SLNL: A novel method for gene selection and phenotype classification

Abstract

One of the central tasks of genome research is to predict phenotypes and discover some important gene biomarkers. However, there are three main problems in analyzing genomics data to predict phenotypes and gene marker selection. Such as large p and small n, low reproducibility of the selected biomarkers, and high noise. To provide a unified solution to alleviate the problems as mentioned above, we propose a self‐paced learning L 1 / 2 ${{\rm{L}}}_{1/2}$ absolute network‐based logistic regression model, called SLNL. Through the L 1 / 2 ${L}_{1/2}$ regularization, the model can get a more sparse result, which provides better interpretability. The absolute network‐based penalty enables the model to integrate the feature network knowledge and helps select higher reproducibility genes. Moreover, this proposed penalty overcomes the drawback of a traditional network penalty without considering the sign of the coefficient. By the self‐paced learning strategy, the model can now consider the noise level in gene expression data, lower the impact of high noise samples in data to model training, and provide better prediction accuracy. We compare the proposed method with six alternative approaches in various experimental scenarios, including a comprehensive simulation, four benchmark gene expression datasets, one lung cancer data set, and three lung cancer validation sets. Results show that SLNL can identify fewer meaningful biomarkers and obtain the best or equivalent prediction performance. Moreover, biological analysis shows that the genes selected by the SLNL might be helpful to tumor diagnosis and treatment.

Bitter-RF: A random forest machine model for recognizing bitter peptides

“…However, one may notice that the features were not optimized. In the future, we will use various of feature selection techniques (83)(84)(85)(86) to pick out the best features for improving model's performance.…”

Feature Genes in Neuroblastoma Distinguishing High-Risk and Non-high-Risk Neuroblastoma Patients: Development and Validation Combining Random Forest With Artificial Neural Network

“…All the three classifiers employed in the study are state-of-the-art machine learning techniques that show good performances in various applications. However, some recent machine learning or feature selection methods are not discussed in our study, for example, Huang et al proposed a novel method for gene selection and phenotype classification and an efficient tool for survival analysis and biomarker selection (8,9).…”

“…With the rapid development of bioinformatics technology, we have a deeper understanding of neuroblastoma. A large amount of biological data has exploded, various biological databases have been established, and various prediction models can be established using mathematical knowledge (8)(9)(10)(11). But there are thousands of genetic data, and screening out the signature genes will help us more quickly and easily distinguish between high-risk and non-high-risk neuroblastoma patients.…”

Construction of a miRNA Signature Using Support Vector Machine to Identify Microsatellite Instability Status and Prognosis in Gastric Cancer

“…In recent years, a number of critical discoveries have highlighted the growing interest in understanding the mechanisms of miRNAs. And with the development of artificial intelligence, more new tools have been applied in the life sciences [9,10]. However, the specific role and prog-nostic value of miRNAs in GC have not been clearly elucidated.…”