[논문]Application of Random Forests to Association Studies Using Mitochondrial Single Nucleotide Polymorphisms

Kim, Yoon-Hee; Kim, Ho

[국내논문] Application of Random Forests to Association Studies Using Mitochondrial Single Nucleotide Polymorphisms 원문보기

Genomics & informatics, v.5 no.4, 2007년, pp.168 - 173

Kim, Yoon-Hee (Department of Biostatistics and Epidemiology, School of Public Health, Seoul National University) , Kim, Ho (Inherited Disease Research Branch, NHGRI)

Abstract ▼ AI-Helper

In previous nuclear genomic association studies, Random Forests (RF), one of several up-to-date machine learning methods, has been used successfully to generate evidence of association of genetic polymorphisms with diseases or other phenotypes. Compared with traditional statistical analytic methods, such as chi-square tests or logistic regression models, the RF method has advantages in handling large numbers of predictor variables and examining gene-gene interactions without a specific model. Here, we applied the RF method to find the association between mitochondrial single nucleotide polymorphisms (mtSNPs) and diabetes risk. The results from a chi-square test validated the usage of RF for association studies using mtDNA. Indexes of important variables such as the Gini index and mean decrease in accuracy index performed well compared with chi-square tests in favor of finding mtSNPs associated with a real disease example, type 2 diabetes.

주제어

AI 본문요약
AI-Helper

* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.

문제 정의

In previous nuclear genomic studies, we hypothesized that RF is appropriate for association studies using mtSNPs with unique characteristics such as haploid number and lack of recombination, unlike nuclear SNPs. To our knowledge, this paper is the first application of RF to investigate the association of mtSNPs with disease. We validate the usage of RF compared with the results from the chi-square test using example data searching the association between mtDNA and type 2 diabetes.

가설 설정

3. Iterate steps 2-1 and 2-2 until the tree is fully grown (no pruning).
In this paper, we investigated the feasibility of RF as a tool for detection of association using the mitochondrial genome. Because mtDNA does not undergo recombination, which may lead to a lack of independence between mtSNP sites, RF methods that do not require strong independence assumptions among predictor variables are particularly applicable to mtDNA markers.

제안 방법

Selection of non-diabetes was based on the following criteria: matched age with diabetes cases, no past history of diabetes, no diabetes in first-degree relatives. One hundred thirty-two mitochondrial biallelic SNPs among whole mtDNA were selected for subsequent analyses using Restriction Fragment Length Polymorphism (RFLP). We used the revised Cambridge Reference Sequence (rCRS, (Chinnery et al.
Separation of the cases from the controls based on the models derived from the RF procedure was quite good. Concerning gene x gene interaction, we chose the simplest interaction model, the two-locus interaction model, for multivariate logistic regression analyses (₁₃₂C₂= 8,646; number of two-locus interaction terms among 132 mtSNPs, data not shown). However, we cannot be sure about only the efficiency of this analysis using never-ending terms in a model, but also the interpretation of results.
, 1986) and a logistic regression method to our example data (data not shown); thus, only bp4985 was detected as a predictor because of the perfect contrast of observations between cases and controls. When we performed multivariate logistic regression model including bp4985 to assess the relative effect of each SNP, we could not estimate the effects of other mtSNPs except for bp4985 owing to too much information of only this mtSNP in a regression model. In this case, information on other mtSNPs, which had a relatively small effect to big effect from the only perfect one mtSNP, cannot be provided to researchers.

대상 데이터

Demographic information of example data is not consented to publish. One hundred thirty unrelated patients with type 2 diabetes and 65 well-matched, unaffected control subjects were used in the analyses. Type 2 diabetes was diagnosed according to World Health Organization criteria.

