Accounting for cellular heterogeneity is critical in epigenome-wide association studies.
Jaffe AE, Irizarry RA. · 2014
Read Accounting for cellular heterogeneity is critical in epigenome-wide association studies. by Jaffe AE, Irizarry RA. (2014) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4053810.txt.
Introduction
Andrew E. Jaffe and Rafael A. Irizarry warn that blood cell mix can fake epigenetic findings, in Genome Biology in 2014. Studies of DNA methylation in whole blood are booming. Blood, though, is a blend of cell types, and each type carries its own methylation pattern.
Using a method that estimates cell proportions from the methylation data itself, the pair re-examined five published studies of aging. Cell composition itself changed with age. Much of what had been read as age-linked methylation shift was really that blend shifting. At single sites, the two effects tangled badly. The lesson reaches beyond aging: any study on mixed tissue can mistake a change in the recipe for a change in the cells. The authors supply software for the standard array platform so teams can estimate and correct the mix.
Catalog record
| CREATOR | Jaffe AE, Irizarry RA. |
|---|---|
| YEAR | 2014 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Accounting for cellular heterogeneity is critical in epigenome-wide association studies By Andrew E Jaffe, Rafael A Irizarry Published in: Genome Biology DOI: 10.1186/gb-2014-15-2-r31 ABSTRACT Background Epigenome-wide association studies of human disease and other quantitative traits are becoming increasingly common. A series of papers reporting age-related changes in DNA methylation profiles in peripheral blood have already been published. However, blood is a heterogeneous collection of different cell types, each with a very different DNA methylation profile. Results Using a statistical method that permits estimating the relative proportion of cell types from DNA methylation profiles, we examine data from five previously published studies, and find strong evidence of cell composition change across age in blood. We also demonstrate that, in these studies, cellular composition explains much of the observed variability in DNA methylation. Furthermore, we find high levels of confounding between age-related variability and cellular composition at the CpG level. Conclusions Our findings underscore the importance of considering cell composition variability in epigenetic studies based on whole blood and other heterogeneous tissue sources. We also provide software for estimating and exploring this composition confounding for the Illumina 450k microarray. BACKGROUND Epigenome-wide association studies (EWAS) of human disease are becoming increasingly common. DNA methylation (DNAm) is of particular interest because it is dynamic across the lifetime, affected by environmental insults, and previously implicated in developmental disorders and cancer [1]. In these studies, DNAm levels are measured genome-wide at thousands to millions of sites in hundreds of individuals to identify loci where these levels are associated with quantitative traits or disease [1,2]. Because existing cohort studies that extensively characterize participants often store blood samples, the most widely available tissue for subsequent/retrospective EWAS is whole blood. Furthermore, many studies measure genome-wide DNAm in blood as obtaining disease-relevant tissues is often invasive and/or impossible. With many of these studies completed, few disease-associated loci have been reported outside of cancer [3], type 1 diabetes [4], and rheumatoid arthritis [5]. Instead a series of papers reporting age-related changes of DNAm profiles have been published [6-14]. Age-related changes in DNAm have been previously reported and functionally described by Chu et al. [15]. In this carefully designed study, fluorescence-activated cell sorting (FACS) was used to separate peripheral blood into pure cellular populations. DNAm was measured in four genomic regions, selected using biological insight, and modest age-related changes were found in CD4+ and CD8+ T cells. In contrast, the above-mentioned EWAS measured DNAm for all CpGs selected by the array manufacturers and used whole blood as a source tissue. Whole blood is a heterogeneous collection of different cell types, each with a very different DNA methylation profile [16,17]. Observed whole blood DNAm profiles are therefore mixtures of the cell type profiles. In a seminal paper, Houseman et al. [16] describe a statistical method that can accurately estimate relative proportions