According to the authors, current methylation-based deconvolution approaches are sensitive to sequencing depth, the quality of reference data, and variability between sequencing platforms. ReadsRatio was designed to sidestep these limitations by relying on read-level information within marker regions rather than on calibrated methylation intensity values. In evaluations combining computer simulations with clinical cohorts, the method detected specific tissue involvement in patients with lupus, in kidney transplant recipients, and in individuals with COVID-19, where it was compared against conventional deconvolution approaches.

A key practical advantage reported by the team is robustness across sequencing platforms and at reduced sequencing depths. Because the method maintains its accuracy when less sequencing data is available, preliminary estimations by the researchers suggest it could enable substantial cost savings compared with approaches that require deeper sequencing. The authors position ReadsRatio as a versatile, cost-efficient framework for cell-free DNA deconvolution with broad applicability in both research and clinical settings.

The findings are presented as an open-access methodology article, and the version published is a peer-reviewed, accepted manuscript carrying a permanent DOI, subject to further editorial edits before the final version of record. The study spans simulation work and clinical cohorts covering three distinct disease and transplantation contexts, providing early evidence that the read-ratio strategy can identify tissue-specific signals where established approaches may falter. As with any newly reported method, the preliminary nature of the cost estimates and the specific cohorts examined define the current scope of the evidence, and broader validation across additional platforms and clinical settings remains a task for future work.

Subject of Research: A computational method for cell type deconvolution of cell-free DNA using methylation read ratios

Article Title: ReadsRatio: a facile method for cell type deconvolution with cross-platform applications

Article References: Jing, Q., Li, Q., Chan, S. C. W., Chen, W., Choi, N. F. Y., Tang, I. Y. K., Lau, C.-S., Ling, G. S., Li, P. H., & Wong, J. W. H. (2026). ReadsRatio: a facile method for cell type deconvolution with cross-platform applications. Genome Biology. https://doi.org/10.1186/s13059-026-04310-9

Image Credits: AI Generated

DOI: 10.1186/s13059-026-04310-9

Keywords: cell-free DNA, deconvolution, DNA methylation, ReadsRatio, sequencing, Genome Biology, lupus, kidney transplantation, COVID-19, liquid biopsy, computational biology, biomarkers