction hits its bottlenecks. In this article, different evaluation scores of prediction efficiency are compared. Next, the state of the art and future perspectives of contact map methods are reviewed and special attention is paid to those relying on machine learning algorithms. Details of neural network based methods as well as a list of machine learning based methods are given. Finally, bottlenecks and potential improvements of contact map predictions are discussed.

Every performance, in an officially sanctioned meet, by a registered USA swimmer is recorded into an online database with times dating back to 1980. For the first time, statistical analysis and machine learning methods are systematically applied to 4,022,631 swim records. In this study, we investigate performance features for all strokes as a function of age and gender. The variances in performance of males and females for different ages and strokes were studied, and the correlations of performances for different ages were estimated using the Pearson correlation. Regression analysis show the performance trends for both males and females at different ages and suggest critical ages for peak training. Moreover, we assess twelve popular machine learning methods to predict or classify swimmer performance. Each method exhibited different strengths or weaknesses in different cases, indicating no one method could predict well for all strokes. To address this problem, we propose a new method by combining multiple inference methods to derive Wisdom of Crowd Classifier (WoCC). Our simulation experiments demonstrate that the WoCC is a consistent method with better overall prediction accuracy. Our study reveals several new age-dependent trends in swimming and provides an accurate method for classifying and predicting swimming times.

Network alignment is an important bridge to understanding human protein-protein interactions (PPIs) and functions through model organisms. However, the underlying subgraph isomorphism problem complicates and increases the time required to align protein interaction networks (PINs). Parallel computing technology is an effective solution to the challenge of aligning large-scale networks via sequential computing. In this study, the typical Hungarian-Greedy Algorithm (HGA) is used as an example for PIN alignment. The authors propose a HGA with 2-nearest neighbours (HGA-2N) and implement its graphics processing unit (GPU) acceleration. Numerical experiments demonstrate that HGA-2N can find alignments that are close to those found by HGA while dramatically reducing computing time. The GPU implementation of HGA-2N optimises the parallel pattern, computing mode and storage mode and it improves the computing time ratio between the CPU and GPU compared with HGA when large-scale networks are considered. By using HGA-2N in GPUs, conserved PPIs can be observed, and potential PPIs can be predicted. Among the predictions based on 25 common Gene Ontology terms, 42.8% can be found in the Human Protein Reference Database. Furthermore, a new method of reconstructing phylogenetic trees is introduced, which shows the same relationships among five herpes viruses that are obtained using other methods.

Phishing attacks are evolving with more sophisticated techniques, posing significant threats. Considering the potential of machine-learning-based approaches, our research presents a similar modern approach for web phishing detection by applying powerful machine learning algorithms. An efficient layered classification model is proposed to detect websites based on their URL structure, text, and image features. Previously, similar studies have used machine learning techniques for URL features with a limited dataset. In our research, we have used a large dataset of 20,000 website URLs, and 22 salient features from each URL are extracted to prepare a comprehensive dataset. Along with this, another dataset containing website text is also prepared for NLP-based text evaluation. It is seen that many phishing websites contain text as images, and to handle this, the text from images is extracted to classify it as spam or legitimate. The experimental evaluation demonstrated efficient and accurate phishing detection. Our layered classification model uses support vector machine (SVM), XGBoost, random forest, multilayer perceptron, linear regression, decision tree, naïve Bayes, and SVC algorithms. The performance evaluation revealed that the XGBoost algorithm outperformed other applied models with maximum accuracy and precision of 94% in the training phase and 91% in the testing phase. Multilayer perceptron also worked well with an accuracy of 91% in the testing phase. The accuracy results for random forest and decision tree were 91% and 90%, respectively. Logistic regression and SVM algorithms were used in the text-based classification, and the accuracy was found to be 87% and 88%, respectively. With these precision values, the models classified phishing and legitimate websites very well, based on URL, text, and image features. This research contributes to early detection of sophisticated phishing attacks, enhancing internet user security. [ABSTRACT FROM AUTHOR]

The uneven distribution of sample points is a common problem in medical datasets. How to improve the classification accuracy with these datasets remains to be solved. Based on the density-based spatial clustering of applications with noise (DBSCAN) algorithm, a weight learning approach is proposed to utilize the density information of datasets for the accurate prediction of cardiovascular diseases (CVDs). The approach selects important features by the random forest (RF) algorithm, divides the sample points into three types and weights them using different values by weight learning based on the density. Thus, the constructed machine learning models that combine the original features and weight feature can learn density information, more effectively identify decision boundaries, and achieve better performance. Compared with conventional machine learning models, the cross-validation approach showed that the performance of machine learning models with weight learning could achieve improved accuracy by 3 percentage points with the Stroke dataset and more than 10 percentage points with the University of California, Irvine (UCI) dataset. [ABSTRACT FROM AUTHOR]

A cellular automata (CA) method assisted by a back-propagation neural network (BPNN), named CA-BPNN, is proposed to simulate grain and pore growth. First, CA-BPNN uses the BPNN to detect the relations between porosity and solidification parameters and then uses the relations to establish extra transformation rules of pore growth, which are finally implemented on A356 alloy directly. Compared with the computational results, the shapes and volume fraction of pores from experimental observation are consistent. CA-BPNN can reduce the difficulty of simulating the whole solidification process of casting without solving the high-dimensional continuous governing equations of porosity. This work can be further extended to other useful industrial alloys, such as aluminum alloys and various grades of industrial steels, if the experimental data sets are improved and other machine learning algorithms are introduced. [ABSTRACT FROM AUTHOR]

An extensive, in-depth study of diabetes risk factors (DBRF) is of crucial importance to prevent (or reduce) the chance of suffering from type 2 diabetes (T2D). Accumulation of electronic health records (EHRs) makes it possible to build nonlinear relationships between risk factors and diabetes. However, the current DBRF researches mainly focus on qualitative analyses, and the inconformity of physical examination items makes the risk factors likely to be lost, which drives us to study the novel machine learning approach for risk model development. In this paper, we use Bayesian networks (BNs) to analyze the relationship between physical examination information and T2D, and to quantify the link between risk factors and T2D. Furthermore, with the quantitative analyses of DBRF, we adopt EHR and propose a machine learning approach based on BNs to predict the risk of T2D. The experiments demonstrate that our approach can lead to better predictive performance than the classical risk model. [ABSTRACT FROM AUTHOR]

The major histocompatibility complex class II (MHC-II) molecule plays a crucial role in immunology. Computational prediction of MHC-II binding peptides can help researchers understand the mechanism of immune systems and design vaccines. Most of the prediction algorithms for MHC-II to date have made large efforts in human leukocyte antigen (HLA, the name of MHC in Human) molecules encoded in the DR locus. However, HLA DQ molecules are equally important and have only been made less progress because it is more difficult to handle them experimentally. In this study, we propose an artificial neural network-based approach called MHC2NNZ to predict peptides binding to HLA DQ molecules. Unlike previous artificial neural network-based methods, MHC2NNZ not only considers sequence similarity features but also captures the chemical and physical properties, and a novel method incorporating these properties is proposed to represent peptide flanking regions (PFR). Furthermore, MHC2NNZ improves the prediction accuracy by combining with amino acid preference at more specific positions of the peptides binding core. By evaluating on 3549 peptides binding to six most frequent HLA DQ molecules, MHC2NNZ is demonstrated to outperform other state-of-the-art MHC-II prediction methods. [ABSTRACT FROM AUTHOR]