Our goal for this study was to compare different tagging methods, haplotype blocking and association testing using different sample populations for tag SNP with respect to the power. We found that there were no significant differences in estimated power between the two tag SNP samples, 50 and 100. Large association samples would have to be recruited in order to offset the lower power. Then we would give some advise according to figure 6 to 11 in every situation and sample. We considered about number of tag SNPs and power’s level to choose a tagging method with an association test that cost less and powerful. When we had association sample = 500, four methods of tag NP match with the multi-SNP+1-Block had power about the same either in case or control, so we selected haplotype+1-Block because it had less number of tag SNPs.
Although haplotype+CI had least number of tag SNPs in second situation, it’s power was lower than haplotype+1-Block about 10% in control and 40% in case. When matching with multi-SNP+SSLD we may select haplotype+1-Block in case and haplotype+CI of second situation in control. The powers in multi-SNP+CI and single-SNP were too small, so we would use the two methods to do association analysis. Then we consider about association sample = 200, we selected tag SNP method of SSLD match with multi-SNP+1-Block in second situation because it had the best power. Haplotype+CI in control and haplotype+1-Block in case were our choices when matching with multi-SNP+SSLD. These conclusions were applied only on rare allele, but other allele types may be not suitable. These conclusions will facilitate future decisions in this same population, and when analyze real data may make decisions refer to this.
Table 3. Association sample = 500cases-500controls in casual region Table 3.a. haplotype+CI and haplotype+SSLD using first situation
sample
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
tagger multi-SNP+1-Block 0.98 0.99 0.97 0.99
multi-SNP+SSLD 0.6 0.96 0.58 0.98
multi-SNP+CI 0.27 0.27 0.29 0.28
single-SNP 0.19 0.19 0.2 0.22
haplotype+CI multi-SNP+1-Block 0.99 0.99 0.99 0.99
multi-SNP+SSLD 0.98 0.98 0.98 0.98
multi-SNP+CI 0.29 0.27 0.29 0.23
single-SNP 0.2 0.19 0.22 0.21
haplotype+SSLD multi-SNP+1-Block 0.99 0.99 0.99 0.99
multi-SNP+SSLD 0.98 0.98 0.98 0.98
multi-SNP+CI 0.27 0.26 0.26 0.26
single-SNP 0.19 0.18 0.2 0.2
haplotype+1-Block multi-SNP+1-Block 0.95 0.97 0.97 0.99
multi-SNP+SSLD 0.57 0.89 0.7 0.95
multi-SNP+CI 0.26 0.21 0.23 0.26
single-SNP 0.21 0.2 0.18 0.21
Table 3.b. haplotype+CI and haplotype+SSLD using second situation.
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
haplotype+CI multi-SNP+1-Block 0.88 0.58 0.91 0.58
multi-SNP+SSLD 0.88 0.6 0.91 0.61
multi-SNP+CI 0.19 0.14 0.17 0.16
single-SNP 0.16 0.13 0.17 0.15
haplotype+SSLD multi-SNP+1-Block 0.99 0.99 0.99 0.99
multi-SNP+SSLD 0.98 0.98 0.98 0.97
multi-SNP+CI 0.26 0.25 0.26 0.26
single-SNP 0.2 0.21 0.21 0.2
Table 4. Association sample = 200cases-200controls in casual region Table 4.a. haplotype+CI and haplotype+SSLD used first situation
sample
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
tagger multi-SNP+1-Block 0.53 0.69 0.52 0.67
multi-SNP+SSLD 0.28 0.63 0.33 0.72
multi-SNP+CI 0.16 0.19 0.17 0.21
single-SNP 0.11 0.11 0.09 0.11
haplotype+CI multi-SNP+1-Block 0.68 0.69 0.68 0.68
multi-SNP+SSLD 0.73 0.71 0.75 0.73
multi-SNP+CI 0.21 0.18 0.21 0.22
single-SNP 0.09 0.1 0.1 0.11
haplotype+SSLD multi-SNP+1-Block 0.67 0.68 0.68 0.68
multi-SNP+SSLD 0.71 0.72 0.76 0.72
multi-SNP+CI 0.23 0.19 0.2 0.22
single-SNP 0.09 0.1 0.11 0.11
haplotype+1-Block multi-SNP+1-Block 0.48 0.63 0.56 0.64
multi-SNP+SSLD 0.41 0.67 0.42 0.71
multi-SNP+CI 0.11 0.14 0.15 0.17
single-SNP 0.1 0.09 0.11 0.11
Table 4.b. haplotype+CI and haplotype+SSLD using second situation.
