Multi-scale supervised clustering-based feature selection for tumor classification and identification of biomarkers and targets on genomic data

Abstract

Background The small number of samples and the curse of dimensionality hamper the better application of deep learning techniques for disease classification. Additionally, the performance of clustering-based feature selection algorithms is still far from being satisfactory due to their limitation in using unsupervised learning methods. To enhance interpretability and overcome this problem, we developed a novel feature selection algorithm. In the meantime, complex genomic data brought great challenges for the identification of biomarkers and therapeutic targets. The current some feature selection methods have the problem of low sensitivity and specificity in this field. Results In this article, we designed a multi-scale clustering-based feature selection algorithm named MCBFS which simultaneously performs feature selection and model learning for genomic data analysis. The experimental results demonstrated that MCBFS is robust and effective by comparing it with seven benchmark and six state-of-the-art supervised methods on eight data sets. The visualization results and the statistical test showed that MCBFS can capture the informative genes and improve the interpretability and visualization of tumor gene expression and single-cell sequencing data. Additionally, we developed a general framework named McbfsNW using gene expression data and protein interaction data to identify robust biomarkers and therapeutic targets for diagnosis and therapy of diseases. The framework incorporates the MCBFS algorithm, network recognition ensemble algorithm and feature selection wrapper. McbfsNW has been applied to the lung adenocarcinoma (LUAD) data sets. The preliminary results demonstrated that higher prediction results can be attained by identified biomarkers on the independent LUAD data set, and we also structured a drug-target network which may be good for LUAD therapy. Conclusions The proposed novel feature selection method is robust and effective for gene selection, classification, and visualization. The framework McbfsNW is practical and helpful for the identification of biomarkers and targets on genomic data. It is believed that the same methods and principles are extensible and applicable to other different kinds of data sets.

A novel dissimilarity metric based on feature‐to‐feature scatter frequencies for clustering‐based feature selection in biomedical data

“…The general framework of CBFS is to first group the attributes into a set of distinct clusters of highly correlated features and then select a representative feature from each cluster. Feature selection by feature grouping has recently attracted the interest of researchers in pattern recognition and data mining 33,38‐44 as it is less susceptible to obtaining a suboptimal feature subset. These methods are distinguished by the clustering algorithm, similarity and relevance metric.…”

“…In addition, disregarding the class labels and the loss of information in the final stage are the other drawbacks of this technique. The multiscale clustering‐based feature selection algorithm is a wrapper approach with a novel dissimilarity function which exploits the potential of feature clustering for genomics data analysis 42 . It is in fact a supervised learning method, with a multiscale dissimilarity function formed by multiple distance functions.…”

“…The parameters related to the importance of distance functions and the feature weights are obtained by solving the corresponding optimizing problem. This method, although employing a novel supervised approach for feature clustering, is a wrapper technique that is computationally demanding 42 …”

Analysis on the Penetration of Emotional Education in College Physical Education Based on Emotional Feature Clustering

“…In the past, physical education in colleges and universities mainly focused on competitive minority sports, but now our requirements should develop into popular sports of art and ecological sports so that sports can become a necessity for people to survive. erefore, in college physical education, physical education teachers should add emotional education instead of simply teaching physical education [3], improve interest in learning, harmonize the relationship between teachers and students, form a harmonious teaching atmosphere, and improve teaching quality, which even has an important influence on his whole life. e modern educational process emphasizes the imparting of rational knowledge and neglects the accumulation of emotional experience.…”

MDAKRLS: Predicting human microbe-disease association based on Kronecker regularized least squares and similarities

“…In 2017, Huang et al [ 24 ] presented the NGRHMDA method by integrating two single recommendation methods (graph‑based scoring and neighbor‑based collaborative filtering prediction model), and achieved a good prediction result. With the fast development of machine learning technology [ 25 , 26 ], some machine learning-based methods were also presented for MDAs prediction. For example, in 2017, Wang et al [ 27 ] proposed a semi-supervised method called LRLSHMDA based on the Laplacian regularized least squares method.…”