Supplement Fig. S1. QQ-Plot for Hispanic Genome Wide Data

Supplement Fig. S1. QQ-plot for Hispanic genome wide data

Quantile-quantile (Q-Q) plot comparing the distributing of the observed versus the expected –log10 P values (log-additive model, adjusted for age and sex) of the 634,037 SNPs. The black line indicatesdistribution expected under the null hypothesis of no association and the red line indicates the inflation of the test statistics (λ=1.02)

Supplement Fig. S2: Association and recombination plots for IKZF1 (7p12.2 loci) associated with childhood ALL risk in the Hispanic population, CCLS, 1995-2008

Genotyped SNPs are represented as diamonds and a larger diamond indicated the top-hit association in each region. The strength of linkage disequilibrium between each SNPs and the top hit is indicated by the color of the symbol. Recombination rates, plotted in light blue, are based on HapMap3MEX samples, and genomic coordinates are based on National Center for Biotechnology Information (NCBI) Build 36 of the human genome (released in March 2006).

Supplement Fig. S3: Association and recombination plots for ARID5B (10q21.2 loci) associated with childhood ALL risk in the Hispanic population, CCLS, 1995-2008

Supplement Fig. S4: Association and recombination plots for CEBPE (14q11.2 loci) associated with childhood ALL risk in the Hispanic population, CCLS, 1995-2008

Supplement Table S1: P-values for gene by environment interaction of six candidate SNPs† and daycare attendance by age of 6 months with the risk of ALL in Hispanic Children, CCLS, 1995-2008

Gene / SNP / Censored at Age 6 months / Cases N‡ / Controls N‡ / OR (95%CI)a / Pinteraction
CEBPE / rs4982731 / No / 81/198 / 162/237 / 1.64 (1.17-2.29) / 0.07
Yes / 4/30 / 16/22 / 6.42 (1.56-26.48)
CEBPE / rs17794251 / No / 85/194 / 167/231 / 1.65 (1.17-2.31) / 0.99
Yes / 9/25 / 14/24 / 2.45 (0.64-9.42)
ARID5B / rs4506592 / No / 101/175 / 66/333 / 0.35 (0.24-0.51) / 0.70
Yes / 13/21 / 8/30 / 0.35 (0.09-1.31)
IKZF1 / rs4132601 / No / 128/151 / 226/173 / 1.55 (1.14-2.13) / 0.71
Yes / 15/19 / 22/16 / 2.36 (0.69-8.03)
IKZF1 / rs6964823 / No / 165/104 / 185/214 / 0.59 (0.43-0.82) / 0.61
Yes / 14/20 / 15/23 / 0.59 (0.17-2.09)
IKZF1 / rs4917017 / No / 93/186 / 101/298 / 0.69 (0.49-0.97) / 0.99
Yes / 6/28 / 4/34 / 0.93 (0.18-4.73)

Odds ratios (OR) and 95% confidence intervals (CI) calculated using logistic regression model adjusting for child’s age, gender, annual household income and 5 genetic principal components (PCs).

† Six candidate SNPs were selected from all significant SNPs using a conditional haplotype analysis.

‡ Dominant models used to provide counts of cases with wild genotype versus having any copy of minor alleles and counts of controls with wild genotype versus having any copy of minor alleles by daycare attendance by 6 months (yes/no).

Supplement Table S2. P-values for gene by environment interaction of six candidate SNPs†and daycare attendance by age one with the risk of ALL in Hispanic Children, CCLS, 1995-2008

Gene / SNP / Censored at Age one / Cases N‡ / Controls N‡ / OR (95%CI)a / Pinteraction
CEBPE / rs4982731 / No / 78/192 / 153/223 / 1.65 (1.17-2.33) / 0.15
Yes / 7/36 / 25/36 / 3.08 (1.07-8.85)
rs17794251 / No / 81/89 / 155/220 / 1.62 (1.15-2.29) / 0.78
Yes / 13/30 / 26/35 / 2.12 (0.78-5.76)
ARID5B / rs4506592 / No / 98/172 / 65/311 / 0.37 (0.25-0.53) / 0.70
Yes / 12/27 / 9/52 / 0.26 (0.08-0.82)
IKZF1 / rs4132601 / No / 123/147 / 216/160 / 1.62 (1.17-2.23) / 0.74
Yes / 20/23 / 32/29 / 1.32 (0.52-3.33)
rs6964823 / No / 162/108 / 175/200 / 0.58 (0.42-0.81) / 0.35
Yes / 17/26 / 24/37 / 1.13 (0.42-3.02)
rs4917017 / No / 91/179 / 94/282 / 0.67 (0.47-0.96) / 0.65
Yes / 8/35 / 11/50 / 1.290.42-4.02)

Odds ratios (OR) and 95% confidence intervals (CI) calculated using logistic regression model adjusting for child’s age, gender, annual household income and 5 genetic principal components (PCs).

† Six candidate SNPs were selected from all significant SNPs using a conditional haplotype analysis.