Forward selection method with regression analysis for optimal gene selection in cancer classification

Park, Han-Saem,Yoo, Si-Ho,Cho, Sung-Bae Taylor Francis 2007 International journal of computer mathematics Vol.84 No.5

<P> The development of DNA microarray technology has facilitated in-depth research into cancer classification, and has made it possible to process thousands of genes simultaneously. Since not all genes are crucial for classifying cancer, it is necessary to select informative genes which are associated with cancer. Many gene selection methods have been investigated, but none is perfect. In this paper we investigate methods of finding optimal informative genes for classification of gene expression profiles. We propose a new gene selection method based on the forward selection method with regression analysis in order to find informative genes which predict cancer. The genes selected by this method tend to have information about the cancer that does not overlap with the other genes selected. We have measured the sensitivity, specificity, and recognition rate of the selected genes with the $k$-nearest-neighbour classifier for the colon cancer dataset and the lymphoma dataset. In most cases, the proposed method produces better results than gene selection based on other feature selection methods, yielding a high accuracy of 90.3% for the colon cancer dataset and 72% for the lymphoma dataset.</P>

Identification and Characterization of Human Genes Targeted by Natural Selection

류하정,김영주,박영규,김재정,서을주,유한욱,박인숙,오범석,이종극 한국유전체학회 2008 Genomics & informatics Vol.6 No.4

The human genome has evolved as a consequence of evolutionary forces, such as natural selection. In this study, we investigated natural selection on the human genes by comparing the numbers of nonsynonymous (NS) and synonymous (S) mutations in individual genes. We initially collected all coding SNP data of all human genes from the public dbSNP. Among the human genes, we selected 3 different selection groups of genes: positively selected genes (NS/S≥3), negatively selected genes (NS/S≤1/3) and neutral selection genes (0.9 <NS/S<1.1). We characterized human genes targeted by natural selection. Negatively selected human genes were markedly associated with disease occurrence, but not positively selected genes. Interestingly, positively selected genes displayed an increase in potentially deleterious nonsynonymous SNPs with an increased frequency of tryptophan and tyrosine residues, suggesting a correlation with protective effects against human disease. Furthermore, our nonsynonymous/synonymous ratio data imply that specific human genes, such as ALMS1 and SPTBN5 genes, are differentially selected among distinct populations. We confirmed that inferences of natural selection using the NS/S ratio can be used extensively to identify functional genes selected during the evolutionary adaptation process.

Identification and Characterization of Human Genes Targeted by Natural Selection

Ryu, Ha-Jung,Kim, Young-Joo,Park, Young-Kyu,Kim, Jae-Jung,Park, Mi-Young,Seo, Eul-Ju,Yoo, Han-Wook,Park, In-Sook,Oh, Berm-Seok,Lee, Jong-Keuk Korea Genome Organization 2008 Genomics & informatics Vol.6 No.4

The human genome has evolved as a consequence of evolutionary forces, such as natural selection. In this study, we investigated natural selection on the human genes by comparing the numbers of nonsynonymous (NS) and synonymous (S) mutations in individual genes. We initially collected all coding SNP data of all human genes from the public dbSNP. Among the human genes, we selected 3 different selection groups of genes: positively selected genes (NS/S${\geq}$3), negatively selected genes (NS/S${\leq}$1/3) and neutral selection genes (0.9<NS/S<1.1). We characterized human genes targeted by natural selection. Negatively selected human genes were markedly associated with disease occurrence, but not positively selected genes. Interestingly, positively selected genes displayed an increase in potentially deleterious nonsynonymous SNPs with an increased frequency of tryptophan and tyrosine residues, suggesting a correlation with protective effects against human disease. Furthermore, our nonsynonymous/synonymous ratio data imply that specific human genes, such as ALMS1 and SPTBN5 genes, are differentially selected among distinct populations. We confirmed that inferences of natural selection using the NS/S ratio can be used extensively to identify functional genes selected during the evolutionary adaptation process.

