Open Access Highly Accessed Open Badges Research

Quantitative reconstruction of leukocyte subsets using DNA methylation

William P Accomando1, John K Wiencke2, E Andres Houseman3, Heather H Nelson4 and Karl T Kelsey15*

Author Affiliations

1 Department of Pathology and Laboratory Medicine, Brown University, Providence, RI 02912, USA

2 Department of Neurological Surgery, University of California San Francisco, San Francisco, CA 94158, USA

3 Department of Public Health, Oregon State University, Corvallis, OR 97331, USA

4 Department of Epidemiology, University of Minnesota, Minneapolis, MN 55455, USA

5 Department of Epidemiology, Brown University, Providence, RI 02912, USA

For all author emails, please log on.

Genome Biology 2014, 15:R50  doi:10.1186/gb-2014-15-3-r50

Published: 5 March 2014



Cell lineage-specific DNA methylation patterns distinguish normal human leukocyte subsets and can be used to detect and quantify these subsets in peripheral blood. We have developed an approach that uses DNA methylation to simultaneously quantify multiple leukocyte subsets, enabling investigation of immune modulations in virtually any blood sample including archived samples previously precluded from such analysis. Here we assess the performance characteristics and validity of this approach.


Using Illumina Infinium HumanMethylation27 and VeraCode GoldenGate Methylation Assay microarrays, we measure DNA methylation in leukocyte subsets purified from human whole blood and identify cell lineage-specific DNA methylation signatures that distinguish human T cells, B cells, NK cells, monocytes, eosinophils, basophils and neutrophils. We employ a bioinformatics-based approach to quantify these cell types in complex mixtures, including whole blood, using DNA methylation at as few as 20 CpG loci. A reconstruction experiment confirms that the approach could accurately measure the composition of mixtures of human blood leukocyte subsets. Applying the DNA methylation-based approach to quantify the cellular components of human whole blood, we verify its accuracy by direct comparison to gold standard immune quantification methods that utilize physical, optical and proteomic characteristics of the cells. We also demonstrate that the approach is not affected by storage of blood samples, even under conditions prohibiting the use of gold standard methods.


Cell mixture distributions within peripheral blood can be assessed accurately and reliably using DNA methylation. Thus, precise immune cell differential estimates can be reconstructed using only DNA rather than whole cells.