Subset Of Relevant Genes Research Articles

Optimizing cancer classification: a hybrid RDO-XGBoost approach for feature selection and predictive insights

The identification of relevant biomarkers from high-dimensional cancer data remains a significant challenge due to the complexity and heterogeneity inherent in various cancer types. Conventional feature selection methods often struggle to effectively navigate the vast solution space while maintaining high predictive accuracy. In response to these challenges, we introduce a novel feature selection approach that integrates Random Drift Optimization (RDO) with XGBoost, specifically designed to enhance the performance of cancer classification tasks. Our proposed framework not only improves classification accuracy but also offers valuable insights into the underlying biological mechanisms driving cancer progression. Through comprehensive experiments conducted on real-world cancer datasets, including Central Nervous System (CNS), Leukemia, Breast, and Ovarian cancers, we demonstrate the efficacy of our method in identifying a smaller subset of unique and relevant genes. This selection results in significantly improved classification efficiency and accuracy. When compared with popular classifiers such as Support Vector Machine, K-Nearest Neighbor, and Naive Bayes, our approach consistently outperforms these models in terms of both accuracy and F-measure metrics. For instance, our framework achieved an accuracy of 97.24% in the CNS dataset, 99.14% in Leukemia, 95.21% in Ovarian, and 87.62% in Breast cancer, showcasing its robustness and effectiveness across different types of cancer data. These results underline the potential of our RDO-XGBoost framework as a promising solution for feature selection in cancer data analysis, offering enhanced predictive performance and valuable biological insights.

AI-driven transcriptomic encoders: From explainable models to accurate, sample-independent cancer diagnostics

In the rapidly evolving domain of medical technology, the utilization of sophisticated algorithms for deciphering transcriptional data has emerged as a critical aspect, especially in the oncology sector. These algorithms, drawing upon methodologies from fields such as natural language processing and advanced image analysis, can significantly enhance the accuracy in predicting cancer-related molecular states. Notably, Transformer models, renowned for their proficiency in handling extensive datasets, are now being adapted for breakthroughs in medical diagnostics or in stratifying patients according to prognostic levels. Our study contributes to the field of precision medicine by integrating Transformer-based learning, exemplified by the Geneformer model, with explainable AI techniques. These techniques are employed to find out the input variables (genes resulting from genomic transcription) most correlated with the decisions of neural network systems. This insight, a key goal in genomic research, aims to select the most relevant gene subset for each specific task in which a neural network is employed. This selection approach has proven to be effective in two classification tasks: cell type classification and breast cancer type classification. Such effectiveness has been demonstrated even across various cohorts of patients. When applying Geneformer-like architecture analyses solely to the selected gene subsets, the outcomes either maintain their accuracy or significantly improve. This approach, aims not only to contribute to the identification of vital genetic markers in cancer genomics, but also to exemplify the adaptability of AI models to different datasets, marking a significant step towards the development of accurate and universally applicable diagnostic tools for precision medicine.

Database-assisted screening of autism spectrum disorder related gene set

Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by social and communication difficulties, along with repetitive behaviors. While genetic factors play a significant role in ASD, the precise genetic landscape remains complex and not fully understood, particularly in non-syndromic cases. The study performed an in silico comparison of three genetic databases. ClinVar, SFARI Gene, and AutDB were utilized to identify relevant gene subset and genetic variations associated with non-syndromic ASD. Gene set enrichment analysis (GSEA) and protein–protein interaction (PPI) network analysis were conducted to elucidate the biological significance of the identified genes. The integrity of ASD-related gene subset and the distribution of their variations were statistically assessed. A subset of twenty overlapping genes potentially specific for non-syndromic ASD was identified. GSEA revealed enrichment of biological processes related to neuronal development and differentiation, synaptic function, and social skills, highlighting their importance in ASD pathogenesis. PPI network analysis demonstrated functional relationships among the identified genes. Analysis of genetic variations showed predominance of rare variants and database-specific distribution patterns. The results provide valuable insights into the genetic landscape of ASD and outline the genes and biological processes involved in the condition, while taking into account that the study relied exclusively on in silico analyses, which may be subject to biases inherent to database methodologies. Further research incorporating multi-omics data and experimental validation is warranted to enhance our understanding of non-syndromic ASD genetics and facilitate the development of targeted research, interventions and therapies.

