• Journal of Physiology & Pathology in Korean Medicine
• /
• v.19 no.4
• /
• pp.913-921
• /
• 2005

In this study, we investigated gene expression patterns induced by Ailanthus altissima extract and compared it with Gleevec, a well-known anti-leukemia drug, in K562 chromic leukemia cells. Ailanthus altissima extract(100 ug/ml) and Gleevec(50 ug/ml) were treated to cells for 1h, 2h, 4h, and 16h and total RNA was extracted. Gene expressions were evaluated using cDMA microarray, in which 24,000 genes were spotted. Hierarchical clustering analysis showed that expression of genes included in two clusters were increased or decreased time dependently by both Ailanthus altissima extract and Gleevec. Genes included in another cluster were induced by Ailanthus altissima extract but not by Gleevec. In biological process analysis, expression of genes involved in apoptosis, growth arrest and DNA-damage were increased, but genes stimulating cell cycle were decreased. This study provides comprehensive comparison of the patterns of gene expression changes induced by Ailanthus altissima extract and Gleevec in K-562 leukemia cells.

• Molecular & Cellular Toxicology
• /
• v.1 no.4
• /
• pp.275-280
• /
• 2005

Carbon tetrachloride ($CCl_4$) is well known hepatotoxicant. Its overdose cause severe centrilobular hepatic necrosis in human and experimental animals. We administered $CCl_{4}$ at low (0.2 mL/kg p.o.) and high (2 mL/kg p.o.) doses to mice. Mice were sacrificed at 24 h after administration. We evaluated liver toxicity by serum AST and ALT level and by microscopic observation. Using cDNA chip, we conducted gene expression analysis in liver. Mean serum activities of the hepatocellular leakage enzymes, ALT and AST, were significantly increased compare to control, respectively, in the low and high dose groups. H&E evaluation of stained liver sections revealed $CCl_{4}-related$ histopathological findings in mice. Moderate centrilobular hepatocellular necrosis was present in all $CCl_{4}$ treated mice. We found that gene expression pattern was very similar between low and high dose group. However, some stress related genes were differently expressed. These results could be a molecular signature for the degree of liver injury. Our data suggest that the degree of severity could be figure out by gene expression profiling.

• Environmental Analysis Health and Toxicology
• /
• v.21 no.1 s.52
• /
• pp.71-79
• /
• 2006

Epidemiologic studies have showed a close correlation between arsenic exposure and heart disease such as, cardiovascular problem, ischemic heart disease, infarction, atherosclerosis and hypertension in human. It may increase the mortality of high risk group with heart disease. Regarding the mechanism studies of heart failure, blood vessel, vascular smooth muscle cells and endothelial cells have long been focused as the primary targets in arsenic exposure but there are only a few studies on the cardiomyocytes. In this study, the generation of oxidative stress by low dose of arsenic trioxide was investigated in rat cardiomyocytes. By direct measurement of reactive oxygen species and fluorescent microscopic observation using fluorescent dye 2',7'-dichlorofluorescin diacetate, reactive oxygen species were found to be generated without cell death, where cells are treated with 0.1 ppm arsenic for 24 hours. With the induction of reactive oxygen species, GSH level was decreased by the same treatment. However, DNA damage did not seem to be serious by DAPI staining, while high dose of arsenic (2 ppm for 24 hrs) caused fragmentation of DNA. To identify the molecular biomarkers of low-dose arsenic exposure, gene expression was also investigated with whole genome microarray. As results, 9,022 genes were up-regulated including heme oxygenase-l and glutathione S-transrerase, which are well-known biomarkers of oxidative stress. 9,404 genes were down-regulated including endothelial type gp 91-phox gene by the treatment of 0.1 ppm arsenic for 24 hours. This means that biological responses of cardiomyocytes may be altered by ROS induced by low level arsenic without cell death, and this alteration may be detected clearly by molecular biomarkers such as heme oxygenase-1.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.409-414
• /
• 2013

Naive Bayesian classifiers are a powerful and well-known type of classifiers that can be easily induced from a dataset of sample cases. However, the strong conditional independence assumptions can sometimes lead to weak classification performance. Normally, naive Bayesian classifiers use Gaussian distributions to handle continuous attributes and to represent the likelihood of the features conditioned on the classes. The probability density of attributes, however, is not always well fitted by a Gaussian distribution. Another eminent type of classifier is the neuro-fuzzy classifier, which can learn fuzzy rules and fuzzy sets using supervised learning. Since there are specific structural similarities between a neuro-fuzzy classifier and a naive Bayesian classifier, the purpose of this study is to apply learning distribution graphs constructed by a neuro-fuzzy network to naive Bayesian classifiers. We compare the Gaussian distribution graphs with the fuzzy distribution graphs for the naive Bayesian classifier. We applied these two types of distribution graphs to classify leukemia and colon DNA microarray data sets. The results demonstrate that a naive Bayesian classifier with fuzzy distribution graphs is more reliable than that with Gaussian distribution graphs.

