DNA Microarray Data Research Articles

Comparative Study of Classification Algorithms for Various DNA Microarray Data.

Microarrays are applications of electrical engineering and technology in biology that allow simultaneous measurement of expression of numerous genes, and they can be used to analyze specific diseases. This study undertakes classification analyses of various microarrays to compare the performances of classification algorithms over different data traits. The datasets were classified into test and control groups based on five utilized machine learning methods, including MultiLayer Perceptron (MLP), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and k-Nearest Neighbors (KNN), and the resulting accuracies were compared. k-fold cross-validation was used in evaluating the performance and the result was analyzed by comparing the performances of the five machine learning methods. Through the experiments, it was observed that the two tree-based methods, DT and RF, showed similar trends in results and the remaining three methods, MLP, SVM, and DT, showed similar trends. DT and RF generally showed worse performance than other methods except for one dataset. This suggests that, for the effective classification of microarray data, selecting a classification algorithm that is suitable for data traits is crucial to ensure optimum performance.

Augmentation of Transcriptomic Data for Improved Classification of Patients with Respiratory Diseases of Viral Origin.

To better understand the molecular basis of respiratory diseases of viral origin, high-throughput gene-expression data are frequently taken by means of DNA microarray or RNA-seq technology. Such data can also be useful to classify infected individuals by molecular signatures in the form of machine-learning models with genes as predictor variables. Early diagnosis of patients by molecular signatures could also contribute to better treatments. An approach that has rarely been considered for machine-learning models in the context of transcriptomics is data augmentation. For other data types it has been shown that augmentation can improve classification accuracy and prevent overfitting. Here, we compare three strategies for data augmentation of DNA microarray and RNA-seq data from two selected studies on respiratory diseases of viral origin. The first study involves samples of patients with either viral or bacterial origin of the respiratory disease, the second study involves patients with either SARS-CoV-2 or another respiratory virus as disease origin. Specifically, we reanalyze these public datasets to study whether patient classification by transcriptomic signatures can be improved when adding artificial data for training of the machine-learning models. Our comparison reveals that augmentation of transcriptomic data can improve the classification accuracy and that fewer genes are necessary as explanatory variables in the final models. We also report genes from our signatures that overlap with signatures presented in the original publications of our example data. Due to strict selection criteria, the molecular role of these genes in the context of respiratory infectious diseases is underlined.

Hematopoietic mosaic chromosomal alterations in the New England Centenarian Study.

Mosaic chromosomal alterations (mCAs) are structural alterations that include deletions, duplications, or copy-neutral loss of heterozygosity. mCAs are reported to be associated with survival, age, cancer, and cardiovascular disease. Previous studies of mCAs in large population-based cohorts (UK Biobank, MGBB, BioBank Japan, and FinnGen) have demonstrated a steady increase of mCAs as people age. The distribution of mCAs in centenarians and their offspring is not well characterized. We applied MOsaic CHromosomal Alteration (MoChA) caller on 2298 genome-wide genotype samples of 1582 centenarians, 443 centenarians’ offspring, and 273 unrelated controls from the New England Centenarian Study (NECS). Integrating Log R ratio and B-allele frequency (BAF) intensities with genotype phase information, MoChA employs a Hidden Markov Model to detect mCA-induced deviations in allelic balance at heterozygous sites consistent with genotype phase in the DNA microarray data. We analyzed mCAs spanning over 100 k base pairs, with an estimated cell fraction less than 50%, within samples with genome-wide BAF phase concordance across phased heterozygous sites less than 0.51, and with LOD score of more than 10 for the model based on BAF and genotype phase. Our analysis showed that somatic mCAs increase with older age up to approximately 102 years, but the prevalence of the subjects with mCAs tend to decrease after that age, thus suggesting that accumulation of mCAs is less prevalent in long-lived individuals. We also used Poisson regression to show that centenarians and their offspring tend to accumulate less mCA (RR = 0.63, p=0.045) compared to the controls.

Microarray Data Preprocessing: From Experimental Design to Differential Analysis.