Gene selection for high dimensional biological datasets using hybrid island binary artificial bee colony with chaos game optimization

Microarray technology, as applied to the fields of bioinformatics, biotechnology, and bioengineering, has made remarkable progress in both the treatment and prediction of many biological problems. However, this technology presents a critical challenge due to the size of the numerous genes present in the high-dimensional biological datasets associated with an experiment, which leads to a curse of dimensionality on biological data. Such high dimensionality of real biological data sets not only increases memory requirements and training costs, but also reduces the ability of learning algorithms to generalise. Consequently, multiple feature selection (FS) methods have been proposed by researchers to choose the most significant and precise subset of classified genes from gene expression datasets while maintaining high classification accuracy. In this research work, a novel binary method called iBABC-CGO based on the island model of the artificial bee colony algorithm, combined with the chaos game optimization algorithm and SVM classifier, is suggested for FS problems using gene expression data. Due to the binary nature of FS problems, two distinct transfer functions are employed for converting the continuous search space into a binary one, thus improving the efficiency of the exploration and exploitation phases. The suggested strategy is tested on a variety of biological datasets with different scales and compared to popular metaheuristic-based, filter-based, and hybrid FS methods. Experimental results supplemented with the statistical measures, box plots, Wilcoxon tests, Friedman tests, and radar plots demonstrate that compared to prior methods, the proposed iBABC-CGO exhibit competitive performance in terms of classification accuracy, selection of the most relevant subset of genes, data variability, and convergence rate. The suggested method is also proven to identify unique sets of informative, relevant genes successfully with the highest overall average accuracy in 15 tested biological datasets. Additionally, the biological interpretations of the selected genes by the proposed method are also provided in our research work.

A multitasking multi-objective differential evolution gene selection algorithm enhanced with new elite and guidance strategies for tumor identification

A key preprocessing step in tumor recognition based on microarray expression profile data and machine learning is to identify tumor marker genes. Gene selection aims to select the most relevant gene subset from the original ultra-high dimensional microarray expression profile data to improve tumor identification performance. Inspired by evolutionary multitasking (EMT) and multi-objective optimization, this paper puts forward a novel multitasking multi-objective differential evolution gene selection algorithm (MMODE) which uses new elite and guidance strategies to select the best gene subsets. MMODE initializes two different populations according to different filtering criteria to increase the diversity of the search. These two populations guide their respective populations to search in the optimal direction through knowledge transfer in the evolutionary process. In addition, MMODE employs new elite and guidance strategies that enables individuals to narrow the search range and jump out of local optima. The proposed algorithm is validated on 13 publicly available microarray expression datasets in comparison with state-of-the-art gene selection algorithms. The experimental results show that MMODE can find smaller gene subsets and achieve higher classification accuracy compared with other algorithms.

A hybrid wrapper spider monkey optimization-simulated annealing model for optimal feature selection

In this research, a hybrid wrapper model is proposed to identify the featured gene subset from the gene expression data. To balance the gap between exploration and exploitation, a hybrid model with a popular meta-heuristic algorithm named spider monkey optimizer (SMO) and simulated annealing (SA) is applied. In the proposed model, ReliefF is used as a filter to obtain the relevant gene subset from dataset by removing the noise and outliers prior to feeding the data to the wrapper SMO. To enhance the quality of the solution, simulated annealing is deployed as local search with the SMO in the second phase, which will guide to the detection of the most optimal feature subset. To evaluate the performance of the proposed model, support vector machine (SVM) as a fitness function to recognize the most informative biomarker gene from the cancer datasets along with University of California, Irvine (UCI) datasets. To further evaluate the model, 4 different classifiers (SVM, na¨ıve Bayes (NB), decision tree (DT), and k-nearest neighbors (KNN)) are used. From the experimental results and analysis, it’s noteworthy to accept that the ReliefF-SO-SA-SVM performs relatively better than its state-of-the-art counterparts. For cancer datasets, our model performs better in terms of accuracy with a maximum of 99.45%.

