Highlights• Structural and functional differences between the maternal and paternal genomes, including similar allelic bias in related subsets of genes.• Coupling between the dynamics of gene expression and genome architecture for specific alleles illuminates an allele specific 4D Nucleome.• A novel allele specific phasing algorithm for genome architecture and a quantitative framework for integration of gene expression and genome architecture.
AbstractMillions of genetic variants exist between the paternal and maternal genomes in human cells (1, 2), which result in unequal allelic contributions to gene transcription (3, 4). However, it remains poorly understood how allelic bias affects the interplay between transcription and the 3D organization of the genome. We sought to understand how transcription and genome architecture differ between the maternal and paternal genomes across the cell cycle. We collected and analyzed haplotype-resolved genome-wide data from B-Lymphocytes (NA12878) in G1, S, and G2/M, using RNA sequencing (RNAseq), bromouridine sequencing (Bru-seq), and genome wide chromosome conformation capture (Hi-C). In the past, separation of allele specific data was done only through heterozygous single nucleotide variations (SNVs), insertions, and deletions (InDels), as these unique variations allowed DNA sequencing reads to be mapped back to their parental origins. In this paper, we introduce a novel method of phasing Hi-C data using reads assigned through SNVs/InDels to predict the parental origin of nearby reads of unknown origin. This method allows for more structural data to be systematically assigned to a parental origin, and therefore reduces the sparsity of the allele specific Hi-C contact matrices. By integrating allele specific RNA-seq, Bru-seq, and Hi-C data through three phases of the cell cycle, along with publicly available protein binding data, we provide a more comprehensive understanding of architectural and transcriptional differences between the two genomes. These analyses reveal specific patterns in allelic bias, including similar bias characteristics in some groups of related genes. The integration of these data enabled construction of an allele specific 4D Nucleome.