데이터처리

Thus, we can generate a 2 x 2 table of mtSNP versus disease status at the test locus directly. Fisher’s exact chi-square tests and univariate logistic regression analyses for calculating odds ratios were performed using SAS v9.1 for each mtSNP site to detect mtSNPs that show different allele frequency distributions between cases and controls.
The Gini index for a given variable is the sum over all trees of the decrease in Gini impurity after each split that involved that variable. We validated these measures with p-values from the Fisher’s exact chi-square tests. We have plotted the negative log p values and the Gini index values (Fig.
To detect which mtSNP associates with disease, Table 1 shows the 6 significant mtSNPs (p values < 0.0001) among 132 mtSNPs using Fisher’s exact test and odds ratio estimates from logistic regression. Due to zero cells in the contingency table except for bp10310, odds ratios were not calculated for 5 mtSNPs.
Under the goal of this paper, we validated the RF results compared with Fisher’s exact test. We plotted the scatter plot between Gini index from RF and negative p-values from chi-square tests of 132 mtSNP sites in Fig.

이론/모형

One hundred thirty-two mitochondrial biallelic SNPs among whole mtDNA were selected for subsequent analyses using Restriction Fragment Length Polymorphism (RFLP). We used the revised Cambridge Reference Sequence (rCRS, (Chinnery et al., 1999)) to denote which allele is the “common” allele at each mtSNP locus. Interestingly, one of mtSNPs, bp4985, has all variant alleles in type 2 diabetes patient cases and all common alleles in unaffected controls.

성능/효과

2, adding up the Gini decrease for each individual variable over all trees). Both the accuracy measure and the Gini index detected the 6 mtSNPs (red boxes), which had significant p-values less than 0.0001 for the Fisher’s exact test within the 20 most important variables.

참고문헌 (26)

Bureau, A., Dupuis, J., Falls, K., Lunetta, K.L., Hayward, B., Keith, T.P., and Van Eerdewegh, P. (2005). Identifying SNPs predictive of phenotype using random forests. Genet Epidemiol. 28, 171-82

상세보기
Bureau, A., Dupuis, J., Hayward, B., Falls, K., and Van Eerdewegh, P. (2003). Mapping complex traits using Random Forests. BMC Genet. 4 Suppl 1, S64

상세보기
Burger, G., Gray, M.W., and Lang, B.F. (2003). Mitochondrial genomes: anything goes. Trends Genet. 19, 709-16

상세보기
Chinnery, P.F., Howell, N., Andrews, R.M., and Turnbull, D.M. (1999). Clinical mitochondrial genetics. J Med Genet. 36, 425-36

상세보기
Cho, Y.M., Ritchie, M.D., Moore, J.H., Park, J.Y., Lee, K.U., Shin, H.D., Lee, H.K., and Park, K.S. (2004). Multifactor-dimensionality reduction shows a two-locus interaction associated with Type 2 diabetes mellitus. Diabetologia. 47, 549-54

상세보기
Diaz-Uriarte, R., and Alvarez de Andres, S. (2006). Gene selection and classification of microarray data using random forest. BMC Bioinformatics. 7, 3

상세보기
Grajski, K. A., Breiman, L., Viana Di Prisco, G., and Freeman, W.J. (1986). Classification of EEG spatial patterns with a tree-structured methodology: CART. IEEE Trans Biomed Eng. 33, 1076-86

상세보기
Guo, L.J., Oshida, Y., Fuku, N., Takeyasu, T., Fujita, Y., Kurata, M., Sato, Y., Ito, M., and Tanaka, M. (2005). Mitochondrial genome polymorphisms associated with type-2 diabetes or obesity. Mitochondrion. 5, 15-33

상세보기
Ingman, M., Kaessmann, H., Paabo, S., and Gyllensten, U. (2000). Mitochondrial genome variation and the origin of modern humans. Nature 408, 708-13

상세보기
Kahn, C.R., Vicent, D., and Doria, A. (1996). Genetics of non-insulin-dependent (type-II) diabetes mellitus. Annu Rev Med. 47, 509-31

상세보기
Kato, Y., Miura, Y., Inagaki, A., Itatsu, T., and Oiso, Y. (2002). Age of onset possibly associated with the degree of heteroplasmy in two male siblings with diabetes mellitus having an A to G transition at 3243 of mitochondrial DNA. Diabet Med. 19, 784-6