A Comparative Study Of Algorithmic Efficiency Of Feature Selection Algorithm On Microarray

A key challenge before classification can take place is feature selection. An effective feature selection method would increase classification accuracy and simultaneously reduce computation costs and time. A variety of filter approaches, along with different search algorithms, were considered in this study. Five traditional classifiers were evaluated on the selected gene subsets: Random Forest, Sequential minimal optimization algorithm, Naive Bayes, Decision Trees, and K-Nearest Neighbour. The datasets chosen for this analysis are the microarray gene expression data of two types of cancers: Acute Lymphocytic Leukaemia (ALL)/Acute Myeloid Leukaemia (AML) and Lung cancer. According to the experimental results, a fuzzy rough subset combined with Genetic Search selects optimal relevant gene subsets and produces significantly good classifier accuracy. Compared to classical classifiers described here, this research finds that Random Forest classifiers yield 94.33% on the raw dataset and 100% classifier accuracy after applying feature selection methods. Utilizing conventional methods like Precision, Recall, F-Score, and Region of Characteristics, MCC Matthews correlation coefficient, results are validated.

Multilevel Feature Selection Method for Improving Classification of Microarray Gene Expression Data

Microarray gene expression profiles provide valuable answers to a variety of problems, and contributes to advances in clinical medicine. Gene expression data typically has a high dimension and a small sample size. Gene selection from microarray gene expression data is a challenge due to high dimensionality of the data. The number of samples in the microarray dataset is much smaller compared to the number of genes as features. To extract useful gene information from cancer microarray data and reduce dimensionality, selection of significant genes is necessary. An effective method of gene feature selection helps in dimensionality reduction and improves the classification performance. Experimental results suggest that appropriate combination of filter gene selection methods is more effective than individual techniques for microarray data classification. In this paper, we propose a two-layered feature selection method. In the first layer, t-test statistical method is used to remove the features that have little correlation with the classification results. In the second layer, line segment approximation method is used to transform the feature subset into a less dimensional feature space. Four well known classifiers kNN, SVM, NBC, DT were used to verify the performance of the proposed feature selection algorithm on binary class microarray data. The experimental results show that the proposed method can effectively select relevant gene subsets, and achieves higher classification accuracy.

PRMT1, a Key Modulator of Unliganded Progesterone Receptor Signaling in Breast Cancer.

The progesterone receptor (PR) is a key player in major physiological and pathological responses in women, and the signaling pathways triggered following hormone binding have been extensively studied, particularly with respect to breast cancer development and progression. Interestingly, growing evidence suggests a fundamental role for PR on breast cancer cell homeostasis in hormone-depleted conditions, with hormone-free or unliganded PR (uPR) involved in the silencing of relevant genes prior to hormonal stimulation. We herein identify the protein arginine methyltransferase PRMT1 as a novel actor in uPR signaling. In unstimulated T47D breast cancer cells, PRMT1 interacts and functions alongside uPR and its partners to target endogenous progesterone-responsive promoters. PRMT1 helps to finely tune the silencing of responsive genes, likely by promoting a proper BRCA1-mediated degradation and turnover of unliganded PR. As such, PRMT1 emerges as a key transcriptional coregulator of PR for a subset of relevant progestin-dependent genes before hormonal treatment. Since women experience periods of hormonal fluctuation throughout their lifetime, understanding how steroid receptor pathways in breast cancer cells are regulated when hormones decline may help to determine how to override treatment failure to hormonal therapy and improve patient outcome.

