• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.289-291
• /
• 2005

최근 유전자 칩의 발전으로 다양하고 방대한 양의 유전자 정보를 이용한 정확하고 신뢰성 높은 분류, 군집 및 질병을 예측하는 분석 기법이 증가하고 있다. 하지만 특징적인 유전자를 선택하는 Gene Selection 기법의 종류는 많지가 않으며 주로 통계적인 방법에 의존하여 유전자를 선택하는 기법을 많이 사용하고 있다. 본 논문에서는 유전자 알고리즘과 신경망의 결합을 통한 데이터마이닝을 기반으로 신뢰성 높은 특징적인 유전자를 선택하는 Gene Selection 기법에 대하여 연구을 진행하였다.

• Research in Plant Disease
• /
• v.14 no.3
• /
• pp.165-170
• /
• 2008

One hundred and three isolates of Xanthomonas oryzae pv. oryzae in Korea were evaluated for their virulence on four near-isogenic lines (NILs) containing a single resistance gene, and Korean differential varieties. The resistant gene backgrounds of Cheongcheongbyeo, Pungsanbyeo, Hangangchalbyeo, Milyang42 were not completely understood and they were not suited for the classification of X. oryzae pv. oryzae. Four NILs, IRBB101, IRBB103, IRBB105, and IRBB107 were difference for characterizing races of X. oryzae pv. oryzae because they have a single resistance gene. These NILs may be useful differential set in examining pathogenic races of X. oryzae pv. oryzae in Korea. Based on the virulence of 103 isolates to new differential varieties, they were classified into 3 races.

• Korean Journal of Microbiology
• /
• v.47 no.3
• /
• pp.268-274
• /
• 2011

Anabaena (Cyanobacteria, Nostocales) are important for water quality controls, because they are often responsible for freshwater green tides; moreover, some species are reported to produce hepatotoxin. In this study, we sequenced RNA polymerase beta subunit (rpoB) gene of Anabaena, and evaluated their sequences for the potential use of a molecular taxonomic marker in this taxon. Anabaena rpoB showed low DNA similarity and high genetic divergences when compared those of 16S rRNA, and the molecular differences were statistically significant (Student t-test, p<0.01). Parsimony analyses showed the rpoB gene evolves 4.8-fold faster than 16S rRNA. In addition, phylogeny of the rpoB gene separated each Anabaena strain more clearly compared with a 16S rRNA tree. These results suggest that the rpoB gene is a useful marker for the molecular phylogenetics and the species discrimination of Anabaena.

• Journal of KIISE:Software and Applications
• /
• v.28 no.11
• /
• pp.822-832
• /
• 2001

We have implemented an inductive learning system that learns PROSPECTOR-rule-style classification rules from sets of examples. In our a approach, a genetic algorithm is used in which a population consists of rule-sets and rule-sets generate offspring through the exchange of rules relying on genetic operators such as crossover, mutation, and inversion operators. In this paper, we describe our learning environment centering on the syntactic structure and meaning of classification rules, the structure of a population, and the implementation of genetic operators. We also present a method to evaluate the performance of rules and a heuristic approach to generate rules, which are developed to implement mutation operators more efficiently. Moreover, a method to construct a classification system using multiple learned rule-sets to enhance the performance of a classification system is also explained. The performance of our learning system is compared with other learning algorithms, such as neural networks and decision tree algorithms, using various data sets.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.447-450
• /
• 2005

오늘날 웹서비스에 대한 정보 검색 시스템들은 UDDI 레지스트리에 대한 전문적인 지식을 필요로 한다. 즉 웹서비스를 명세하기 위해 사용된 카테고리와 이에 대한 값, 이름 등을 사전에 알고 있어야만 검색이 가능하다. 그러나 일반 사용자들은 이러한 사전지식을 충분히 알고 있지 못하기 때문에 웹서비스에 대한 정보 검색이 쉽지가 않다. 그러므로 일반 사용자들을 위해 웹서비스에 대한 정보를 카테고리에 맞게 분류하면 웹서비스 검색을 보다 용이하게 할 수가 있다. 따라서 본 논문에서는 유전자 알고리즘과 신경망을 이용하여 보다 효율적으로 웹서비스 정보를 분류하는 클러스터링 기법을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.826-828
• /
• 2004

홍채인식 시스템은 영상획득, 전처리, 특징 추출, 패턴 정합의 단계로 이루어져 있다. 이 중 특징 추출은 특징 차원의 감소뿐만 아니라 분류 정착도의 증가를 위한 필수적인 과정이다. 본 논문에서는 특징을 추출하는데 있어서, 홍채데이타에 웨이블렛 변환의 다해상도 분석 기법을 시도하여 일정 영역을 추출한 후, 그 영역에 유전자 알고리즘(Genetic Algorithm)을 적용하여 가장 분별력 있는 특징들만을 추출 및 사용하는 홍채인식 시스템을 제안한다. 유전자 알고리즘의 선택연산자로는 적응도 비례 방식과 전역 엘리트 방식을 사용하였으며, 적합도 함수로는 Gaussian Kernel을 사용하는 Support Vector Machine(SVM)을 사용하였다. 본 시스템을 통해 나온 최적의 특징집합을 이용한 SVM분류기로 인식률을 알아본 결과 웨이블렛만을 사용했을 때 보다 대략 1.5%정도 더 좋은 인식률을 얻을 수 있었다.

