Abstract Motivation: Microarrays are capable of determining the expression levels thousands genes simultaneously. One important application gene data is classification samples into categories. In combination with methods, this technology can be useful to support clinical management decisions for individual patients, e.g. in oncology. Standard statistic methodologies or prediction do not work well when number variables p (genes) far too exceeds n. So, modification existing statistical...

Identifying protein-protein interactions (PPIs) is essential for elucidating protein functions and understanding the molecular mechanisms inside cell. However, experimental methods detecting PPIs are both time-consuming expensive. Therefore, computational prediction of becoming increasingly popular, which can provide an inexpensive way predicting most likely set at entire proteome scale, be used to complement approaches. Although much progress has already been achieved in this direction,...

A reliable and accurate identification of the type tumors is crucial to proper treatment cancers. In recent years, it has been shown that sparse representation (SR) by l1-norm minimization robust noise, outliers even incomplete measurements, SR successfully used for classification. This paper presents a new SR-based method tumor classification using gene expression data. set metasamples are extracted from training samples, then an input testing sample represented as linear combination these...

Non-negative matrix factorization (NMF) has become one of the most powerful methods for clustering and feature selection. However, performance traditional NMF method severely degrades when data contain noises outliers or manifold structure is not taken into account. In this article, a novel called correntropy-based hypergraph regularized (CHNMF) proposed to solve above problem. Specifically, we use correntropy instead Euclidean norm in loss term CHNMF, which will improve robustness...

A reliable and precise identification of the type tumors is crucial to effective treatment cancer. With rapid development microarray technologies, tumor clustering based on gene expression data becoming a powerful approach cancer class discovery. In this paper, we apply penalized matrix decomposition (PMD) extract metasamples for clustering. The extracted capture inherent structures samples belong same class. At time, PMD factors sample over can be used as its indicator in return. Compared...

Microarray techniques have been used to delineate cancer groups or identify candidate genes for prognosis. As such problems can be viewed as classification ones, various methods applied analyze interpret gene expression data. In this paper, we propose a novel method based on robust principal component analysis (RPCA) classify tumor samples of Firstly, RPCA is utilized highlight the characteristic associated with special biological process. Then, and RPCA+LDA (robust linear discriminant...

Non-negative Matrix Factorization (NMF), a classical method for dimensionality reduction, has been applied in many fields. It is based on the idea that negative numbers are physically meaningless various data-processing tasks. Apart from its contribution to conventional data analysis, recent overwhelming interest NMF due newly discovered ability solve challenging mining and machine learning problems, especially relation gene expression data. This survey paper mainlyfocuses research examining...

Abstract Motivation MicroRNAs (miRNAs) are a class of non-coding RNAs that play critical roles in various biological processes. Many studies have shown miRNAs closely related to the occurrence, development and diagnosis human diseases. Traditional experiments costly time consuming. As result, effective computational models become increasingly popular for predicting associations between diseases, which could effectively boost disease prevention. Results We propose novel framework, called...

Predicting disease-related long non-coding RNAs (lncRNAs) is beneficial to finding of new biomarkers for prevention, diagnosis and treatment complex human diseases. In this paper, we proposed a machine learning techniques-based classification approach identify lncRNAs by graph auto-encoder (GAE) random forest (RF) (GAERF). First, combined the relationship lncRNA, miRNA disease into heterogeneous network. Then, low-dimensional representation vectors nodes were learned from network GAE, which...

Principal component analysis (PCA) has been used to study the pathogenesis of diseases. To enhance interpretability classical PCA, various improved PCA methods have proposed date. Among these, a typical method is so-called sparse which focuses on seeking loadings. However, performance these still far from satisfactory due their limitation using unsupervised learning methods; moreover, class ambiguity within sample high. overcome this problem, paper developed new method, named supervised...

Abstract Background Deep learning algorithms significantly improve the accuracy of pathological image classification, but breast cancer classification using only single-mode images still cannot meet needs clinical practice. Inspired by real scenario pathologists reading for diagnosis, we integrate and structured data extracted from electronic medical record (EMR) to further classification. Methods In this paper, propose a new richer fusion network benign malignant based on multimodal data....

miRNAs belong to small non-coding RNAs that are related a number of complicated biological processes. Considerable studies have suggested closely associated with many human diseases. In this study, we proposed computational model based on Similarity Constrained Matrix Factorization for miRNA-Disease Association Prediction (SCMFMDA). order effectively combine different disease and miRNA similarity data, applied network fusion algorithm obtain integrated (composed functional similarity,...

Abstract The advances in single-cell ribonucleic acid sequencing (scRNA-seq) allow researchers to explore cellular heterogeneity and human diseases at cell resolution. Cell clustering is a prerequisite scRNA-seq analysis since it can recognize identities. However, the high dimensionality, noises significant sparsity of data have made big challenge. Although many methods emerged, they still fail fully intrinsic properties cells relationship among cells, which seriously affects downstream...

Accumulating evidence suggests that circRNAs play crucial roles in human diseases. CircRNA-disease association prediction is extremely helpful understanding pathogenesis, diagnosis, and prevention, as well identifying relevant biomarkers. During the past few years, a large number of deep learning (DL) based methods have been proposed for predicting circRNA-disease achieved impressive performance. However, there are two main drawbacks to these methods. The first underutilize biometric...

Graph convolutional networks (GCNs)-based methods for hyperspectral image (HSI) classification have received more attention due to its flexibility in information aggregation. However, most existing GCN-based HSI community rely on capturing fixed K-hops neighbors feature aggregation, which ignores the inherent imbalance class distributions and fails achieve optimal smoothing through graph convolution operator. It is unreasonable apply strategy imbalanced classes, as regions with rich...

Intracranial aneurysm (IA) is a vascular disease of the brain arteries caused by pathological dilation, which can result in subarachnoid hemorrhage if ruptured. Automatically classification and segmentation intracranial aneurysms are essential for their diagnosis treatment. However, majority current research focused on two-dimensional images, ignoring 3D spatial information that also critical. In this work, we propose novel dual-branch fusion network called Point Cloud Multi-View Medical...

Circular RNAs (circRNAs) play vital roles in transcription and translation. Identification of circRNA-RBP (RNA-binding protein) interaction sites has become a fundamental step molecular cell biology. Deep learning (DL)-based methods have been proposed to predict achieved impressive identification performance. However, those cannot effectively capture long-distance dependencies, utilize the information multiple features. To overcome limitations, we propose DL-based model iCRBP-LKHA using deep...