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
haplotype+CI multi-SNP+1-Block 0.59 0.32 0.7 0.39
multi-SNP+SSLD 0.72 0.44 0.74 0.44
multi-SNP+CI 0.16 0.11 0.16 0.13
single-SNP 0.1 0.05 0.09 0.07
haplotype+SSLD multi-SNP+1-Block 0.7 0.67 0.67 0.68
multi-SNP+SSLD 0.72 0.69 0.74 0.7
multi-SNP+CI 0.23 0.18 0.19 0.22
single-SNP 0.1 0.11 0.1 0.09
Table 5. Association sample = 500cases-500controls in null region Table 5.a. haplotype+CI and haplotype+SSLD used first situation
sample
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
tagger multi-SNP+1-Block 0.05 0.04 0.05 0.05
multi-SNP+SSLD 0.05 0.04 0.07 0.05
multi-SNP+CI 0.04 0.08 0.06 0.06
single-SNP 0.05 0.06 0.06 0.07
haplotype+CI multi-SNP+1-Block 0.05 0.07 0.05 0.05
multi-SNP+SSLD 0.05 0.06 0.06 0.06
multi-SNP+CI 0.06 0.05 0.05 0.05
single-SNP 0.03 0.04 0.04 0.04
haplotype+SSLD multi-SNP+1-Block 0.05 0.06 0.05 0.05
multi-SNP+SSLD 0.06 0.05 0.06 0.06
multi-SNP+CI 0.05 0.07 0.05 0.05
single-SNP 0.04 0.05 0.04 0.04
haplotype+1-Block multi-SNP+1-Block 0.02 0.04 0.05 0.05
multi-SNP+SSLD 0.03 0.04 0.06 0.05
multi-SNP+CI 0.03 0.06 0.04 0.03
single-SNP 0.06 0.07 0.06 0.07
Table 5.b. haplotype+CI and haplotype+SSLD using second situation.
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
haplotype+CI multi-SNP+1-Block 0.05 0.07 0.07 0.06
multi-SNP+SSLD 0.06 0.12 0.11 0.07
multi-SNP+CI 0.04 0.09 0.06 0.07
single-SNP 0.04 0.06 0.05 0.05
haplotype+SSLD multi-SNP+1-Block 0.04 0.05 0.08 0.03
multi-SNP+SSLD 0.06 0.06 0.08 0.07
multi-SNP+CI 0.06 0.07 0.06 0.05
single-SNP 0.04 0.05 0.04 0.04
Table 6. Association sample = 500cases-500controls in null region Table 6.a. haplotype+CI and haplotype+SSLD used first situation
sample
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
tagger multi-SNP+1-Block 0 0 0.01 0
multi-SNP+SSLD 0.02 0.01 0.01 0
multi-SNP+CI 0.02 0.01 0.03 0.04
single-SNP 0.02 0.01 0.01 0.02
haplotype+CI multi-SNP+1-Block 0.02 0.02 0.01 0.01
multi-SNP+SSLD 0.01 0 0.01 0.02
multi-SNP+CI 0.01 0 0.03 0.03
single-SNP 0.02 0.01 0.01 0.01
haplotype+SSLD multi-SNP+1-Block 0.01 0.02 0 0
multi-SNP+SSLD 0.01 0 0.01 0.01
multi-SNP+CI 0.02 0.01 0.04 0.03
single-SNP 0.02 0.01 0.01 0.01
haplotype+1-Block multi-SNP+1-Block 0.01 0.01 0.01 0.01
multi-SNP+SSLD 0 0.02 0.01 0.02
multi-SNP+CI 0.02 0.04 0.03 0.03
single-SNP 0.02 0.02 0.02 0.02
Table 6.b. haplotype+CI and haplotype+SSLD using second situation.
Control50 Case50 Control100 Case100 Tag-Method Asso-Method
haplotype+CI multi-SNP+1-Block 0 0 0 0
multi-SNP+SSLD 0.02 0 0.01 0.02
multi-SNP+CI 0.02 0.02 0.03 0.03
single-SNP 0.03 0.02 0.02 0.02
haplotype+SSLD multi-SNP+1-Block 0.03 0.01 0.01 0.02
multi-SNP+SSLD 0 0 0.01 0.01
multi-SNP+CI 0.01 0.01 0.04 0.03
haplotype+CI multi-SNP+1-Block 0.02 0.01 0.02 0.01
Tagger
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 6.a. power of tagger – association methods in casual region
line1-line4 using association sample = 500, line5-line8 using association sample = 500, line1 and line5 using sample = control 50, line2 and line6 using sample = case 50, line3 and line7 using sample = control 100, line4 and line8 using sample = case 100
Tagger
0 0.2 0.4 0.6 0.8 1
Multi-test-1-Block Multi-test-SSLD Multi-test-GAB Single-test
power
line1 line2 line3 line4 line5 line6 line7 line8
Fig 6.b. power of tagger – association methods in null region
haplotype+CI
0 0.2 0.4 0.6 0.8 1 1.2
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 7.a. power of haplotype+CI – association methods in casual region in first situation
haplotype+CI
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 7.b. power of haplotype+CI – association methods in null region in first situation
haplotype+SSLD
0 0.2 0.4 0.6 0.8 1 1.2
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 8.a. power of haplotype+SSLD – association methods in casual region in first situation
haplotype+SSLD
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 8.b. power of haplotype+SSLD – association methods in null region in first stuation
haplotype+1-Block
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 9.a. power of haplotype+1-Block – association methods in casual region in first situation
haplotype+1-Block
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 9.b. power of haplotype+1-Block – association methods in null region in first situation
haplotype+CI
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 10.a. power of haplotype+CI – association methods in casual region in second situation
haplotype+CI
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 10.b. power of haplotype+CI – association methods in null region in second situation
haplotype+SSLD
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 11.a. power of haplotype+SSLD – association methods in casual region in second situation
haplotype+SSLD
0 0.2 0.4 0.6 0.8 1
Multi-SNP+1-Block Multi-SNP+SSLD Multi-SNP+CI Single-SNP
power
line1 line2 line3 line4 line5 line6 line7 line8
Figure 11.b. power of haplotype+SSLD – association methods in null region in second situation
References
1. Wang DG, Fan JB, Siao CJ, Berno A, Young P, Sapolsky R, Ghandour G, et al Large Scale identification, mapping and genotyping of single-nucleotide
polymorphisms in the human genome. 1998; Science 280:1077–1082.
