Statistical Genomics

Marco Antonio Mendoza-Parra, Mohamed-Ashick M. Saleem, Matthias Blum, Pierre-Etienne Cholley, Hinrich Gronemeyer, Mendoza-Parra, Marco Antonio, Saleem, Mohamed-Ashick M, Blum, Matthias, Cholley, Pierre-Etienne, Gronemeyer, Hinrich

Ewy Mathé, Sean Davis

The combination of massive parallel sequencing with a variety of modern DNA/RNA enrichment technologies provides means for interrogating functional protein-genome interactions (ChIP-seq), genome-wide transcriptional activity (RNA-seq; GRO-seq), chromatin accessibility (DNase-seq, FAIRE-seq, MNase-seq), and more recently the three-dimensional organization of chromatin (Hi-C, ChIA-PET). In systems biology-based approaches several of these readouts are generally cumulated with the aim of describing living systems through a reconstitution of the genome-regulatory functions. However, an issue that is often underestimated is that conclusions drawn from such multidimensional analyses of NGS-derived datasets critically depend on the quality of the compared datasets. To address this problem, we have developed the NGS-QC Generator, a quality control system that infers quality descriptors for any kind of ChIP-sequencing and related datasets. In this chapter we provide a detailed protocol for (1) assessing quality descriptors with the NGS-QC Generator; (2) to interpret the generated reports; and (3) to explore the database of QC indicators ( www.ngs-qc.org ) for >21,000 publicly available datasets.