Handling Realistic Assumptions in Hypothesis Testing of 3D Co-Localization of Genomic Elements

• Activity Number: 82
• Type: Contributed
• Date/Time: Sunday, August 4, 2013 : 4:00 PM to 5:50 PM
• Sponsor: Section on Statistical Computing
• Author(s): Tonje Lien*+ and Jonas Paulsen and Geir Kjetil Sandve and Lars Holden and Ørnulf Borgan and Ingrid Glad and Eivind Hovig
• Companies: and Oslo University Hospital, Section for Medical Informatics, The Norwegian Radium Hospital and University of Oslo, Department of Informatics and Norwegian Computing Center and University of Oslo, Department of Mathematics and University of Oslo, Department of Mathematics and Oslo University Hospital, Institute for Cancer Research, Department of Tumor Biology
• Keywords: Hi-C ; chromatin structure ; hypothesis testing ; Genetics ; permutation test

Abstract

The study of chromatin 3D structure has recently gained much focus, due to novel techniques for detecting genome wide chromatin contacts utilizing next-generation sequencing. A deeper understanding of the architecture of the DNA inside the nucleus is crucial for gaining insight into fundamental processes such as transcriptional regulation. Chromatin conformation capture-based methods such as Hi-C and ChIA-PET are now paving the way for routine genome-wide studies of chromatin 3D structure in a range of organisms and tissues. However, appropriate methods for analyzing such data are lacking. We propose a hypothesis test of 3D co-localization of genomic elements that handles intra- or interchromosomal interactions both separately and jointly and that adjusts for biases caused by structural dependencies in the data. We show that maintaining structural properties during resampling is essential in order to obtain valid estimation of p-values. Applying the method on a set of mutated regions in leukemia cells, we found strong evidence of intrachromosomal co-localization of these elements, supporting the role of chromatin 3D structure in shaping the landscape of somatic mutations in cancer.