Cell type deconvolution of methylated cell-free DNA at the resolution of individual reads

Abstract

Cell-free DNA (cfDNA) are DNA fragments originating from dying cells that are detectable in bodily fluids, such as the plasma. Accelerated cell death, for example caused by disease, induces an elevated concentration of cfDNA. As a result, determining the cell type origins of cfDNA molecules can provide information about an individual's health. In this work, we aim to increase the sensitivity of methylation-based cell type deconvolution by adapting an existing method, CelFiE, which uses the methylation beta values of individual CpG sites to estimate cell type proportions. Our new method, CelFEER, instead differentiates cell types by the average methylation values within individual reads. We additionally improved the originally reported performance of CelFiE by using a new approach for finding marker regions that are differentially methylated between cell types. This approach compares the methylation values over 500 bp regions instead of at single CpG sites and solely takes hypomethylated regions into account. We show that CelFEER estimates cell type proportions with a higher correlation (r2 = 0.94 ± 0.04) than CelFiE (r2 = 0.86 ± 0.09) on simulated mixtures of cell types. Moreover, we found that it can find a significant difference between the skeletal muscle cfDNA fraction in four ALS patients and four healthy controls.

Multi-cell type deconvolution using a probabilistic model for single-molecule DNA methylation haplotypes

“…Methylation arrays provide information at a single base level, while methylation sequencing can provide within-read methylation patterns or methylation haplotypes [22]. A single read can only have one cell type of origin, and a recent whole-genome atlas of purified cell types [1,29] allows for general purpose multi cell-type deconvolution. Deconvolution methods developed for methylation array data [7,19] do not take methylation haplotype information into account.…”

“…Deconvolution methods developed for methylation array data [7,19] do not take methylation haplotype information into account. A new read-aware method, "UXM", was published along with the methylation atlas [1,29]. UXM relies on percent methylation within each read, and thus forgoes base-level information.…”

“…Next, the methylation status of each CpG in each read was determined by binomial sampling of its cell type of origin, at the appropriate position, . Unlike the simulations in Caggiano et al [21] and Keukeleire et al [29], the read depth was a set parameter and not sampled. The total number of individual methylation cells ( ) was held constant regardless of read lengths by adjusting the number of reads.…”

Circulating, cell-free methylated DNA indicates cellular sources of allograft injury after liver transplant

“…Also, cell-free methylated DNA can reflect intervention-related changes over time through analyses of serially collected blood samples (38). Fragment-level deconvolution of methylation sequencing data allows for increased sensitivity and specificity of signal localization using CpG pattern analysis of individual cfDNA molecules (35, 38, 51–53). Likewise, hybridization capture to CpG-rich DNA segments maximizes sequencing depth while still maintaining comprehensive coverage (38).…”