DNA microarray data preprocessing is of utmost importance in the analytical path starting from the experimental design and leading to a reliable biological interpretation. In fact, when all relevant aspects regarding the experimental plan have been considered, the following steps from data quality check to differential analysis will lead to robust, trustworthy results. In this chapter, all the relevant aspects and considerations about microarray preprocessing will be discussed. Preprocessing steps are organized in an orderly manner, from experimental design to quality check and batch effect removal, including the most common visualization methods. Furthermore, we will discuss data representation and differential testing methods with a focus on the most common microarray technologies, such as gene expression and DNA methylation.

Nutrigenomic Studies on the Ameliorative Effect of Enzyme-Digested Phycocyanin in Alzheimer's Disease Model Mice.

Alzheimer’s disease (AD) is the most common form of dementia, and the cognitive impairments associated with this degenerative disease seriously affect daily life. Nutraceuticals for the prevention or delay of AD are urgently needed. It has been increasingly observed that phycocyanin (PC) exerts neuroprotective effects. AD model mice intracerebroventricularly injected with amyloid beta-peptide 25–35 (Aβ25–35) at 10 nmol/head displayed significant cognitive impairment in the spontaneous alternation test. Cognitive impairment was significantly ameliorated in mice treated with 750 mg/kg of enzyme-digested (ED) PC by daily oral administration for 22 consecutive days. Application of DNA microarray data on hippocampal gene expression to nutrigenomics studies revealed that oral EDPC counteracted the aberrant expression of 35 genes, including Prnp, Cct4, Vegfd (Figf), Map9 (Mtap9), Pik3cg, Zfand5, Endog, and Hbq1a. These results suggest that oral administration of EDPC ameliorated cognitive impairment in AD model mice by maintaining and/or restoring normal gene expression patterns in the hippocampus.

Gene selection for microarray data classification via multi-objective graph theoretic-based method

In recent decades, the improvement of computer technology has increased the growth of high-dimensional microarray data. Thus, data mining methods for DNA microarray data classification usually involve samples consisting of thousands of genes. One of the efficient strategies to solve this problem is gene selection, which improves the accuracy of microarray data classification and also decreases computational complexity. In this paper, a novel social network analysis-based gene selection approach is proposed. The proposed method has two main objectives of the relevance maximization and redundancy minimization of the selected genes. In this method, on each iteration, a maximum community is selected repetitively. Then among the existing genes in this community, the appropriate genes are selected by using the node centrality-based criterion. The reported results indicate that the developed gene selection algorithm while increasing the classification accuracy of microarray data, will also decrease the time complexity.

Candidate genes for congenital malformations in pigs

Candidate genes for congenital malformations in pigs

Gene expression analysis of RCC1, VAV2, RPA3, and SRPK1 for human cervical cancer biomarkers

Cervical cancer is the most common cancer in women worldwide, with developing countries accounting for 80% of the disease burden. Infection with a high-risk group of human papilloma viruses (HPVs) is caused mainly by carcinogenesis of cervical cancer. Although, active diagnosis has been improved treatment efficacy in reducing the incidence and mortality rate. Cervical cancer is generally classified into four histopathological subtypes; adenocarcinoma, squamous cell carcinoma, papilloma and carcinoma. In this study, the gene expression profiles of these four subtypes in cervical cancer cell lines were compared to normal cervical tissue using quantitative real-time PCR (qPCR). Sixty-one genes from public DNA microarray data were selected. The expression of 30 candidate genes including a housekeeping gene, GAPDH, were quantified with qPCR in HeLa, SiHa, C-4 I and SW756 cell lines showing more than 1.5-fold difference in their expression. VAV2 gene was found to overexpress in Adenocarcinoma (HeLa), Squamous cell carcinoma (SiHa) and Carcinoma (SW756) and RCC1 gene was overexpressed in papilloma (C-4 I) of cervical cell lines. Based on their upregulation level, VAV2 gene was identified as a candidate marker for adenocarcinoma, squamous cell carcinoma, and carcinoma while RCC1 gene was promising as a candidate marker for papilloma subtype of cervical cancer. In addition, of RCC1, VAV2, RPA3, and SRPK1 were potential gene candidates for cervical cancer markers. The candidate biomarkers may useful for screening and diagnosis.