In silico markers: an evolutionary and statistical approach to select informative genes of human breast cancer subtypes

Shib Sankar Bhowmick,Debotosh Bhattacharjee,Luis Rato 한국유전학회 2019 Genes & Genomics Vol.41 No.12

Background Recent advancement in bioinformatics offers the ability to identify informative genes from high dimensional gene expression data. Selection of informative genes from these large datasets has emerged as an issue of major concern among researchers. Objective Gene functionality and regulatory mechanisms can be understood through the analysis of these gene expression data. Here, we present a computational method to identify informative genes for breast cancer subtypes such as Basal, human epidermal growth factor receptor 2 (Her2), luminal A (LumA), and luminal B (LumB). Methods The proposed In Silico Markers method is a wrapper feature selection method based on Least Absolute Shrinkage and Selection Operator (LASSO), Covariance Matrix Adaptation Evolution Strategy (CMA-ES) and Support Vector Machine (SVM) as a classifier. Moreover, the composite measure consisting of relevance, redundancy, and rank score of frequently appeared genes are used to select informative genes. Results The informative genes are validated by statistical and biologically relevant criteria. For a comparative evaluation of the proposed approach, biological similarity score designed on semantic similarity measure of GO terms are investigated. Further, the proposed technique is evaluated with 7 existing gene selection techniques using two-class annotated breast cancer subtype datasets. Conclusion The utilization of this method can bring about the discovery of informative genes. Furthermore, under multiple criteria decision-making set-up, informative genes selected by the In Silico Markers are found to be admirable than the compared methods selected genes.

A Novel Feature Gene Selection Method Based On Neighborhood Mutual Information

Tao Chen,Zenglin Hong,Hui Zhao,Xiao Yang,Jun Wei 보안공학연구지원센터 2015 International Journal of Hybrid Information Techno Vol.8 No.7

DNA microarray technique can detect tens of thousands of genes activity in cells and has been widely used in clinical diagnosis. However, microarray data has characteristics of high dimension and small samples, moreover many irrelevant and redundant genes also decrease performance of classification algorithm .Mutual information is very effective method and has widely been used in feature gene selection, but it cannot directly deal with continuous features. Therefore, this paper proposes a novel feature gene selection method to resolve this problem. Firstly, a lot of irrelevant genes are eliminated from original data by using reliefF algorithm , and the candidate subset of genes is obtained; Secondly, a algorithm based on neighborhood mutual information and forward greedy search strategy which deals with directly continuous features is proposed to select feature genes in above genes subset. Here, because radius of neighborhood greatly affects reduction performance, differential evolution algorithm is applied to optimize radius before reduction. The simulation results on six benchmark microarray datasets show that our method can obtain higher classification accuracy using as few genes as possible, especially neighborhood mutual information can directly continuous features. Feature genes selected has an important meaning for understanding microarray data and finding pathogenic genes of cancer. It is an effective and efficient method for feature genes selection.

Construction of Synaptic Neural Network for Genetic Interaction Analysis

Jaeyong Yee,Mira Park 한국자료분석학회 2021 Journal of the Korean Data Analysis Society Vol.23 No.4

Contribution by a single gene to the association with trait may be either independent or through interactions with other genes. Examining all available genes for the main effect should be carried out without the time constraint. However the number of possible interacting combinations would soon become formidably large with the growing number of genes. Therefore it is often coerced to identify a group of candidate genes for the interaction and investigate only within it. Such an identification process should be able to select the group of genes having possibilities to interact with each other. Main effect of each gene should not necessarily be the criterion for the selection. We devised a neural network process that was quite sensitive to the interaction of a particular gene to the remaining ones. Contribution of each gene to the association by the genes as a whole was estimated. Selection was made based on the statistical significance for the existence of such contribution. It was demonstrated that this process might perform reliable candidate gene selection for the interaction even when the selected genes did not show significant main effect, through single scan of each individual gene.