• Journal of Life Science
• /
• v.13 no.1
• /
• pp.128-135
• /
• 2003

102 ESTs (Expressed Sequence Tags) were obtained by sequencing clones from a library of rainbow trout kidney cDNAs. Of the sequences generated, 55.8% of the ESTs were represented by 37 known genes. The 45 clones of unknown gene products potentially represent 40 novel genes. The genes involved in structural function (14.5%) and transcription/translation (11.6%) account for the major gene expression activities in the kidney Microarray experiment was conducted to compare gene expression of the unique ESTs in young and adult rainbow trout kidneys. While mitochondrion, cytochrome b, rho G, spastin protein, and three unknown genes were down-regulated in the mature fish kidney, calponin 1, calcium binding protein, histone deacetylase 1, and an unknown gene were up-regulated in the mature fish kidney. This research demonstrates the feasibility and power of functional genomics in rainbow trout.

• Korean Journal of Microbiology
• /
• v.48 no.4
• /
• pp.319-324
• /
• 2012

In this study, we investigated molecular divergences and phylogenetic characteristics of the 16S ribosomal RNA (rRNA) and RNA polymerase beta subunit (rpoB) gene sequences from the order Oscillatoriales (Cyanobacteria). The rpoB of Oscillatoriales showed higher genetic divergence when compared with those of 16S rRNA (p-distance: rpoB=0.270, 16S=0.109), and these differences were statistically significant (Student t-test, p<0.001). Phylogenetic trees of 16S rRNA and rpoB were generally compatible; however, rpoB tree clearly separated the compared Oscillatoriales taxa, with higher phylogenetic resolution. In addition, parsimony analyses showed that rpoB gene evolved 2.40-fold faster than 16S rRNA. These results suggest that the rpoB is a useful gene for the molecular phylogenetics and species discrimination in the order Oscillatoriales.

• Journal of KIISE
• /
• v.45 no.1
• /
• pp.61-68
• /
• 2018

The identification of genes that contribute to the prediction of prognosis in patients with cancer is one of the challenges in providing appropriate therapies. To find the prognostic genes, several classification models using gene expression data have been proposed. However, the prediction accuracy of cancer prognosis is limited due to the heterogeneity of cancer. In this paper, we integrate microarray data with biological network data using a modified PageRank algorithm to identify prognostic genes. We also predict the prognosis of patients with 6 cancer types (including breast carcinoma) using the K-Nearest Neighbor algorithm. Before we apply the modified PageRank, we separate samples by K-Means clustering to address the heterogeneity of cancer. The proposed algorithm showed better performance than traditional algorithms for prognosis. We were also able to identify cluster-specific biological processes using GO enrichment analysis.

• Korean Journal of Ecology and Environment
• /
• v.50 no.1
• /
• pp.137-147
• /
• 2017

The identity of toxin producers remains only hypothesis unless there were identified by strain isolation and analytical confirmation of both the cyanotoxin production and the genetic identity of the monoculture. The purposes of this study were to identify a morphologic and phylogenetic classification in Aphanizomenon flos-aquae strains isolated from the Nakdong River and to investigate the potential ability of the strains to produce toxins such as saxitoxin and cylindrospermopsin using target genes. The 16S rRNA and sxtA, sxtI, cyrA, cyrJ genes were analyzed on two strains (DGUC001, DGUC003) isolated from the Nakdong River. Morphological features of the strains were observed a shape of aggregated trichomes in parallel fascicles which can reach up to macroscopic size and a hyaline terminal cell without aerotope. In addition, the 16S rRNA phylogenetic analyses showed that the strains were identified as the same species with high genetic similarity of 98.4% and grouped within a monospecific andsupported cluster I of Aphanizomenon flos-aquae selected from GenBank of the NCBI. The cyrA and cyrJ genes encoding for the cylindrospermopsin-biosynthesis were not detected in the present study. The sxtA gene was in detected both the two strains, whereas the sxtI gene which had been suggested as a suitable molecular marker to detect saxitoxin-producing cyanobacteria was not found both the strains. Thus, the two strains isolated from Nakdong River were identified as the same species of Aphanizomenon flos-aquae Ralfs ex Bornet et Flahault 1888, the two strains were confirmed as potential non-producing strains of the saxitoxin and cylindrospermopsin.