• Molecular & Cellular Toxicology
• /
• v.1 no.2
• /
• pp.92-98
• /
• 2005

DNA microarray analysis of gene expression in toxicogenomics typically requires relatively large amounts of total RNA. This limits the use of DNA microarray when the sample available is small. To confront this limitation, different methods of linear RNA amplification that generate antisense RNA (aRNA) have been optimized for microarray use. The target preparation strategy using amplified RNA in DNA microarray protocol can be divided into direct-incorporation labeling which resulted in cDNA targets (Cy-dye labeled cDNA from aRNA) and indirect-labeling which resulted in cRNA targets (i.e. Cy-dye labeled aRNA), respectively. However, despite the common use of amplified targets (cDNA or cRNA) from aRNAs, no systemic assessment for the use of amplified targets and bias in terms of hybridization performance has been reported. In this investigation, we have compared the hybridization performance of cRNA targets with cDNA targets from aRNA on a 10 K cDNA microarrays. Under optimized hybridization conditions, we found that 43% of outliers from cDNA technique and 86% from the outlier genes were reproducibly detected by both targets hybridization onto cDNA microarray. This suggests that the cRNA labeling method may have a reduced capacity for detecting the differential gene expression when compared to the cDNA target preparation. However, further validation of this discordant result should be pursued to determine which techniques possesses better accuracy in identifying truly differential genes.

• The Journal of the Korea Contents Association
• /
• v.20 no.3
• /
• pp.73-80
• /
• 2020

Using 8,499 observations from 5 years-1,878 firms panel data during 2014 to 2018 in Korean stock exchanges, this study examines the impact of National Pension's ownership on corporate philanthropic giving. The empirical study finds that National Pension's ownership has positive relations with the extent of corporate philanthropic giving in terms of the amounts per employee, the expenditures with respect to total asset and total sales, implying that National Pension plays a monitoring role in promoting target firms to increase the extent of corporate philanthropic giving, which lead to increase in target firms' long-term values. The empirical study also finds that when National Pension is a blockholder holding more than 5% ownership in the target firms, it has positive relations with the extent of corporate philanthropic giving in terms of the amounts per employee, the expenditures with respect to total asset, implying that it exercises disciplinary roles on focal firms in promoting the extent of corporate philanthropic giving in order to increase target firms' long-term values. The results overall support that National Pension plays positive effects on target firms in promoting the extent of corporate philanthropic giving, which lead to increase in target firms' long-term values.

• Journal of the Korean Applied Science and Technology
• /
• v.26 no.1
• /
• pp.38-44
• /
• 2009

We proposed the new nano-carbon ball (NCB) materials for eliminating the total volatile organic compounds(TVOCs) from the felt which is built in the car. The concentrations of acetaldehyde and formaldehyde of the original felts were varied upon the different production lots. Acetaldehyde in the felt can be eliminated to target level($0.2{\mu}g$) after introducing 0.5 wt% of NCB into the felt. Detector tube method for analyzing formaldehyde gas was more accurate than HPLC method. Formaldehyde can be eliminated to target level (64 ppb) after introducing 0.5 wt% of NCB into the felt. We also found that TVOC can be reduced to target level ($0.32{\mu}g$) after introducing 2.0 wt% of NCB. Upon introducing small amounts of NCB into the felt, it was possible that the level of formaldehyde, acetaldehyde and TVOC formed from the felts can be reduced to the target level. We also suggest the effective analyzing method of TVOCs.

• Proceedings of the EASDL Conference
• /
• 2004.10a
• /
• pp.13-30
• /
• 2004

There are two groups of significant functional constituents in sesame seeds on the whole; one is the vegetable oils and another is the anti-oxidative compounds. However, although high amounts of major fatty acids are synthesized in sesame seeds, their composition is unfavorable because the contents of alpha- and gamma-linolenic acid, the essential fatty acids, are very low or do not produced in sesame seeds. So, to increase these fatty acids in sesame seeds, one strategy is to overexpress their genes, ${\omega}$-3 fatty acid desaturase for alpha-linolenic acid and delta-6 fatty acid desaturase for gamma-linolenid acid, in them. Another molecular target is to enhance alpha-tocopherol, vitamin E, because its content is very low in sesame seeds. The enzyme, gamma-tocopherol methyltransferase, catalyzes the conversion of gamma-tocophero to alpha-tocopherol. Overexpression of this enzyme in sesame seeds could be also a good molecular breeding target. Reduction of phytic acid is also another molecular target in sesame seeds because phosphorus pollution may be caused by its high content in sesame seeds. Accordingly, to do so, one of target enzymes could be myo-inositol 1-phosphate synthase which is a key regulatory enzyme in the pathway of phytic aicd biosyntheses. In this lecture, a molecular strategy for development of value-added sesame crop is described in association with some results of our experiments involved in the molecular characterizations of the genes mentioned above.