전진선택법에 의해 선택된 부분 상관관계의 유전자들을 이용한 암 분류

유시호,조성배 대한전자공학회 2004 電子工學會論文誌-SP (Signal processing) Vol.43 No.1

유전 발현 데이터는 생명체의 특정 조직에서 채취한 샘플을 마이크로어레이상에서 측정한 것으로, 유전자들의 발현 정도가 수치로 나타난 데이터이다. 일반적으로 정상조직과 이상조직에서 관련 유전자들의 발현 정도는 차이를 보이기 때문에 유전 발현 데이터를 통하여 암을 분류할 수 있다. 그러나 분류에 모든 유전자가 관여하지는 않으므로 효율적인 암의 분류를 위해서는 관련성 있는 소수의 유전자만을 선별해내는 작업인 특징선택 방법이 필요하다. 본 논문에서는 회귀분석의 변수선택방법중 하나인 전진 선택법(forward selection method)을 사용하여 유전자들을 선하고 분류하는 방법을 제안한다. 이 방법은 선택되는 유전자들의 중복된 정보를 최소화시켜 암의 분류에 있어 보다 효과적인 유전자 선택을 한다. 실험데이터는 대장암 데이터(Colon cancer dataset)를 사용하였고, 분류기는 k-최근접 이웃(KNN)을 사용하였다. 이 방법과 상관계수를 이용한 특징 선택방법인 피어슨 상관계수와 스피어맨 상관계수방법과 비교해본 결과 전진 선택법에 의한 특징선택 방법이 암의 분류에 있어서 더 효과적인 유전자 선택을 한다는 사실을 확인하였다. 실험결과 90.3%의 높은 인식률을 보였다. 추가적으로 림프종 데이터에 대한 실험을 하였고, 그 결과 전진 선택법의 유용성을 확인할 수 있었다. Gene expression profile is numerical data of gene expression level from organism measured on the microarray. Generally, each specific tissue indicates different expression levels in related genes, so that we can classify cancer with gene expression profile. Because not all the genes are related to classification, it is needed to select related genes that is called feature selection. This paper proposes a new gene selection method using forward selection method in regression analysis. This method reduces redundant information in the selected genes to have more efficient classification. We used k-nearest neighbor as a classifier and tested with colon cancer dataset. The results are compared with Pearson's coefficient and Spearman's coefficient methods and the proposed method showed better performance. It showed 90.3% accuracy in classification. The method also successfully applied to lymphoma cancer dataset.

A Hybrid Feature Gene Selection Method based on Fuzzy Neighborhood Rough Set with Information Entropy

Tao Chen,Zenglin Hong,Fang-an Deng,Man Cui 보안공학연구지원센터(IJSIP) 2014 International Journal of Signal Processing, Image Vol.7 No.6

DNA microarray technique can detect tens of thousands of genes activity in cells and has been widely used in clinical diagnosis. However, microarray data has the characteristics of high dimension and small samples, moreover many irrelevant and redundant genes also decrease performance of classification algorithm. Feature gene selection is an effective method to solve this problem. This paper proposes a hybrid feature gene selection method. Firstly, a lot of irrelevant genes from original data were eliminated by using reliefF algorithm, and the candidate feature genes subset is obtained; Secondly, Fuzzy neighborhood rough set with information entropy which deals directly with continuous data is proposed to reduce redundant genes among genes subset above. Here, differential evolution algorithm is used to optimize radius before reduction by using fuzzy neighborhood rough set, because radius of neighborhood greatly affects reduction performance. The simulation results on six microarray datasets indicate that our method can obtain higher classification accuracy by using as few genes as possible, especially feature genes selected are important for understanding microarray data and identifying the pathogenic genes. The results demonstrated that this method is effective and efficient for feature genes selection.

