Semi-supervised Ensemble Learning for Efficient Cancer Sample Classification from miRNA Gene Expression Data

Low-Rank Representation (LRR) is a powerful subspace clustering method because of its successful learning of low-dimensional subspace of data. With the breakthrough of “OMics” technology, many LRR-based methods have been proposed and used to cancer clustering based on gene expression data. Moreover, studies have shown that besides gene expression data, some other genomic data in TCGA also contain important information for cancer research. Therefore, these genomic data can be integrated as a comprehensive feature source for cancer clustering. How to establish an effective clustering model for comprehensive analysis of integrated TCGA data has become a key issue. In this paper, we develop the traditional LRR method and propose a novel method named Block-constraint Laplacian-Regularized Low-Rank Representation (BLLRR) to model multigenome data for cancer sample clustering. The proposed method is dedicated to extracting more abundant subspace structure information from multiple genomic data to improve the accuracy of cancer sample clustering. Considering the heterogeneity of different genome data, we introduce the block-constraint idea into our method. In BLLRR decomposition, we treat each genome data as a data block and impose different constraints on different data blocks. In addition, graph Laplacian is also introduced into our method to better learn the topological structure of data by preserving the local geometric information. The experiments demonstrate that the BLLRR method can effectively analyze integrated TCGA data and extract more subspace structure information from multigenome data. It is a reliable and efficient clustering algorithm for cancer sample clustering.

Download Full-text

Network-based cancer genomic data integration for pattern discovery

BMC Genomic Data ◽

10.1186/s12863-021-01004-y ◽

2021 ◽

Vol 22 (S1) ◽

Author(s):

Fangfang Zhu ◽

Jiang Li ◽

Juan Liu ◽

Wenwen Min

Keyword(s):

Gene Expression ◽

Gene Expression Data ◽

Expression Profiles ◽

Mirna Gene ◽

Interaction Network ◽

Gene Interaction ◽

Functional Modules ◽

Expression Data ◽

Gene Interaction Network ◽

Gene Modules

Abstract Background Since genes involved in the same biological modules usually present correlated expression profiles, lots of computational methods have been proposed to identify gene functional modules based on the expression profiles data. Recently, Sparse Singular Value Decomposition (SSVD) method has been proposed to bicluster gene expression data to identify gene modules. However, this model can only handle the gene expression data where no gene interaction information is integrated. Ignoring the prior gene interaction information may produce the identified gene modules hard to be biologically interpreted. Results In this paper, we develop a Sparse Network-regularized SVD (SNSVD) method that integrates a prior gene interaction network from a protein protein interaction network and gene expression data to identify underlying gene functional modules. The results on a set of simulated data show that SNSVD is more effective than the traditional SVD-based methods. The further experiment results on real cancer genomic data show that most co-expressed modules are not only significantly enriched on GO/KEGG pathways, but also correspond to dense sub-networks in the prior gene interaction network. Besides, we also use our method to identify ten differentially co-expressed miRNA-gene modules by integrating matched miRNA and mRNA expression data of breast cancer from The Cancer Genome Atlas (TCGA). Several important breast cancer related miRNA-gene modules are discovered. Conclusions All the results demonstrate that SNSVD can overcome the drawbacks of SSVD and capture more biologically relevant functional modules by incorporating a prior gene interaction network. These identified functional modules may provide a new perspective to understand the diagnostics, occurrence and progression of cancer.

Download Full-text

Using Rule-Based Machine Learning for Candidate Disease Gene Prioritization and Sample Classification of Cancer Gene Expression Data

PLoS ONE ◽

10.1371/journal.pone.0039932 ◽

2012 ◽

Vol 7 (7) ◽

pp. e39932 ◽

Cited By ~ 54

Author(s):

Enrico Glaab ◽

Jaume Bacardit ◽

Jonathan M. Garibaldi ◽

Natalio Krasnogor

Keyword(s):

Gene Expression ◽

Machine Learning ◽

Gene Expression Data ◽

Disease Gene ◽

Cancer Gene ◽

Expression Data ◽

Rule Based ◽

Sample Classification ◽

Disease Gene Prioritization

Download Full-text

An ensemble learning approach for modeling the systems biology of drug-induced injury

Biology Direct ◽

10.1186/s13062-020-00288-x ◽

2021 ◽

Vol 16 (1) ◽

Author(s):

Joaquim Aguirre-Plans ◽

Janet Piñero ◽

Terezinha Souza ◽

Giulia Callegaro ◽

Steven J. Kunnen ◽

...

Keyword(s):

Gene Expression ◽

Adverse Reaction ◽

Ensemble Learning ◽

Gene Expression Data ◽

Drug Target ◽

Learning Approach ◽

Expression Data ◽

Drug Induced ◽

Chemical Structures

Abstract Background Drug-induced liver injury (DILI) is an adverse reaction caused by the intake of drugs of common use that produces liver damage. The impact of DILI is estimated to affect around 20 in 100,000 inhabitants worldwide each year. Despite being one of the main causes of liver failure, the pathophysiology and mechanisms of DILI are poorly understood. In the present study, we developed an ensemble learning approach based on different features (CMap gene expression, chemical structures, drug targets) to predict drugs that might cause DILI and gain a better understanding of the mechanisms linked to the adverse reaction. Results We searched for gene signatures in CMap gene expression data by using two approaches: phenotype-gene associations data from DisGeNET, and a non-parametric test comparing gene expression of DILI-Concern and No-DILI-Concern drugs (as per DILIrank definitions). The average accuracy of the classifiers in both approaches was 69%. We used chemical structures as features, obtaining an accuracy of 65%. The combination of both types of features produced an accuracy around 63%, but improved the independent hold-out test up to 67%. The use of drug-target associations as feature obtained the best accuracy (70%) in the independent hold-out test. Conclusions When using CMap gene expression data, searching for a specific gene signature among the landmark genes improves the quality of the classifiers, but it is still limited by the intrinsic noise of the dataset. When using chemical structures as a feature, the structural diversity of the known DILI-causing drugs hampers the prediction, which is a similar problem as for the use of gene expression information. The combination of both features did not improve the quality of the classifiers but increased the robustness as shown on independent hold-out tests. The use of drug-target associations as feature improved the prediction, specially the specificity, and the results were comparable to previous research studies.

Download Full-text