• Journal of the Korean Magnetic Resonance Society
• /
• v.23 no.4
• /
• pp.104-107
• /
• 2019

Many proteins are expressed as an insoluble form during the production using Escherichia coli (E. coli) system. Although various methods are applied to increase their amounts of soluble expression, refolding is the only feasible way to obtain a target protein in some cases. Moreover, protein NMR experiments require ¹³C/¹⁵N-labeled proteins that can only be obtained from E. coli systems in terms of cost and technical difficulty. The finding of appropriate refolding conditions for a target protein is a time-consuming process. In particular, it is very difficult to determine whether the refolded protein has a native structure, when a target protein has no enzymatic activity and its refolding yield is very low. Here, we showed that 1-dimensional ¹H-¹⁵N heteronuclear single quantum correlation (1D ¹H-¹⁵N HSQC) experiment can be efficiently used to screen an optimal condition for the refolding of a target protein by monitoring both the structure and concentration of the refolded protein.

• Korean journal of food and cookery science
• /
• v.26 no.1
• /
• pp.54-61
• /
• 2010

This study was performed to determine the optimal manufacturing conditions adding green laver, which has a rich aroma and unique taste, to Sulgidduk. The variables in green laver Sulgidduk production were amounts of green laver powder and water. Six sensory characteristics were used for sensory evaluations, including color, green laver flavor, green laver taste, moistness, softness, and elasticity. The optimal amounts of the powder and water were found to be 6.84 g for green laver powder and 31.08 g for water, satisfying a target sensory score (7.0/9.0) according to a response surface method. Sulgidduk with these optimal amounts of green laver and water was tasted by 118 consumers and showed a high acceptability score (6.94). Older consumers ($\geq$30 years old) preferred the color and flavor of green laver significantly more than younger consumers ($\leq$29 years old) did (p<0.01). Color analysis results of green laver Sulgidduk were significantly different in brightness, redness, and yellowness from those of Sulgidduk (control)(p<0.01). Texture analysis scores of green laver Sulgidduk were significantly lower than those of Sulgidduk (control) in hardness, adhesiveness, springiness, cohesiveness, gumminess, and chewiness, and they were significantly different in adhesiveness and chewiness (p<0.05).

• Phonetics and Speech Sciences
• /
• v.4 no.3
• /
• pp.111-117
• /
• 2012

In this paper, we compare the performance of several speaker adaptation methods for a HMM-based Korean speech synthesis system with small amounts of adaptation data. According to objective and subjective evaluations, a hybrid method of constrained structural maximum a posteriori linear regression (CSMAPLR) and maximum a posteriori (MAP) adaptation shows better performance than other methods, when only five minutes of adaptation data are available for the target speaker. During the objective evaluation, we find that the duration models are insufficiently adapted to the target speaker as the spectral envelope and pitch models. To alleviate the problem, we propose the duration rectification method and the duration interpolation method. Both the objective and subjective evaluations reveal that the incorporation of the proposed two methods into the conventional speaker adaptation method is effective in improving the performance of the duration model adaptation.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.881-882
• /
• 2008

This paper presents an algorithm of target detection of an underwater acoustic signal by estimating the background. At first, subtract the estimated background from the underwater acoustic signal. To estimate the background, this paper uses an algorithm of Denoising. By using Thresholding and Power analysis, we extract targets from the signal to eliminate the background. The proposed method is valuable as an algorithm to reduce calculation amounts of multi frames we will apply.

• 전자공학회논문지 IE
• /
• v.46 no.4
• /
• pp.42-48
• /
• 2009

In this paper, we propose adaptive near-lossless image coding algorithm according to bandwidth while maintaining image quality. The proposed method adjusts error range using amounts of encoded bits and target bits at a slice encoding interval. Experimental results show that our proposed method not only almost fits compression into bandwidth, but also has better image quality.

• Journal of information and communication convergence engineering
• /
• v.21 no.2
• /
• pp.139-144
• /
• 2023

Engineers prefer deep neural networks (DNNs) for solving computer vision problems. However, DNNs pose two major problems. First, neural networks require large amounts of well-labeled data for training. Second, the covariate shift problem is common in computer vision problems. Domain adaptation has been proposed to mitigate this problem. Recent work on adversarial-learning-based unsupervised domain adaptation (UDA) has explained transferability and enabled the model to learn robust features. Despite this advantage, current methods do not guarantee the distinguishability of the latent space unless they consider class-aware information of the target domain. Furthermore, source and target examples alone cannot efficiently extract domain-invariant features from the encoded spaces. To alleviate the problems of existing UDA methods, we propose the mixup regularization in adversarial discriminative domain adaptation (ADDA) method. We validated the effectiveness and generality of the proposed method by performing experiments under three adaptation scenarios: MNIST to USPS, SVHN to MNIST, and MNIST to MNIST-M.