Authors: Mian Umair Ahsan, MS (1); Kai Wang, PhD (1, 2). Affiliation: 1) Raymond G. Perelman Center for Cellular and Molecular Therapeutics, Children's Hospital of Philadelphia. 2) Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania.
Poster # 84
DNA methylation is an important biological marker that affects a range of physiological and cellular processes such as differentiation and pluripotency. Aberrant methylation patterns are linked to several diseases such as cancer and imprinted disorders, and analysis of differentially methylated regions can provide crucial insights into how certain genomic regions or biological pathways are affected by different biological or medical conditions.In recent years, Oxford Nanopore Technologies (ONT) sequencing has emerged as a reliable and direct technology in DNA methylation detection through the analysis of ionic current signal. Moreover, long ONT reads allow accurate haplotyping of genomes into parental haplotypes. In this study, we propose a haplotype-aware framework for detecting differentially methylated regions between two genomes or haplotypes of the same genomes. The proposed framework uses NanoCaller for variant calling and read haplotyping, then uses DeepMod2 to predict methylation levels for CpG sites on each haplotype. Lastly, DSS is used to detect differentially methylated regions between all haplotypes of the samples. We apply this framework to Nanopore reads of HG001 and Ashkenazim trio (HG002, HG003 and HG004) to detect differentially methylated and imprinted regions. We also use these datasets to simulate genomes with imprinting disorders such as Silver-Russell and Beckwith-Wiedemann syndromes and evaluate the ability of the proposed framework to detect imprinted control regions implicated in these diseases.
Comments