Determining Clinical Course of Diffuse Large B-Cell Lymphoma Using Targeted Transcriptome and Machine Learning Algorithms

Determining Clinical Course of Diffuse Large B-Cell Lymphoma Using Targeted Transcriptome and Machine Learning Algorithms

SARA: A memetic algorithm for high-dimensional biomedical data

Over the past two decades, large amounts of biomedical and clinical data have been generated. These high dimensional datasets contain thousands of genes. However, such datasets contain many irrelevant genes which influence the predictive accuracy of diagnosis. Therefore, to select the relevant genes from the dataset and to accurately identify the patterns in the genes, it is necessary to employ some gene selection and classification algorithms. In this work, a hybrid algorithm is proposed using simulated annealing (SA) and Rao algorithm (RA) for selecting the optimal gene subset and classifying cancer. SA works as a local search strategy and RA works as a global optimization framework. The reason for combining SA in RA is to improve the exploitation capability of RA. The proposed method consists of two stages. In the first stage, minimum redundancy maximum relevance (mRMR) is employed to select the relevant gene subsets from the microarray dataset. Then, SA is hybridized with RA to improve the quality of solutions after every iteration of RA. Log sigmoidal function is introduced as an encoding scheme to transform the continuous version of Simulated annealing-Rao algorithm (SARA) to a discrete optimization algorithm. The performance of our approach is tested on three binary-class and four multi-class datasets. A comparative study is carried out with eighteen existing techniques. Results from the experiments have shown that our proposed approach selects discriminating genes with high classification accuracy. Particularly, it achieves high classification accuracy on the SRBCT dataset with 99.81% with only five informative genes.

Hybridization of Moth flame optimization algorithm and quantum computing for gene selection in microarray data

Ever-increasing data in various fields like Bioinformatics field, which has led to the need to find a way to reduce the data dimensionality. Gene selection problem has a large number of genes (relevant, redundant or noise), which needs an effective method to help us in detecting diseases and cancer. In this problem, computational complexity is reduced by selecting a small number of genes, but it is necessary to choose the relevant genes to keep a high level of accuracy. Therefore, in order to find the optimal gene subset, it is essential to devise an effective exploration approach that can investigate a large number of possible gene subsets. In addition, it is required to use a powerful evaluation method to evaluate the relevance of these gene subsets. In this paper, we present a novel swarm intelligence algorithm for gene selection called quantum moth flame optimization algorithm (QMFOA), which based on hybridization between quantum computation and moth flame optimization (MFO) algorithm. The purpose of QMFOA is to identify a small gene subset that can be used to classify samples with high accuracy. The QMFOA has a simple two-phase approach, the first phase is a pre-processing that uses to address the difficulty of high-dimensional data, which measure the redundancy and the relevance of the gene, in order to obtain the relevant gene set. The second phase is a hybridization among MFOA, quantum computing, and support vector machine with leave-one-out cross-validation, etc., in order to solve the gene selection problem. We use quantum computing to guarantee a good trade-off between the exploration and the exploitation of the search space, while a new update moth operation using Hamming distance and Archimedes spiral allows an efficient exploration of all possible gene-subsets. The main objective of the second phase is to determine the best relevant gene subset of all genes obtained in the first phase. In order to assess the performance of the proposed QMFOA, we test QMFOA on thirteen microarray datasets (six binary-class and seven multi-class) to evaluate and compare the classification accuracy and the number of genes selected by the QMFOA against many recently published algorithms. Experimental results show that QMFOA provides greater classification accuracy and the ability to reduce the number of selected genes compared to the other algorithms.

