2005 Fiscal Year Final Research Report Summary
High dimensional data mining of drug therapeutic effects by the analysis of single nucleotide polymorphism and chemical structure of drugs
| Project/Area Number |
14208032
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Kwansei Gakuin University |
|---|
Principal Investigator |
OKADA Takashi Kwansei Gakuin University, School of Science and Technology, Professor, 理工学部, 教授 (00103135)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ICHIISHI Eiichiro Tohoku University, New Industry Creation Hatchery Center, Associate Professor, 未来科学技術共同研究センター, 助教授 (20361206)
WASHIO Takashi Osaka University, The Institute of Scientific and Industrial Research, Associate Professor, 産業科学研究所, 助教授 (00192815)
OYAMA Mayumi Kwansei Gakuin University, School of Humanities, Professor, 文学部, 教授 (90103134)
|
|---|
| Project Period (FY) |
2002 – 2005
|
| Keywords | Data mining / Clinical database / Single nucleotide polymorphism / Cascade model / CVP450 / Structure-activity relationships / Numerical association rule / Fingertip pulse wave |
|---|
| Research Abstract |
The objectives of this research project are to develop a methodology of data mining for high dimensional data, and to apply it to the analysis of structure-activity relationships of drugs, the discovery of important genes with SNPs for medical care as well as finding essential factors influencing the quality of patients' life. Various experiments are done to accomplish the research, including wet lab works using DNA microarray. Several results achieved to date are shown below.
1.Attribute generation system from the chemical structure of drugs is developed to give many linear fragments. The attribute selecting scheme is also introduced to facilitate the effective and efficient mining.
2.The cascade model, a mining method developed by Okada, was extended to organize rules into principal and relative rules. Topographical expression of rules was also introduced to provide the better understanding of datascape.
3.The application of the above method to Dopamine receptor ligands resulted in the discovery of agonist and antagonist pharmacophores.
4.Microarray experiments of insulin-resistant diabetic patients led to the identification relevant SNPs. Further, important genes were discovered, which are evoked by locomotor stimulation, Helicobacter pylori and retinoic acid used to prevent cancer.
5.Software has been developed to evaluate SNPs effects in receptor-ligand interaction and transcription to RNA.
6.Clinical records in the hospital for the old are collected and their CYP450 SNPs were investigated. The analysis has shown that pharmaceutical interactions are important for the patients' quality of life.
7.A method was developed to mine numerical association rules and to utilize them for the classification.
8.Measurement of fingertip pulse waves was established as a simple and convenient method to analyze the human's psychological conditions.
|