Abstract 10784: Circulating Obesity Associated Pro-Fibrotic Protein Promotes Fibrosis in Liver and Heart

Introduction: Non-alcoholic steatohepatitis (NASH), driven by the obesity epidemic, has become the most common form of liver disease. Obesity contributes inflamed visceral adipose tissue secretes pro-inflammatory adipokines that is causal for systemic metabolic disorders. Despite its importance, the mechanism of NASH in the presence of morbid obesity related adipokines from BAT remains unclear. Hypothesis: Various kinds of white adipokines are known in obesity. However, brown adipokines still remain to be explored. Methods: To identify the function of obesity associated pro-fibrotic protein (OAFP), we generated a murine obese NASH model by imposing a high fat diet in C57BL6/NCr mice, and murine systemic OAFP knockout (KO) model. We also conducted functional in vitro studies with differentiated brown adipocytes. Results: Analyzing two sets of DNA microarray data with bioinformatics, we identified a secreted form OAFP expressed in dysfunctional brown adipose tissue (BAT) in mice. Testing our biobank samples, we found this protein increased in plasma of NASH patients or aged individuals. We generated a murine obese NASH model by imposing a high fat diet in C57BL6/NCr mice, and found that OAFP is produced predominantly by BAT. In this model, we also found that OAFP increased in plasma. For this reason, we generated a murine systemic OAFP knockout (KO) model and found that liver fibrosis ameliorated in OAFP-KO model. In addition, vaccination therapy also inhibited liver fibrosis. In contrast, liver fibrosis augmented in OAFP gain of function (GOF) model, and these results suggested that OAFP has causal role for the progression of fibrotic response in liver. In the obese NASH model, we found that cardiac fibrosis also developed and this ameliorated in OAFP-KO model, indicating that OAFP may have pathological roles for heart failure with preserved ejection fraction (HFpEF) related with age-related disorders. In vitro studies with differentiated brown adipocytes showed that OAFP was upregulated through by PERK/JNK/c-fos/AP1 pathways. Conclusions: Our results suggest that OAFP contributes for the progression of fibrotic responses in obese or aged models. Inhibition of OAFP may become a therapy for NASH or HFpEF.

Gene expression analysis of human induced pluripotent stem cells cryopreserved by vitrification using StemCell Keep

In recent years, regenerative medicine research using human somatic and induced pluripotent stem cells has advanced considerably, promoting clinical applications. However, it is essential that these cells are cryopreserved safely and effectively. Most cryopreservation solution agents contain dimethyl sulfoxide (DMSO), which exhibits strong toxicity and can potentially promote cell differentiation. Hence, it is important to explore substitutes for DMSO in cryoprotectant solutions. One such alternative is StemCell Keep (SCK), a DMSO-free solution that has been reported to effectively cryopreserve human induced pluripotent stem cells (hiPS cells). To clarify the effect of cryopreservation agents on cells, DNA microarray analysis is useful, as it can identify a large number of gene expression differences in cryopreserved cells, as well as functional increases in gene groups. In this study, we performed gene expression analysis of SCK-cryopreserved hiPS cells using a DNA microarray gene chip. The hiPS cells vitrified with SCK or DMSO-based vitrification solutions were thawed and cultured on Matrigel under feeder-free conditions, and RNA was extracted for DNA microarray analysis. Genes obtained from DNA microarray data were classified by the keywords of Gene Ontology Biological Process Term, and their relationships were analyzed using DAVID or the GeneMANIA database.SCK-cryopreserved hiPS cells expressed several anti-apoptotic genes, as well as genes related to cell adhesion or proliferation at levels that were nearly equivalent to those of non-frozen hiPS cells. Gene enrichment analysis with selected genes of SCK-cryopreserved hiPS cells whose expression differences were superior to those of DAP-cryopreserved showed strong interactions of negative regulation of apoptotic process, cell adhesion and positive regulation of cell proliferation in DAVID analysis. We demonstrated that SCK successfully maintained the key functions of hiPS cells, including anti-apoptosis, cell adhesion, and cell proliferation, during cryopreservation.

