Discovering biomarkers from gene expression data for predicting cancer subgroups using neural networks and relational fuzzy clustering

Abstract

Background: The four heterogeneous childhood cancers, neuroblastoma, non-Hodgkin lymphoma, rhabdomyosarcoma, and Ewing sarcoma present a similar histology of small round blue cell tumor (SRBCT) and thus often leads to misdiagnosis. Identification of biomarkers for distinguishing these cancers is a well studied problem. Existing methods typically evaluate each gene separately and do not take into account the nonlinear interaction between genes and the tools that are used to design the diagnostic prediction system. Consequently, more genes are usually identified as necessary for prediction. We propose a general scheme for finding a small set of biomarkers to design a diagnostic system for accurate classification of the cancer subgroups. We use multilayer networks with online gene selection ability and relational fuzzy clustering to identify a small set of biomarkers for accurate classification of the training and blind test cases of a well studied data set.

A fuzzy rule based approach to identify biomarkers for diagnostic classification of cancers

“…These genes are listed in Table 1. Note that these 20 genes have many redundant genes as reported in [22].…”

“…At the beginning of training, these gates are kept almost closed, and the training algorithm opens the required gates (allows features to enter the network) depending on the ability of features to reduce the training error. The same set of twenty genes selected by FSMLP in [22] are used here. These genes are listed in Table 1.…”

“…Fuzzy clustering is done on this reduced data set. Such neural networks have also been used to select genes in [22]. We use a modified multilayered perceptron (MLP) network [11] with online feature selection capability.…”

“…In [22] we have proposed a method that uses neural networks and relational fuzzy clustering to discern a small set of genes with adequate discriminative power. These selected genes can be used to design useful classifiers using several machine learning tools.…”

Bioinformatics and Biostatistics in Mining Epigenetic Disease Markers and Targets

“…Pal et al proposed a scheme that integrated neural network classifiers and fuzzy clustering for identifying a small set of biomarkers to distinguish four usually misdiagnosed childhood cancers. They have shown that their approach was capable of identifying a very small set of genes as biomarkers with high performance, by taking the interaction between genes into consideration [49]. In another study, Chambwe et al performed unsupervised hierarchical clustering on genome-wide DNA methylation data of 140 DLBCL samples and ten normal germinal center B cells.…”