• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.967-969
• /
• 2005

유전자의 발현은 전사인자와 전사인자 결합부위의 결함에 의해 조절된다. 따라서 이러한 결합부위를 예측하는 것은 유전학 분야에서 중요한 이슈이다. 본 논문에서는 접미사 배열을 이용하여 전사인자가 결합할 것으로 예상되는 DNA 서열들의 공통서열을 추출하고, 이를 다시 입력 서열과 부분 정렬을 수행함으로써 전사인자가 결합하는 부위를 예측하는 알고리즘을 제시한다. 그리고 알려진 전사인자 결합부위를 가진 데이터로 실험한 결과를 통해 제시된 추출 방법의 성능에 대하여 논의한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.435-438
• /
• 2006

Single Nucleotide Polymorphism(SNP)는 인간 유전자 서열의 0.1%에 해당하는 부분으로 이는 각 개인의 체질 및 각종 유전질환과 밀접한 관련이 있다고 알려져 있으며 이 SNP 정보를 이용 각종 질환의 유전적 원인규명에 대한 많은 생물학적 연구가 진행되고 있다. 그러나 아직 SNP를 이용한 효율적인 분석방법에 대한 전산학적 연구는 많지 않다. 본 논문에서는 대표적인 패턴인식기 중 하나인 Support Vector Machine(SVM)을 이용 한국인의 대표적인 유전질환으로 알려진 위암에 대한 예측율을 실험하였다. 실험 데이터는 간 및 소화기 질환 유전체 센터에서 얻어진 위 질환 환자를 대상으로 하였으며 실험 결과 예측율은 67.3%로 이는 Case Based Reasoning(CBR)방법의 55% 보다 더 좋은 예측 결과를 보였다.

• Proceedings of the IEEK Conference
• /
• 2003.11b
• /
• pp.239-242
• /
• 2003

모든 생명체는 유전자의 최종 산물인 다양한 단백질들이 각각의 복잡한 기능을 수행함과 동시에 그들 사이의 긴밀한 상호작용에 의해 생명을 유지한다. 도메인 (Domain)은 단백질의 기능적 단위로서 한 개 단백질은 최대 수십 개의 도메인을 가지는데 이들 도메인에 대한 정보는 단백질의 기능을 예측하는데 도움이 될 수 있다. 본 논문에서는 종양을 억제하는 기능을 가지는 단백질과 그러한 기능을 가질 것으로 추정되어지는 단백질의 아미노산 서열, 또 기능이 밝혀지지 않은 미지의 아미노산 서열을 가지고 이미 밝혀져 있는 도메인 서열과 비교 검색하여 이들 사이에 일치하는 도메인을 통하여 표적 단백질의 기능 동정에 관한 연구에 도움이 되며, 또한 기능이 밝혀지지 않은 아미노산 서열의 도메인을 검색하여 새로운 기능을 예측함으로써 다른 실험적 방법과 비교하여 시간과 비용을 절약할 수 있는 효과적인 방법을 얻었기에 제안하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.433-434
• /
• 2008

기존의 실험을 통한 전통적인 생물학의 연구와는 달리, 미세 RNA (microRNA)의 연구에 있어 컴퓨터를 통한 프로그램 개발과 정보기술의 이용은 필수 불가결한 요소가 되었다. 컴퓨터를 바탕으로 한 대부분의 연구는 미세 RNA를 발현하는 유전자와 미세 RNA의 타겟 (target)을 예측하는 두가지 분야로 나누어 이루어지고 있다. 본 연구에서는 미세 RNA의 타겟을 예측하는 프로그램 개발시 이용되는 몇 가지 원칙들과 그 원칙들의 문제점을 서술하며, 현재 인터넷상에서 이용 가능한 프로그램들을 소개한다. 또한 컴퓨터를 통해 예측된 미세 RNA 의 타겟을 실험을 통해 검증하는 방법에 대해 논한다.

