We propose a novel and resource-efficient approach that enables joint imputation of multiple epigenetic data tracks through denoising autoencoders neural networks that allow prediction of missing data by learning their relations with observed data.
A brief description is included in the Write-up folder.