Boundaries tuned support vector machine (BT-SVM) classifier for cancer prediction from gene selection

In recent days, the identified genes which are detecting cancer-causing diseases are plays a crucial part in the microarray data analysis. Huge volume of data required since the disease changed often. Conventional data mining techniques are lacking in space concern and time complexity. Based on big data the proposed work is executed. Using the ISPCA – Improved Supervised Principal Component Analysis, feature extraction is developed in this study. For gene expression, co-variance matrix is generated and through feature selection cancer classification is performed by IPSCA. Further feature selection process by boundaries tuned support vector machines (BT-SVM) classifier and modified particle swarm optimization with novel wrapper model algorithm are performed. The experimentation is carried out by utilizing different datasets like leukaemia, breast cancer dataset, brain cancer, colon, and lung carcinoma from the UCI repository. The proposed work is executed on six benchmark dataset for DNA microarray data in terms of accuracy, recall, and precision to evaluate the performance of the proposed work. For evaluating the proposed work effectiveness, it is compared with various traditional techniques and resulted in optimum accuracy, recall, precision and training time with and without feature selection effectively.

PD37-06 LATROPHILIN-3 AS A KEY DOWNSTREAM EFFECTOR OF THE ANDROGEN RECEPTOR INDUCES UROTHELIAL TUMORIGENESIS

You have accessJournal of UrologyBladder Cancer: Basic Research & Pathophysiology I (PD37)1 Sep 2021PD37-06 LATROPHILIN-3 AS A KEY DOWNSTREAM EFFECTOR OF THE ANDROGEN RECEPTOR INDUCES UROTHELIAL TUMORIGENESIS Takuro Goto, Yuki Teramoto, Yujiro Nagata, Guiyang Jiang, Taichi Mizushima, Satoshi Inoue, Hiroki Ide, George Netto, and Hiroshi Miyamoto Takuro GotoTakuro Goto More articles by this author , Yuki TeramotoYuki Teramoto More articles by this author , Yujiro NagataYujiro Nagata More articles by this author , Guiyang JiangGuiyang Jiang More articles by this author , Taichi MizushimaTaichi Mizushima More articles by this author , Satoshi InoueSatoshi Inoue More articles by this author , Hiroki IdeHiroki Ide More articles by this author , George NettoGeorge Netto More articles by this author , and Hiroshi MiyamotoHiroshi Miyamoto More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000002047.06AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: We and others have indicated that androgen receptor (AR) signaling plays a critical role in urothelial cancer outgrowth. Meanwhile, our DNA microarray data suggest that latrophilin-3 (LPHN3), one of latrophilins that are a group of the G-protein-coupled receptor where a spider venom latrotoxin (LTX) is known to bind, represents a downstream target of AR in bladder cancer cells. The present study aims to determine the functional role of LPHN3 in urothelial carcinogenesis. METHODS: We determined the expression of LPHN3 in 145 bladder tumors and paired normal bladder tissues (by immunohistochemistry) as well as in human normal urothelial SVHUC sublines stably expressing wild-type AR or vector only (by RT-PCR and western blot). We then assessed the effects of ligand (e.g. LTX, FLRT3) treatment on neoplastic/malignant transformation of SVHUC-derived cells, as well as bladder tumor development in C57BL/6 mice, induced by chemical carcinogens (e.g. MCA, BBN). RESULTS: Immunohistochemistry showed that LPHN3 was positive in 52% (47% 1+, 5% 2+) of tumors, which was significantly (P=0.049) higher than in non-neoplastic urothelial tissues [37% (all 1+)], and that LPHN3 positivity in non-muscle-invasive tumors was associated with the risk of tumor recurrence after transurethral surgery (P=0.051). The level of LPHN3 expression was higher in SVHUC-AR cells than in AR-negative SVHUC-vector cells. In addition, significant induction in the LPHN3 expression by LTX or FLRT3 treatment was seen in SVHUC, whereas LPHN3 knockdown via shRNA virus infection in SVHUC-AR resulted in considerable induction or reduction of the expression of the proteins positively (e.g. cleaved caspase-3) or negatively (e.g. Bcl-2, Bcl-xL), respectively, associated with apoptosis. Chromatin immunoprecipitation assay further revealed the binding of AR to the promoter region of LPHN3 in SVHUC-derived cells. The in vitro transformation assay showed that LTX and FLRT3 induced the MCA-mediated oncogenic activity in both SVHUC-AR and SVHUC-vector cells, while the inhibitory effect of LPHN3 knockdown on the neoplastic transformation was observed in MCA-SVHUC-AR cells. LTX/FLRT3 also significantly accelerated the development of bladder tumors in BBN-treated female mice. CONCLUSIONS: These findings suggest that LPHN3, which is activated in bladder tumors as a key downstream effector of the AR, promotes urothelial tumorigenesis. Accordingly, LPHN3 inhibition, along with AR inactivation, has the potential of being an effective chemopreventive approach for urothelial cancer. Source of Funding: Ferring Research Institute © 2021 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 206Issue Supplement 3September 2021Page: e657-e657 Advertisement Copyright & Permissions© 2021 by American Urological Association Education and Research, Inc.MetricsAuthor Information Takuro Goto More articles by this author Yuki Teramoto More articles by this author Yujiro Nagata More articles by this author Guiyang Jiang More articles by this author Taichi Mizushima More articles by this author Satoshi Inoue More articles by this author Hiroki Ide More articles by this author George Netto More articles by this author Hiroshi Miyamoto More articles by this author Expand All Advertisement Loading ...