Development of Resistant Gene-Pyramided Japonica Rice for Multiple Biotic Stresses Using Molecular Marker-Assisted Selection

( Jung Pil Suh ),( Young Chan Cho ),( Yong Jae Won ),( Eok Keun Ahn ),( Man Kee Baek ),( Myeong Ki Kim ),( Bo Kyeong Kim ),( Kshirod K. Jena ) 한국육종학회 2015 Plant Breeding and Biotechnology Vol.3 No.4

Advances in plant molecular techniques have dramatically widened the applicability of gene identification and pyramiding valuable genes. This study was carried out to pyramid five resistance genes for biotic stress into the japonica rice cultivar using marker-assisted selection (MAS) and marker-assisted background analysis of selected progenies using SSR markers. The Pi40, Xa4, xa5, Xa21 and Bph18 genes were combined in Jinbubyeo, a Korean japonica rice variety using MAS. Gene specific co-dominant PCR-based markers were used to select for homozygous recombinant lines in a segregating population derived from a cross between the parental homozygous resistant gene introgression lines. We had successfully developed multiple gene pyramided breeding lines (GPLs) for bacterial blight, blast, and brown planthopper using MAS in rice. The GPLs exhibited high resistance against biotic stress and had around 93% of the genetic background of the recurrent parent Jinbubyeo based on SSR graphical mapping. The yield and agronomic traits of the GPLs were similar to those of the recurrent parent, indicating that there is no apparent agronomic trait penalty associated with the presence of the resistance genes. The strategy of simultaneous foreground and phenotypic selection to introduce multiple R genes is very useful to reduce the cost and the time required for the isolation of desirable recombinants with target resistance genes in rice. The GPLs could be useful to enhance effective resistance for biotic stress and produce stable grain yield in japonica rice breeding programs.

유전자 발현량 차이를 이용한 네트워크 기반 질병 관련 유전자 탐색 기법

김현진,안재균,박상현 한국정보과학회 2012 데이타베이스 연구 Vol.28 No.3

One of general method for microarray analysis is discovering differentially expressed genes. The differentially expressed gene is a gene which shows different expression levels between two conditions. However, existing methods for finding differentially expressed genes have a limitation. They cannot consider influences among genes. Specifically, they can hardly discover the biologically proved genes which are related to specific diseases. To get over the limitation, we propose a novel approach to discover disease-related genes using differentially expressed genes and protein-protein interaction network. The proposed approach uses protein-protein interaction to reflect the influences among genes. The approach showed better accuracy and AUC(Area Under Curve) value than a method which does not consider the influences among genes and showed lower p-value than other feature selection methods. 마이크로어레이 데이터를 분석하는 대표적인 방법 중 한가지는 차등 발현 유전자(Differentially Expressed Gene)들을 찾는 것이다. 차등 발현 유전자란 두 실험 조건 하에서 샘플 집합의 유전자 발현량이 많이 차이나는 유전자를 의미한다. 하지만 기존의 차등 발현 유전자를 찾는 방법들은 유전자끼리 주고 받는 영향을 고려하지 않아 근본적인 한계를 지니고 있다. 실제로 기존 방법들로, 질병과 관련되어 있다고 생물학적 실험으로 증명된 유전자들을 많이 찾아내지 못하고 있다. 이러한 한계를 극복하기 위해 본 연구에서는 유전자 사이의 상관관계를 고려한 단백질-단백질 상호작용(Protein-Protein Interaction) 네트워크를 적용하여 유전자 간에 미치는 영향을 기존 방법에 추가함으로써 차등 발현 유전자를 검색하는 새로운 방법을 제안한다. 이렇게 찾아낸유전자들로 질병과 관련된 클래스 분류를 시도한 결과, 기존의 네트워크적 접근 방법을 적용하지 않은 차등발현 유전자를 찾는 방법보다 더 높은 정확도와 AUC(Area Under Curve)를 보였다. 또한 점수 값의 상위에위치해있는 유전자들이 해당 질병과 얼마나 관련되어 있는지에 대해서 다른 특성 선택(Feature Selection)방법들과 비교해보았을 때 더 낮은 p-value를 나타냄으로써 본 연구의 방법이 질병 관련 유전자를 잘 검색한다는 사실을 보여주었다.