• Tuberculosis and Respiratory Diseases
• /
• v.64 no.4
• /
• pp.285-292
• /
• 2008

Background: A diagnosis of malignant pleural effusion is clinically important, as the prognosis of lung cancer patients with malignant pleural effusion is poor. The diagnosis will be difficult if a cytological test is negative. This study was performed to investigate whether the detection of hypermethylation of the p16 (CDKN2A) and retinoic acid receptor b2 (RARB2) genes in pleural fluid is useful for a diagnosis of malignant pleural effusion. Methods: Pleural effusion was collected from 43 patients and was investigated for the aberrant promoter methylation of the RARB2 and CDKN2A genes by use of methylation-specific PCR. Results were compared with findings from a pleural biopsy and from pleural fluid cytology. Results: Of 43 cases, 17 cases of pleural effusion were due to benign diseases, and 26 cases were from lung cancer patients with malignant pleural effusion. Hypermethylation of the RARB2 and CDKN2A genes was not detected in the case of benign diseases, independent of whether or not the patients had ever smoked. In 26 cases of malignant pleural effusion, hypermethylation of RARB2, CDKN2A or either of these genes was detected in 14, 5 and 15 cases, respectively. The sensitivities of a pleural biopsy, pleural fluid cytology, hypermethylation of RARB2, hypermethylation of CDKN2A, or hypermethylation of either of the genes were 73.1%, 53.8%, 53.8%, 19.2%, and 57.7%, respectively; negative predictive values were 70.8%, 58.6%, 58.6%, 44.7%, and 60.7%, respectively. If both genes are considered together, the sensitivity and negative predictive value was lower than that for a pleural biopsy, but higher than that for pleural fluid cytology. The sensitivity of hypermethylation of the RARB2 gene for malignant pleural effusion was lower in small cell lung cancers than in non-small cell lung cancers. Conclusion: These results demonstrate that detection of hypermethylation of the RARB2 and CDKN2A genes showed a high specificity, and sensitivity was higher than for pleural fluid cytology. With a better understanding of the pathogenesis of lung cancer according to histological types at the molecular level, and if appropriate genes are selected for hypermethylation testing, more precise results may be obtained.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.1039-1042
• /
• 2006

Because many corporations computerize process figure enemy who is introducing eCRM fast and are used mainly at past by purpose to detect and analyze and forecast systematic analysis of customer information and various pattern of customer recently, ordinary peoples are trend that is alternated gradually by data mining that can drawand forecast result of good quality easily. Field that this data mining is used representatively is eCRM. In this treatise customer data of A discount store and sale data of 1 years experimenting that forecast customer contribution to base next year through data mining actuality data and data mining through comparison with predicted data are how effective to eCRM prove.

• Journal of Life Science
• /
• v.18 no.7
• /
• pp.1011-1014
• /
• 2008

The ability or inability to taste phenylthiocarbamide (PTC) is a classic inherited trait that has been best-studied in human populations. Also, variation in PTC perception has been correlated with dietary preferences and thus may have important consequence for diet-related diseases in modem populations. The recent identification of the TAS2R38 gene (PTC gene) which is a member of TAS2R family of bitter taste receptor genes and three common polymorphisms in the gene is highly correlated with taste sensitivity to PTC. Balancing natural selection has acted to maintain high frequency of both alleles of the gene in human population. Future detailed studies of the relationships between molecular mechanisms and taste function may have therapeutic implications, such as helping patients to consume beneficial bitter-tasting compounds.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2006.11a
• /
• pp.285-288
• /
• 2006

Genetic Algorithms(GA) which searches optimum solution using natural selection and the law of heredity has been applied to teaming algorithms in order to estimate performance deterioration of the aircraft gas turbine engine. The compressor, gas generation turbine and power turbine are considered for estimation for performance deterioration of a complex component at design point was conducted. As a result of that, complex defect diagnostics has been conducted. As a result, the accuracy of diagnostics were verified with its relative error with in 10% at each component.

• Journal of the Korean Geotechnical Society
• /
• v.18 no.6
• /
• pp.117-127
• /
• 2002

A deterministic approach of slope stability, which is generally corresponding to the model of a simple non-linear function for slopes, is problematic in that it does not account the versatile characteristics of ground layers in an effective way. To resolve this problem, this study proposes a new way of analyzing slope stability, so-called “genetic algorithm method, ” so as to reflect some particular conditions pertaining to the grounds under concern. Similarities and differences in slope stability that may exist between homogeneous and multiple ground layers are examined in a competitive manner, Overall, though similarities deemed a little bit salient, the algorithm method turned out to be very applicable to estimating the validity of slope stability. Furthermore, an additional effort to consider long-standing sequential and dynamic changes in both the amount of rainfall and the underground water level is made in order to improve the results.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.9
• /
• pp.25-34
• /
• 2005

This paper proposes a selection methodology composed of neural network (NN) and genetic algorithm (GA) to design optimal NN structure. We combine the characteristics of GA and NN to reduce the computational complexity of artificial intelligence applications and increase the precision of NN' prediction in the design of NN structure. Genetic selection approach of design parameters of NN is introduced to obtain optimal NN structure. Analyzed results for geotechnical problems are given to evaluate the performance of the proposed hybrid methodology.