A systematic evaluation of normalization methods and probe replicability using infinium EPIC methylation data

BACKGROUND

The Infinium EPIC array measures the methylation status of > 850,000 CpG sites. The EPIC BeadChip uses a two-array design: Infinium Type I and Type II probes. These probe types exhibit different technical characteristics which may confound analyses. Numerous normalization and pre-processing methods have been developed to reduce probe type bias as well as other issues such as background and dye bias.

METHODS

This study evaluates the performance of various normalization methods using 16 replicated samples and three metrics: absolute beta-value difference, overlap of non-replicated CpGs between replicate pairs, and effect on beta-value distributions. Additionally, we carried out Pearson's correlation and intraclass correlation coefficient (ICC) analyses using both raw and SeSAMe 2 normalized data.

RESULTS

The method we define as SeSAMe 2, which consists of the application of the regular SeSAMe pipeline with an additional round of QC, pOOBAH masking, was found to be the best performing normalization method, while quantile-based methods were found to be the worst performing methods. Whole-array Pearson's correlations were found to be high. However, in agreement with previous studies, a substantial proportion of the probes on the EPIC array showed poor reproducibility (ICC < 0.50). The majority of poor performing probes have beta values close to either 0 or 1, and relatively low standard deviations. These results suggest that probe reliability is largely the result of limited biological variation rather than technical measurement variation. Importantly, normalizing the data with SeSAMe 2 dramatically improved ICC estimates, with the proportion of probes with ICC values > 0.50 increasing from 45.18% (raw data) to 61.35% (SeSAMe 2).

African ancestry protects against Alzheimer's disease-related neuropathology

Previous studies in dementia epidemiology have reported higher Alzheimer's disease rates in African-Americans when compared with White Americans. To determine whether genetically determined African ancestry is associated with neuropathological changes commonly associated with dementia, we analyzed a population-based brain bank in the highly admixed city of São Paulo, Brazil. African ancestry was estimated through the use of previously described ancestry-informative markers. Risk of presence of neuritic plaques, neurofibrillary tangles, small vessel disease, brain infarcts and Lewy bodies in subjects with significant African ancestry versus those without was determined. Results were adjusted for multiple environmental risk factors, demographic variables and apolipoprotein E genotype. African ancestry was inversely correlated with neuritic plaques (P=0.03). Subjects with significant African ancestry (n=112, 55.4%) showed lower prevalence of neuritic plaques in the univariate analysis (odds ratio (OR) 0.72, 95% confidence interval (CI) 0.55-0.95, P=0.01) and when adjusted for age, sex, APOE genotype and environmental risk factors (OR 0.43, 95% CI 0.21-0.89, P=0.02). There were no significant differences for the presence of other neuropathological alterations. We show for the first time, using genetically determined ancestry, that African ancestry may be highly protective of Alzheimer's disease neuropathology, functioning through either genetic variants or unknown environmental factors. Epidemiological studies correlating African-American race/ethnicity with increased Alzheimer's disease rates should not be interpreted as surrogates of genetic ancestry or considered to represent African-derived populations from the developing nations such as Brazil.

Functional polymorphisms of DEFB1 gene in type 1 diabetes Brazilian children

We analyzed three functional 5' un-translated region beta-defensin 1 (DEFB1) single nucleotide polymorphism (SNPs) in a group of 170 type 1 diabetes (T1D) patients. In order to evaluate the SNPs influence on the disease onset and the development of other autoimmune disorder, such as celiac disease (CD) and autoimmune thyroid disease (AITD), patients were stratified according to the presence of AITD, CD, and both AITD and CD. As control group, we studied 191 healthy children and adolescent not presenting a familiar historic of T1D, CD or AITD. DEFB1 SNPs were in Hardy-Weinberg equilibrium both in healthy controls and T1D patients, as well in the T1D patients stratified according to the presence of other autoimmune disorder(s). Allele, genotype, and haplotype frequencies of T1D patients globally considered were comparable to healthy controls ones. No evidence of any association of DEFB1 SNPs with the onset of AIDT, CD, and both AITD and CD on T1D patients was evidenced. Only a minor trend was found for an increased frequency of the - 20 G allele in T1D patients only presenting AITD vs. T1D patients not presenting AITD or CD, as well as an increase of those haplotypes comprising the - 20 G allele when compared with the GCA haplotype. We also evaluated the influence of functional DEFB1 SNPs on the age of T1D onset: no significant statistical conclusion was achieved. Further studies are envisaged, in order to elucidate the possible role of functional DEFB1 polymorphisms in the onset of TD1 and other autoimmune-related disorders.