Gene Selection and Classification Using Quantum Moth Flame Optimization Algorithm

In this paper, we present a new swarm intelligence algorithm for gene selection called quantum moth flame optimization algorithm (QMFOA), which based on hybridization between quantum computation and moth flame optimization algorithm (MFOA). The purpose of QMFOA is to identify a small gene subset that can be used to classify samples with high accuracy. The QM- FOA has a simple two-phase approach, the first phase is a pre-processing that uses to address the difficulty of high-dimensional data, which measure the redundancy and the relevance of the gene, in order to obtain the relevant gene set. The second phase is hybridization among MFOA, quantum computing, and support vector machine (SVM) with leave-one-out cross-validation (LOOCV), in order to solve the gene selection problem. The main objective of the second phase is to determine the best relevant gene subset of all genes obtained in the first phase.In order to assess the performance of the proposed QMFOA, we test it on six Microarray datasets. Experimental results show that QMFOA provides great classification accuracy in comparison to some known algorithms.

Master Regulators of Signaling Pathways: An Application to the Analysis of Gene Regulation in Breast Cancer.

Analysis of gene regulatory networks allows the identification of master transcriptional factors that control specific groups of genes. In this work, we inferred a gene regulatory network from a large dataset of breast cancer samples to identify the master transcriptional factors that control the genes within signal transduction pathways. The focus in a particular subset of relevant genes constitutes an extension of the original Master Regulator Inference Algorithm (MARINa) analysis. This modified version of MARINa utilizes a restricted molecular signature containing genes from the 25 human pathways in KEGG's signal transduction category. Our breast cancer RNAseq expression dataset consists of 881 samples comprising tumors and normal mammary gland tissue. The top 10 master transcriptional factors found to regulate signal transduction pathways in breast cancer we identified are: TSHZ2, HOXA2, MEIS2, HOXA3, HAND2, HOXA5, TBX18, PEG3, GLI2, and CLOCK. The functional enrichment of the regulons of these master transcriptional factors showed an important proportion of processes related to morphogenesis. Our results suggest that, as part of the aberrant regulation of signaling pathways in breast cancer, pathways similar to the regulation of cell differentiation, cardiovascular system development, and vasculature development may be dysregulated and co-opted in favor of tumor development through the action of these transcription factors.

Gene Selection via a New Hybrid Ant Colony Optimization Algorithm for Cancer Classification in High-Dimensional Data.

The recent advance in the microarray data analysis makes it easy to simultaneously measure the expression levels of several thousand genes. These levels can be used to distinguish cancerous tissues from normal ones. In this work, we are interested in gene expression data dimension reduction for cancer classification, which is a common task in most microarray data analysis studies. This reduction has an essential role in enhancing the accuracy of the classification task and helping biologists accurately predict cancer in the body; this is carried out by selecting a small subset of relevant genes and eliminating the redundant or noisy genes. In this context, we propose a hybrid approach (MWIS-ACO-LS) for the gene selection problem, based on the combination of a new graph-based approach for gene selection (MWIS), in which we seek to minimize the redundancy between genes by considering the correlation between the latter and maximize gene-ranking (Fisher) scores, and a modified ACO coupled with a local search (LS) algorithm using the classifier 1NN for measuring the quality of the candidate subsets. In order to evaluate the proposed method, we tested MWIS-ACO-LS on ten well-replicated microarray datasets of high dimensions varying from 2308 to 12600 genes. The experimental results based on ten high-dimensional microarray classification problems demonstrated the effectiveness of our proposed method.

Analysis of high-dimensional genomic data employing a novel bio-inspired algorithm