• Archives of Plastic Surgery
• /
• v.37 no.4
• /
• pp.317-322
• /
• 2010

Purpose: There is no clear evidence of the original cause of hypertrophic scar, and the effective method of treatment is not yet established. Recently the steps of searching in gene and molecular level are proceeding. we are trying to recognize the difference between keratinocytes of hypertrophic scar and normal skin. Then we do support the comprehension of the scar formation mechanism and scar management. Methods: Total RNAs were extracted from cultured keratinocytes from 4 hypertrophic scars and normal skins. The cDNA chips were prepared. A total of 3063 cDNAs from human cDNA library were arrayed. And the scanning data were analyzed. Results: On microarray, heat shock protein, pyruvate kinase, tumor rejection antigen were more than 2 fold intensity genes. Among them, heat shock 70 kd protein showed the strongest intensity difference. Conclusion: In this study, it can be concluded that heat shock proteins play an important role in the process of wound healing and scar formation. This study provides basic biologic information for scar research. The new way of the prevention and treatment of scar formation would be introduced with further investigations.

• The Korean Journal of Applied Statistics
• /
• v.35 no.5
• /
• pp.667-684
• /
• 2022

Gene expression data present the level of mRNA abundance of each gene, and analyses of gene expressions have provided key ideas for understanding the mechanism of diseases and developing new drugs and therapies. Nowadays high-throughput technologies such as DNA microarray and RNA-sequencing enabled the simultaneous measurement of thousands of gene expressions, giving rise to a characteristic of gene expression data known as high dimensionality. Due to the high-dimensionality, learning models to analyze gene expression data are prone to overfitting problems, and to solve this issue, dimension reduction or feature selection techniques are commonly used as a preprocessing step. In particular, we can remove irrelevant and redundant genes and identify important genes using gene selection methods in the preprocessing step. Various gene selection methods have been developed in the context of machine learning so far. In this paper, we intensively review recent works on gene selection methods using machine learning approaches. In addition, the underlying difficulties with current gene selection methods as well as future research directions are discussed.

• Clinical and Experimental Reproductive Medicine
• /
• v.33 no.2
• /
• pp.125-132
• /
• 2006

Objective: In the previous study, we complied the differentially expressed genes during early folliculogenesis. Objective of the present study was to identify downstream target genes of transcription factors (TFs) using bioinformatics for selecting the target TFs among the gene lists for further functional analysis. Materials & Methods: By using bioinformatics tools, constituent domains were identified from database searches using Gene Ontology, MGI, and Entrez Gene. Downstream target proteins/genes of each TF were identified from database searches using TF database ($TRANSFAC^{(R)}$ 6.0) and eukaryotic promoter database (EPD). Results: DNA binding and trans-activation domains of all TFs listed previously were identified, and the list of downstream target proteins/genes was obtained from searches of TF database and promoter database. Based on the known function of identified downstream genes and the domains, 3 (HNF4, PPARg, and TBX2) out of 26 TFs were selected for further functional analysis. The genes of wee1-like protein kinase and p21WAF1 (cdk inhibitor) were identified as potential downstream target genes of HNF4 and TBX2, respectively. PPARg, through protein-protein interaction with other protein partners, acts as a transcription regulator of genes of EGFR, p21WAF1, cycD1, p53, and VEGF. Among the selected 3 TFs, further study is in progress for HNF4 and TBX2, since wee1-like protein kinase and cdk inhibitor may involved in regulating maturation promoting factor (MPF) activity during early folliculogenesis. Conclusions: Approach used in the present study, in silico analysis of downstream target genes, was useful for analyzing list of TFs obtained from high-throughput cDNA microarray study. To verify its binding and functions of the selected TFs in early folliculogenesis, EMSA and further relevant characterizations are under investigation.