Improved Lasso (ILASSO) for Gene Selection and Classification in High Dimensional DNA Microarray Data

<p class="0abstract">Classification and selection of gene in high dimensional microarray data has become a challenging problem in molecular biology and genetics. Penalized Adaptive likelihood method has been employed recently for classification of cancer to address both gene selection consistency and estimation of gene coefficients in high dimensional data simultaneously. Many studies from the literature have proposed the use of ordinary least squares (OLS), maximum likelihood estimation (MLE) and Elastic net as the initial weight in the Adaptive elastic net, but in high dimensional microarray data the MLE and OLS are not suitable. Likewise, considering the Elastic net as the initial weight in Adaptive elastic yields a poor performance, because the ridge penalty in the Elastic net grouped coefficient of highly correlated genes closer to each other. As a result, the estimator fails to differentiate coefficients of highly correlated genes that have different sign being grouped together. To tackle this issue, the present study proposed Improved LASSO (ILASSO) estimator which add the ridge penalty to the original LASSO with an Adaptive weight to both and simultaneously. Results from the real data indicated that ILASSO has a better performance compared to other methods in terms of the number of gene selected, classification precision, Sensitivity and Specificity.</p>

St8sia1-deficiency in mice alters tumor environments of gliomas, leading to reduced disease severity.

ABSTRACTGanglioside GD3/GD2 are over-expressed in various neuroectoderm-derived tumors. Previous studies indicated that GD3 is involved in the enhancement of cancer properties such as rapid growth and increased invasiveness. However, little is known about the functions of GD3/GD2 in glioma cells and glioma microenvironments. To clarify the functions of GD3/GD2 in gliomas, we used a mouse glioma model based on the RCAS/Gtv-a system. At first, we compared the gliomas size between wild-type (WT) and GD3 synthase (GD3S) knockout (KO) mice, showing a less malignant histology and slower tumor growth in GD3S-KO mice than in WT mice. Immunohistochemistry of glioma sections from WT and GD3S-KO mice revealed that reactive microglia/macrophages showed different localization patterns between the two genetic types of mice. CD68+ cells were more frequently stained inside glioma tissues of GD3S-KO mice, while they were stained mainly around glioma tissues in WT mice. The number of CD68+ cells markedly increased in tumor tissues of GD3S-KO mice at 2 weeks after injection of transfectant DF-1 cells. Furthermore, CD68+ cells in GD3S(-/-) glioma tissues expressed higher levels of inducible nitric oxide synthase. We observed higher expression levels of pro-inflammatory cytokine genes in primary-cultured glioma cells of WT mice than in GD3S-KO mice. DNA microarray data also revealed differential expression levels of various cytokines and chemokines in glioma tissues between WT and GD3S-KO mice. These results suggest that expression of GD3S allows glioma cells to promote polarization of microglia/macrophages towards M2-like phenotypes by modulating the expression levels of chemokines and cytokines.