Over the last decade, there has been a rapid growth in the generation and analysis of the genomics data. Though the existing data analysis methods are capable of handling a particular problem, they cannot guarantee to solve all problems with different nature. Therefore, there always lie a scope of a new algorithm to solve a problem which cannot be efficiently solved by the existing algorithms. In the present work, a novel hybrid approach is proposed based on the improved version of a recently developed bio-inspired optimization technique, namely, salp swarm algorithm (SSA) for microarray classification. Initially, the Fisher score filter is employed to pre-select a subset of relevant genes from the original high-dimensional microarray dataset. Later, a weighted-chaotic SSA (WCSSA) is proposed for the simultaneous optimal gene selection and parameter optimization of the kernel extreme learning machine (KELM) classifier. The proposed scheme is experimented on both binary-class and multi-class microarray datasets. An extensive comparison is performed against original SSA-KELM, particle swarm optimized-KELM (PSO-KELM), and genetic algorithm-KELM (GA-KELM). Lastly, the proposed method is also compared against the results of sixteen existing techniques to emphasize its capacity and competitiveness to successfully reduce the number of original genes by more than 98%. The experimental results show that the genes selected by the proposed method yield higher classification accuracy compared to the alternative techniques. The performance of the proposed scheme demonstrates its effectiveness in terms of number of selected genes (NSG), accuracy, sensitivity, specificity, Matthews correlation coefficient (MCC), and F-measure. The proposed WCSSA-KELM method is validated using a ten-fold cross-validation technique.

Neuroevolution as a tool for microarray gene expression pattern identification in cancer research.

Microarrays are still one of the major techniques employed to study cancer biology. However, the identification of expression patterns from microarray datasets is still a significant challenge to overcome. In this work, a new approach using Neuroevolution, a machine learning field that combines neural networks and evolutionary computation, provides aid in this challenge by simultaneously classifying microarray data and selecting the subset of more relevant genes. The main algorithm, FS-NEAT, was adapted by the addition of new structural operators designed for this high dimensional data. In addition, a rigorous filtering and preprocessing protocol was employed to select quality microarray datasets for the proposed method, selecting 13 datasets from three different cancer types. The results show that Neuroevolution was able to successfully classify microarray samples when compared with other methods in the literature, while also finding subsets of genes that can be generalized for other algorithms and carry relevant biological information. This approach detected 177 genes, and 82 were validated as already being associated to their respective cancer types and 44 were associated to other types of cancer, becoming potential targets to be explored as cancer biomarkers. Five long non-coding RNAs were also detected, from which four don't have described functions yet. The expression patterns found are intrinsically related to extracellular matrix, exosomes and cell proliferation. The results obtained in this work could aid in unraveling the molecular mechanisms underlying the tumoral process and describe new potential targets to be explored in future works.

Abstract P1-01-12: Expression of autophagy related genes impacts clinical outcomes of human breast carcinoma and is associated with estrogen and progestin receptor status