• Horticultural Science & Technology
• /
• v.33 no.4
• /
• pp.575-584
• /
• 2015

Drought stress is a crucial environmental factor determining crop survival and productivity. The goal of this study was to clearly identify a new drought stress-tolerance gene in Brassica rapa. From KBGP-24K microarray data with the B. rapa ssp. pekinensis inbred line 'Chiifu' under drought stress treatment, a gene which was named BrDSR (B. rapa Drought Stress Resistance) was chosen among 738 drought-responsive unigenes. BrDSR function has yet to be determined, but its expression was induced over 6-fold by drought. To characterize BrDSR, the gene was isolated from B. rapa inbred line 'CT001' and found to contain a 438-bp open reading frame encoding a 145 amino acid protein. The full-length cDNA of BrDSR was used to construct an over-expression vector, 'pSL100'. Tobacco transformation was then conducted to analyze whether the BrDSR gene can increase drought tolerance in plants. The BrDSR expression level in T1 transgenic tobacco plants selected via PCR and DNA blot analyses was up to 2.6-fold higher than non-transgenic tobacco. Analysis of phenotype clearly showed that BrDSR-expressing tobacco plants exhibited more tolerance than wild type under 10 d drought stress. Taking all of these findings together, we expect that BrDSR functions effectively in plant growth and survival of drought stress conditions.

• Journal of the Optical Society of Korea
• /
• v.10 no.3
• /
• pp.130-142
• /
• 2006

Lab-on-a-chip technology is attracting great interest because the miniaturization of reaction systems offers practical advantages over classical bench-top chemical systems. Rapid mixing of the fluids flowing through a microchannel is very important for various applications of microfluidic systems. In addition, highly sensitive on-chip detection techniques are essential for the in situ monitoring of chemical reactions because the detection volume in a channel is extremely small. Recently, a confocal surface enhanced Raman spectroscopic (SERS) technique, for the highly sensitive biological analysis in a microfluidic sensor, has been developed in our research group. Here, a highly precise quantitative measurement can be obtained if continuous flow and homogeneous mixing condition between analytes and silver nano-colloids are maintained. Recently, we also reported a new analytical method of DNA hybridization involving a PDMS microfluidic sensor using fluorescence energy transfer (FRET). This method overcomes many of the drawbacks of microarray chips, such as long hybridization times and inconvenient immobilization procedures. In this paper, our recent applications of the confocal Raman/fluorescence microscopic technology to a highly sensitive lab-on-a-chip detection will be reviewed.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.05a
• /
• pp.91-97
• /
• 2002

최근의 생물학 연구를 위한 기기의 자동화 및 고속화는 생물학 관련 정보량의 급증을 가져오고 있다. 예를 들어, DNA chip에서 얻어지는 마이크로어레이(microarray)는 수천 종류의 유전자의 발현량을 동시에 측정한다. 이러한 기술들은 생물의 세포나 조직에서 일어나는 일련의 다양한 현상을 전체적으로 조망하는 관점에서 관찰할 수 있는 기회를 제공하고 있으며, 이를 통한 생명공학의 전반적인 발전이 기대되고 있다. 따라서 대량의 생물학 관련 정보의 분석이나 데이터 마이닝이 행해지고 있으며 이를 위한 대표적인 기법들로는 각종 클러스터링(clustering) 및 신경망 계열의 모델 등이 있다. 본 논문에서는 확률그래프모델의 하나인 베이지안망(Bayesian network)을 생물정보분석에 이용한다. 구체적으로 유전자 발현패턴과 약물의 활성패턴 및 암 종류 사이의 확률적 관계를 모델링한다. 이러한 모델은 NCI60 dataset(http://discover.nci.nih.gov)에서 베이지안망을 학습함으로써 구성된다. 분석의 대상이 되는 데이터가 sparse하기 때문에 발생하는 어려움들을 해결하기 위한 기법들이 제시되며 학습된 모델에 대한 검증은 이미 생물학적으로 확인되어 있는 사실과의 비교를 통해 이루어진다. 학습된 베이지안망 모델은 각각의 유전자 간, 혹은 유전자와 처리된 약물 간의 실제 생물학적 관계를 다수 표현하며, 이는 제시되는 방법이 생물학적으로 유의미한 가설을 데이터 분석을 통해 효율적으로 생성하는데 유용하게 활용될 수 있음을 보인다.