상세보기
Ladoukakis, E.D., and Eyre-Walker, A. (2004). Evolutionary genetics: direct evidence of recombination in human mitochondrial DNA. Heredity. 93, 321

상세보기
Lee, J.W., Lee, J.B., Park, M., and Song, S.H. (2005). An extensive comparison of recent classification tools applied to microarray data. Comp Stat & Data Analysis. 48, 869-885

상세보기
Lunetta, K.L., Hayward, L.B., Segal, J., and Van Eerdewegh, P. (2004). Screening large-scale association study data: exploiting interactions using random forests. BMC Genet. 5, 32

상세보기
Matsunaga, H., Tanaka, Y., Tanaka, M., Gong, J.S., Zhang, J., Nomiyama, T., Ogawa, O., Ogihara, T., Yamada, Y., Yagi, K., and Kawamori, R. (2001). Antiatherogenic mitochondrial genotype in patients with type 2 diabetes. Diabetes Care. 24, 500-3

상세보기
McKinney, B.A., Reif, D.M., Ritchie, M.D., and Moore, J.H. (2006). Machine learning for detecting gene-gene interactions: a review. Appl Bioinformatics. 5, 77-88

상세보기
Mukae, S., Aoki, S., Itoh, S., Sato, R., Nishio, K., Iwata, T., and Katagiri, T. (2003). Mitochondrial 5178A/C genotype is associated with acute myocardial infarction. Circ J. 67, 16-20

상세보기
Niemi, A.K., Hervonen, A., Hurme, M., Karhunen, P.J., Jylha, M., and Majamaa, K. (2003). Mitochondrial DNA polymorphisms associated with longevity in a Finnish population. Hum Genet. 112, 29-33

상세보기
Nigou, M., Parfait, B., Clauser, E., and Olivier, J.L. (1998). Detection and quantification of the A3243G mutation of mitochondrial DNA by ligation detection reaction. Mol Cell Probes. 12, 273-82

상세보기
Ohkubo, K., Yamano, A., Nagashima, M., Mori, Y., Anzai, K., Akehi, Y., Nomiyama, R., Asano, T., Urae, A., and Ono, J. (2001). Mitochondrial gene mutations in the tRNA(Leu(UUR)) region and diabetes: prevalence and clinical phenotypes in Japan. Clin Chem. 47, 1641-8

상세보기
Park, H.S., and Lee, S.U. (2004). MitGEN: Single Nucleotide Polymorphism DB Browser for Human Mitochondrial Genome. Genomics & Informatics 2(3), 147-148

원문보기 상세보기
Poulton, J., Luan, J., Macaulay, V., Hennings, S., Mitchell, J., and Wareham, N.J. (2002). Type 2 diabetes is associated with a common mitochondrial variant: evidence from a population-based case-control study. Hum Mol Genet. 11, 1581-3

상세보기
Rosenbloom, A.L., Joe, J.R., Young, R.S., and Winter, W.E. (1999). Emerging epidemic of type 2 diabetes in youth. Diabetes Care. 22, 345-54

상세보기
Shi, T., Seligson, D., Belldegrun, A.S., Palotie, A., and Horvath, S. (2005). Tumor classification by tissue microarray profiling: random forest clustering applied to renal cell carcinoma. Mod Pathol, 18, 547-57

상세보기
Suzuki, S. (2004). Diabetes mellitus with mitochondrial gene mutations in Japan. Ann N Y Acad Sci. 1011, 185-92

상세보기
Suzuki, S., Oka, Y., Kadowaki, T., Kanatsuka, A., Kuzuya, T., Kobayashi, M., Sanke, T., Seino, Y., and Nanjo, K. (2003). Clinical features of diabetes mellitus with the mitochondrial DNA 3243 (A-G) mutation in Japanese: maternal inheritance and mitochondria-related complications. Diabetes Res Clin Pract. 59, 207-17

상세보기

저자의 다른 논문 :

LOADING...

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증