Abstract The dual role of autophagy in breast cancer initiation, progression and responsiveness to various therapies is the focus of extensive studies. Our goal is to assess the relationship of expression of certain autophagy related genes in primary breast carcinomas to predict risk of recurrence. Our hypothesis includes the caveat that unique gene expression subsets will be deciphered by utilizing Laser Capture Microdissected breast carcinoma cells from tissue biopsies containing many cell types. Methods: Comprehensive analyses of de-identified biomarker results, characteristics of 247 breast carcinoma specimens and clinical outcomes were performed by univariable Cox regressions, Kaplan Meier plots and LASSO with R software version 3.2.5. Microarray analyses were performed on RNA isolated from LCM-procured carcinoma cells to identify gene signatures associated with breast cancer behavior. Results: Expression levels of 22 autophagy related genes analyzed by univariable Cox regression revealed that RB1CC1, KEAP1, ATG7, RUBCN, NOD2, HMOX1, BECN1, PIK3R4 were significant for predicting Progression Free Survival (PFS) at the discovery level of the adjusted p-value < 0.3. Of these, RB1CC1, KEAP1, & ATG7 were significant for Overall Survival (OS) without regard to ER/PR status. Using Kaplan Meier analyses, expression levels of each of 10 candidate genes predicted PFS of which expression of 6 of these genes also predicted OS using a median split cutoff without regard to ER/PR status. Applying LASSO computations without regard to ER/PR status of the primary breast cancer, a clinically relevant gene expression profile consisting of ATG7, BCL2, HMOX1, KEAP1, NOD2, PTEN, RB1CC1, and ULK4 predicted PFS. Collectively, expression of these 8 genes with TP53 and RUBCN predicted OS without regard to ER/PR status. Violin plots of relative gene expression of each of the 22 candidate genes known to be associated with autophagy pathways revealed that ATG7, ATG12, BCL2 and BECN1 were elevated in ER+ lesions while BCL2 and BECN1 were also elevated in PR+ breast carcinomas. Expression levels of 16 of the 22 autophagy related genes examined by univariable Cox regression were related to either ER or PR status of the primary lesion and predicted either PFS or OS. Refinement of clinically relevant gene subsets was accomplished by LASSO calculations in which either the ER or PR protein status of the primary breast carcinoma was considered. Of the genes in the molecular signatures derived, expression levels in breast carcinomas separated by either ER or PR status of the lesion were tested by Kaplan Meier analyses to assess relationships to PFS and OS. Conclusions: Using gene expression results derived from microarray analyses of LCM-procured breast carcinoma cells of primary lesions, subsets of autophagy related genes were identified that predict a patient's risk of recurrence and overall survival. Expression of a number of candidate genes appears to be related to either/both the ER or PR protein status of the primary lesion. Collectively, results suggest that expression of certain autophagy related genes may serve as biomarkers for assessing prognosis of breast carcinoma thus impacting clinical management of breast cancer. Citation Format: Wittliff JL, Hameed ZR, Daniels MW, Cheng A. Expression of autophagy related genes impacts clinical outcomes of human breast carcinoma and is associated with estrogen and progestin receptor status [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-01-12.

A hybrid gene selection method for microarray recognition

DNA microarray data is expected to be a great help in the development of efficient diagnosis and tumor classification. However, due to the small number of instances compared to a large number of genes, many of the computational learning methods encounter difficulties to select the low subgroups. In order to select significant genes from the high dimensional data for tumor classification, nowadays, several researchers are exploring microarray data using various gene selection methods. However, there is no agreement between existing gene selection techniques that produce the relevant gene subsets by which it improves the classification accuracy. This motivates us to invent a new hybrid gene selection method which helps to eliminate the misleading genes and classify a disease correctly in less computational time. The proposed method composes of two-stage, in the first stage, EGS method using multi-layer approach and f-score approach is applied to filter the noisy and redundant genes from the dataset. In the second stage, adaptive genetic algorithm (AGA) work as a wrapper to identify significant genes subsets from the reduced datasets produced by EGS that can contribute to detect cancer or tumor. AGA algorithm uses the support vector machine (SVM) and Naïve Bayes (NB) classifier as a fitness function to select the highly discriminating genes and to maximize the classification accuracy. The experimental results show that the proposed framework provides additional support to a significant reduction of cardinality and outperforms the state-of-art gene selection methods regarding accuracy and an optimal number of genes.

Development of a two-stage gene selection method that incorporates a novel hybrid approach using the cuckoo optimization algorithm and harmony search for cancer classification

For each cancer type, only a few genes are informative. Due to the so-called ‘curse of dimensionality’ problem, the gene selection task remains a challenge. To overcome this problem, we propose a two-stage gene selection method called MRMR-COA-HS. In the first stage, the minimum redundancy and maximum relevance (MRMR) feature selection is used to select a subset of relevant genes. The selected genes are then fed into a wrapper setup that combines a new algorithm, COA-HS, using the support vector machine as a classifier. The method was applied to four microarray datasets, and the performance was assessed by the leave one out cross-validation method. Comparative performance assessment of the proposed method with other evolutionary algorithms suggested that the proposed algorithm significantly outperforms other methods in selecting a fewer number of genes while maintaining the highest classification accuracy. The functions of the selected genes were further investigated, and it was confirmed that the selected genes are biologically relevant to each cancer type.