of cell type components in whole blood. Using practically the same statistical approach, Guintivano et al. [18] describe a method for estimating neuron and non-neuron components in brain samples. However, currently there are no published statistical solutions to parsing age effects by cell type from observed whole blood DNAm measurements. We examined data from five publicly available studies (Additional file 1) and found strong evidence of cell composition changes across age. Furthermore, we find high levels of confounding between age-related variability and cell composition. We report findings that underscore the importance of accounting for cell composition variability in epigenetic studies based on whole blood and other heterogeneous tissue sources. RESULTS AND DISCUSSION DNAm profiles show large between cell type differences We downloaded Illumina HumanMethylation450 BeadChip (Illumina 450k) data from flow-sorted neutrophils (granulocytes), lymphocytes (CD8+ and CD4+ T cells, CD56+ natural killer cells and CD19+ B cells) and CD14+ monocytes from six adult male samples (mean age 38 ± 13.6 years) as previously described [17] and confirmed that sorted blood cell types have unique DNAm profiles (Figure S1 in Additional file 2). In fact 63.5% of the CpGs on the Illumina 450k array showed differences with P < 0.05 across these cell types (Figure S1C in Additional file 2). We used these data to adapt the statistical method developed by Houseman et al.[16] for the Illumina HumanMethylation27 BeadChip (Illumina 27k) array to estimate cell composition from DNAm profiles obtained with its successor, the Illumina 450k. We select a subset of 600 cell-type-specific CpGs (Figure 1) and then use these to estimate proportions in whole blood samples (see Materials and methods). We provide a table with statistical summaries of cell-type variability for all CpGs on the Illumina 450k array (Additional file 3). [Figure: Illustration of how blood composition drives observed age differences. (A) Heatmap of the cell sorted data shows very clear and consistent DNAm profiles for each cell type. We show 600 probes selected for estimating composition proportions used to demonstrate differences here. (B) To simplify the illustration we selected a section of (A) displaying only the two most abundant cell types: CD4+ T cells and granulocytes. (C) Heatmap of a randomly selected sample of 30 whole blood samples (from the data in Additional file 1) across three age groups (10 per group): between 1 and 5 years of age, between 30 and 40, greater than 60 years. The same probes as in (B) are used. When the samples are ordered by their estimated granulocyte proportion, the samples roughly cluster by age and a similar pattern to (B) is observed. The estimated cell count proportions for each of the samples are shown below. Note the strong confounding between age and cell composition. (D) For the two samples highlighted with an arrow in (C), we show how a weighted average of the cell type profiles can reconstruct the observed DNAm profiles. The numbers shown are the estimated proportions. Note how different weights (cell counts) for old and young result in very different observed DNAm patterns. Note that the differences in CD4+ T cells and granulocytes drive much of the differences in DNAm. NK, CD56+ natural killer cells; CD8T, CD8+ T cells; CD4T, CD4+ T cells, Gran, granulocytes; Bcell, CD19+ B cells; Mono, CD14+ monocytes; DNAm, proportion of DNA methylation at individual CpGs (Illumina 'beta' values, bound between 0 and 1); Prop, cell count proportion, between 0 and 1 for each component, such that they sum to 1.] In sorted samples, cell type explains a larger percentage of variability than age Given these results, for the purposes of our analysis, we assumed that, for the selected 600 CpGs, the cell type-specific DNAm profiles are the same for all ages. Although we know this assumption does not hold true for all CpGs [15], the results of this section suggest that it is reasonable for most CpGs, and our 600 CpG profile in particular. To demonstrate this, we interrogated two publicly available datasets - the Reinius et al. [17] Illumina 450k data on 6 men (sample ages were obtained from the authors) and Illumina 27k data from sorted CD4+ T cells and monocytes [6] on 24 and 26 subjects, respectively (see Materials and methods). First, we removed CpG probes that showed age associations (at P < 0.05) in the Reinius et al. [17] dataset when picking cell-type-discriminating probes for the cell composition estimation. Additionally, in Rakyan et