A high-throughput droplet digital PCR system aiming eight DNA methylation targets for age prediction

Yuxiang Zhou,Yanyun Wang,Mengyuan Song,Lanrui Jiang,Chaoran Sun,Shuangshuang Wang,Hewen Yao,Zefei Wang,Xindi Wang,Chunhui Liu,Haibo Luo,Feng Song

DOI: https://doi.org/10.1016/j.jpba.2023.115943

2024-03-15

Abstract:The droplet digital Polymerase Chain Reaction (ddPCR) has garnered recognition for its distinctive attribute of absolute quantification. And it has found practical utility in age prediction through DNA methylation profiles. However, a prevalent limitation in current ddPCR methodologies is the restricted capacity to detect only two targets concurrently in most instruments, leading to high costs, sample wastage, and labor-intensive procedures. To address the limitations, a novel high-throughput ddPCR system allowing for the simultaneous detection of eight targets was developed. Through the implementation of a new 8-plex ddPCR assay, coupled with comprehensive linear regression analyses involving primers and probes ratios, diverse inputs of single CpG sites with distinct primers and probes, and varying plex assay configurations, stable DNA methylation values for four CpGs and stable measurement precisions for distinct multiplex systems were consistently observed. These findings pave the way for advancing the field of chemistry science by enabling more efficient and cost-effective methods. Furthermore, the comparative validation of ddPCR and SNaPshot demonstrated a remarkable concordance in results, and the system also displayed well in the field of various aspects, including species specificity, DNA input, and aged samples. In this study, the recommended input of bisulfite-converted DNA was determined to be 10-50 ng due to the double-positive droplets. Notably, the Pearson correlation coefficient squared values of four CpGs were 0.4878 (ASPA), 0.4832 (IGSF1), 0.6881 (COL1A1), and 0.6475 (MEIS1-AS3). And the testing set exhibited a mean absolute error of 4.5923 years, indicating the robustness and accuracy of the age-predictive model.