Comparative analysis of ReliefF-SVM and CFS-SVM for microarray data classification

Cancer is one of the main causes of death in the world where the World Health Organization (WHO) recognized cancer as among the top causes of death in 2018. Thus, detecting cancer symptoms is paramount in order to cure and subsequently reduce the casualties due to cancer disease. Many studies have been developed data mining approaches to detect symptoms of cancer through a classifying human gene data expression. One popular approach is using microarray data based on DNA. However, DNA microarray data has many dimensions that can have a detrimental effect on the accuracy of classification. Therefore, before performing classification, a feature selection technique must be used to eliminate features that do not have important information to support the classification process. The feature selection techniques used were ReliefF and correlation-based feature selection (CFS) and a classification technique used in this study is support vector machine (SVM). Several testing schemes were applied in this analysis to compare the performance of ReliefF and CFS with SVM. It showed that the ReliefF outperformed compared with CFS as microarray data classification approach.

Effect of Wnt5a on drug resistance in estrogen receptor-positive breast cancer

BackgroundPreviously, we reported that Wnt5a-positive breast cancer can be classified as estrogen receptor (ER)-positive breast cancer; its prognosis is worse than that of Wnt5a-negative breast cancer. This study aimed to investigate the mechanisms underlying the poor prognosis in Wnt5a-positive breast cancer patients.MethodsIn total, 151 consecutive ER-positive breast cancer patients who underwent resection between January 2011 and February 2014 were enrolled. DNA microarray and pathway analyses were conducted using MCF-7 cells stably expressing Wnt5a [MCF-7/Wnt5a (+)]. Based on the outcomes, cell viability/drug sensitivity assays, and mutation analysis were performed using cell cultures and breast cancer tissues. The relationship between Wnt5a and the PI3K–AKT–mTOR signaling pathway was also examined.ResultsThe relapse-free survival rate in patients with Wnt5a-positive breast cancer was significantly lower than that in patients with Wnt5a-negative breast cancer (P = 0.047). DNA microarray data suggest that only the cytochrome P450 (CYP) pathway was significantly upregulated in MCF-7/Wnt5a (+) cells (P = 0.0440). Additionally, MCF-7/Wnt5a (+) cells displayed reduced sensitivity to the metabolic substrates of CYP, tamoxifen (P < 0.001), paclitaxel (P < 0.001), and cyclophosphamide (P < 0.001). Of note, PIK3CA mutations were not associated with the expression of Wnt5a in breast cancer tissue and culture cells.ConclusionsIn ER-positive breast cancer, Wnt5a upregulates the CYP metabolic pathway and suppresses tamoxifen, paclitaxel, and cyclophosphamide resistance, all of the three, standard treatment methods for ER-positive breast cancer. Wnt5a is thus potentially involved in the poor prognosis of ER-positive breast cancer independently of the PI3K–AKT–mTOR signaling pathway.

Enhanced Pharmaceutically Active Compounds Productivity from Streptomyces SUK 25: Optimization, Characterization, Mechanism and Techno-Economic Analysis.

