Proceeding talk – Theme: Genome.
Abstract
TET proteins oxidise methylated cytosine (5mC) into oxidised methylcytosines (oxi-mCs): 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC). DNA methylation variants have multiple functions, e.g., 5mC is associated with diseases and oxi-mC species have a role in active DNA demethylation. How to analyse bisulphite based 5mC and oxi-mC data sets with convoluted read-outs, experiment-specific variation, confounding effects and arbitrary experimental designs is an open and important problem. We propose the first method to quantify all DNA methylation modifications with arbitrary covariate structures from bisulphite based sequencing data by combining a previously proposed hierarchical generative model and a general linear model component. Our method provides accurate methylation level estimates and accurate detection of differential methylation when compared to existing methods. Analysis of novel and published data gave insights into the demethylation of the Foxp3 locus during the induced T regulatory cell differentiation and allowed us to make accurate methylation level predictions.
Authors
Tarmo Äijö, Center for Computational Biology, Simons Foundation, United States
Harri Lähdesmäki, Aalto University School of Science, Finland
Anjana Rao, La Jolla Institute for Allergy & Immunology, United States
Xiaojing Yue, La Jolla Institute for Allergy & Immunology, United States