2. Zhang K, Calabrese P, Nordborg M, Sun F. Haplotype Block Structure and Its Applications to Association Studies: Power and Study Designs. Am J Hum Genet.
2002; Nov 18;71(6).
3. Balding DJ, A tutorial on statistical methods for population association studies.
2006;Nat Rev Genet. 7:781-791
4. De Bakker PI, Yelensky R, Pe'er I, Gabriel SB, Daly MJ, Altshuler D. Efficiency and power in genetic association studies. 2005; Nat Genet 37:1217-23.
5. Stram DO. Tag SNP selection for association studies. 2004; Genetic Epidemiology 27: 365–374.
6. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Higgins J,DeFelice M, Lochner A, Faggart M, Liu-Cordero SN, Rotimi C, Adeyemo A, Cooper R,Ward R, Lander ES, Daly MJ, Altshuler D. The structure of haplotype blocks in the human genome. 2002; Science 296:2225–2229.
7. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps.2005; Bioinformatics 21: 263–265.
8. Robbins, R.B. Some applications of mathematics to breeding problems III.
1918; Genetics 3: 375-389
9. Lewontin, R.C. and K. Kojima . The evolutionary dynamics of complex polymorphisms. 1960; Evolution 14: 458-472
10. Lewontin, R.C. The interaction of selection and linkage. I. General considerations: heterotic models.1964; Genetics, 49, 49-67.
11. Wang, N., Akey, J.M., Zhang, K., Chakraborty, K., and Jin, L. Distribution of recombination crossovers and the origin of haplotype blocks: The interplay of population history.2002.
12. Patil, N., Berno, A.J., Hinds, D.A., Barrett, W.A., Doshi, J.M., Hacker, C.R., Kautzer,C.R., Lee, D.H., Marjoribanks, C., McDonough, D.P., et al. Blocks of limited haplotype diversity revealed by high-resolution scanning of human chromosome 21. 2001; Science 294: 1719–1723.
13. Zhang, K., Deng, M., Chen, T., Waterman, M.S., and Sun, F. A dynamic programming algorithm for haplotype block partitioning. Proc. Natl. Acad.
2002b; Science. 99: 7335–7339.
15. Clayton D.2005. SNPHAP: a program for estimating frequencies of large haplotypes of SNPs (Version1.3)
[http://www.gene.cimr.cam.ac.uk/clayton/software/]
14. Davs J. Balding. A tutorial on statistical methods for population association studies. 2006; Genetic
16. P.I.W. de Bakker, R. Yelensky, I. Pe'er, S.B. Gabriel, M.J. Daly, D. Altshuler Efficiency and power in genetic association studies. 2005; Nature Genetics. 37:
1217-1223
17. Chi PB, Duggal P, Kao WHL, Mathias RA, GrantAV, Stockton ML, Garcia JGN, Ingersoll RG, Scott AF, Beaty TH, Barnes KC, Fallin MD. Comparison of SNP tagging methods using empirical data: association study of 713 SNPs on
chromosome 12q14.3-12q24.21 for asthma and total serum IgE in an African Caribbean population. 2006; Genetic Epidemiology; 30:609 - 619
18. Stram DO, Haiman CA, Hirschhorn JN, Altshuler D, Kolonel LN, Henderson BE,Pike MC. Choosing haplotype-tagging SNPs based on unphased genotype
multiethnic cohort study. 2002; Hum Hered. 55:27-36
19. Julian Forton, Dominic Kwiatkowski, Kirk Rockett, Gaia Luoni, Martin Kimber, and Jeremy Hull.Accuracy of Haplotype Reconstruction from Haplotype-Tagging Single-Nucleotide Polymorphisms. 2005.
20. Kelly M Bukett, Mercedeh Ghadessi, Brad McNeney, Jinko Graham and Denise Daley. 2004. A comparison of five methods for selecting tagging
single-nucleotide polymorphisms.2005; BMC Genetics, 6:S71
21. Kui Zhang, Zhaohui S. Qin, Jun S. Liu, Ting Chen, Michael S. Waterman and Fengzhu Sun.2004.Haplotype Block Partitioning and Tag SNP Selection Using Genotype Data and Their Applications to Association Studies. 2004; Genome Res 14: 908-916.