Professor Philippe Collas, Institute of Basic Medical, University of Oslo, Oslo, Norway.
Understanding how three-dimensional (3D) genome organization influences the regulation of gene expression is a challenge in genome biology. Chromatin conformation is regulated by intra- and inter-chromosomal interactions and associations of chromatin with nuclear lamins. Lamin-genome interactions are in turn under the influence of epigenetic states. We will discuss how chromatin domains of histone H2B GlcNAcylated on Ser112 ‘pre-pattern’ the formation of lamin-associated domains (LADs) during differentiation of adipocyte progenitors into adipocytes [1,2]. As GlcNAcylation is a nutrient sensing modification, our results couple lamin-chromatin interactions, and hence spatial genome conformation, to cellular metabolic state.
Chromosome-chromosome interactions are dynamic and under influence of stochastic and regulated processes. These premises challenge the deconvolution of individual spatial association patterns from data aggregated from millions of cells. We will discuss a computational strategy we are implementing to infer 3D chromatin structure based on integrated modeling of LADs mapped by ChIP-seq and 3D chromosomal interactions identified by HiC. Resulting 3D models are used to elucidate the interplay between chromosomal interactions at the nuclear periphery and in the nuclear interior. We notably identify topologically-associated and lamin-associated chromatin domains associated with the nuclear periphery or placed in the nuclear interior, and hyper-dynamic domains with variable spatial positioning. It is becoming increasingly clear that Integration of multiple datasets with increasingly performant computation techniques  will be essential to underpin the complexity of spatial genome conformation in dynamic systems such as differentiating stem cells.
 Lund et al. 2013. Genome Res 23, 1580-9
 Rønningen, Shah et al. 2015. Genome Res; PMD 26359231
 Paulsen et al. 2015. PLoS Comput Biol 11, e1004396