The present research aimed to enhance the pharmaceutically active compounds’ (PhACs’) productivity from Streptomyces SUK 25 in submerged fermentation using response surface methodology (RSM) as a tool for optimization. Besides, the characteristics and mechanism of PhACs against methicillin-resistant Staphylococcus aureus were determined. Further, the techno-economic analysis of PhACs production was estimated. The independent factors include the following: incubation time, pH, temperature, shaker rotation speed, the concentration of glucose, mannitol, and asparagine, although the responses were the dry weight of crude extracts, minimum inhibitory concentration, and inhibition zone and were determined by RSM. The PhACs were characterized using GC-MS and FTIR, while the mechanism of action was determined using gene ontology extracted from DNA microarray data. The results revealed that the best operating parameters for the dry mass crude extracts production were 8.20 mg/L, the minimum inhibitory concentrations (MIC) value was 8.00 µg/mL, and an inhibition zone of 17.60 mm was determined after 12 days, pH 7, temperature 28 °C, shaker rotation speed 120 rpm, 1 g glucose /L, 3 g mannitol/L, and 0.5 g asparagine/L with R2 coefficient value of 0.70. The GC-MS and FTIR spectra confirmed the presence of 21 PhACs, and several functional groups were detected. The gene ontology revealed that 485 genes were upregulated and nine genes were downregulated. The specific and annual operation cost of the production of PhACs was U.S. Dollar (U.S.D) 48.61 per 100 mg compared to U.S.D 164.3/100 mg of the market price, indicating that it is economically cheaper than that at the market price.

Performance Analysis of Dimensionality Reduction Techniques in Cancer Detection using Microarray Data

Cancer is one of the major causes of deaths worldwide. This disease is more ghastly as it doesn’t announce itself until it reaches in an advance stage. Still, mortality rate for cancer can be decreased if we diagnose & provide treatment at earliest. Though there are traditional clinical trials to predict a cancer there does not a single test which can correctly identify this disease. In the recent years DNA Microarray technology has been significantly used to analyze & predict the cancer. Analysis of gene expressions is not only interesting but also challenging as it is not only the concern of accuracy but also matter of life or death of a patient. DNA Microarray data is high dimensional, noisy & redundant, it makes task of classification more complicated as high computational cost is involved. Therefore feature selection & feature reduction becomes important task prior to classification. This paper presents comparative performance analysis of different dimensionality reduction techniques implemented on TCGA PANCANCER dataset.

Performance Analysis of Dimensionality Reduction Techniques in Cancer Detection using Microarray Data

Cancer is one of the major causes of deaths worldwide. This disease is more ghastly as it doesn’t announce itself until it reaches in an advance stage. Still, mortality rate for cancer can be decreased if we diagnose & provide treatment at earliest. Though there are traditional clinical trials to predict a cancer there does not a single test which can correctly identify this disease. In the recent years DNA Microarray technology has been significantly used to analyze & predict the cancer. Analysis of gene expressions is not only interesting but also challenging as it is not only the concern of accuracy but also matter of life or death of a patient. DNA Microarray data is high dimensional, noisy & redundant, it makes task of classification more complicated as high computational cost is involved. Therefore feature selection & feature reduction becomes important task prior to classification. This paper presents comparative performance analysis of different dimensionality reduction techniques implemented on TCGA PANCANCER dataset.

Application of active learning in DNA microarray data for cancerous gene identification

Microarray technology has an important role in evaluating gene expression data with unique patterns into existence. In gene-expression based experiments, the expression level of the gene is constantly monitored in order to classify a tissue sample. In microarray technology, the expressions of the genes are altered with respect to pathogenes. The altered expression values can be identified by analyzing the genes of the tissue/cell that are affected along with the tissues/cells that are unaffected are termed as biomarkers. In the current paper, we have developed an Active Learning (AL) model by using Support Vector Machine (SVM) in association with feature-selection (FS) algorithm; called Symmetrical Uncertainty (SU) for the prediction of cancer. The effectiveness of the proposed AL and SU combination is manifested and the biomarkers or cancerous genes identified by the proposed method on four gene-expression data sets are reported. In addition, the biological significance tests are performed for the cancer biomarkers obtained from the data sets.