Gene expression modelling across multiple cell-lines with MapReduce

With the wealth of high-throughput sequencing data generated by recent large-scale consortia, predictive gene expression modelling has become an important tool for integrative analysis of transcriptomic and epigenetic data. However, sequencing data-sets are characteristically large, and previously modelling frameworks are typically inefficient and unable to leverage multi-core or distributed processing architectures. In this study, we detail an efficient and parallelised MapReduce implementation of gene expression modelling. We leverage the computational efficiency of this framework to provide an integrative analysis of over fifty histone modification data-sets across a variety of cancerous and non-cancerous cell-lines. Our results demonstrate that the genome-wide relationships between histone modifications and mRNA transcription are lineage, tissue and karyotype-invariant, and that models trained on matched epigenetic/transcriptomic data from non-cancerous cell-lines are able to predict cancerous expression with equivalent genome-wide fidelity.

TREEOME: A framework for epigenetic and transcriptomic data integration to explore regulatory interactions controlling transcription

Motivation: Predictive modelling of gene expression is a powerful framework for the in silico exploration of transcriptional regulatory interactions through the integration of high-throughput -omics data. A major limitation of previous approaches is their inability to handle conditional and synergistic interactions that emerge when collectively analysing genes subject to different regulatory mechanisms. This limitation reduces overall predictive power and thus the reliability of downstream biological inference. Results: We introduce an analytical modelling framework (TREEOME: tree of models of expression) that integrates epigenetic and transcriptomic data by separating genes into putative regulatory classes. Current predictive modelling approaches have found both DNA methylation and histone modification epigenetic data to provide little or no improvement in accuracy of prediction of transcript abundance despite, for example, distinct anti-correlation between mRNA levels and promoter-localised DNA methylation. To improve on this, in TREEOME we evaluate four possible methods of formulating gene-level DNA methylation metrics, which provide a foundation for identifying gene-level methylation events and subsequent differential analysis, whereas most previous techniques operate at the level of individual CpG dinucleotides. We demonstrate TREEOME by integrating gene-level DNA methylation (bisulfite-seq) and histone modification (ChIP-seq) data to accurately predict genome-wide mRNA transcript abundance (RNA-seq) for H1-hESC and GM12878 cell lines. Availability: TREEOME is implemented using open-source software and made available as a pre-configured bootable reference environment. All scripts and data presented in this study are available online at http://sourceforge.net/projects/budden2015treeome/.