Sur ce site

Sur le Web du CNRS

Accueil du site > Animations Scientifiques > Séminaires 2010 > Aurélien Rappailles — Genome wide study of Human DNA replication

Aurélien Rappailles — Genome wide study of Human DNA replication

Speaker :

Aurélien Rappailles, Eukaryotic chromosome replication group, Ecole Normale Supérieure, Paris

When :

Wednesday 23 June at 11am

Where :

C023 (RDC LR6 côté Centre Blaise Pascal)

Title :

Genome wide study of Human DNA replication.

Abstract :

DNA replication of the human genome starts at around 30000 origins whose position and mechanisms that coordinate their activation during S phase are unclear. It was proposed that during evolution, replication induces nucleotides composition asymmetries in the leading and lagging strand sequences. Using bioinformatic tools, groups of C.Thermes and A.Arneodo analyzed this skew in the human genome. They found characteristic patterns covering 30% of the genome and consisting of successions of N-shaped structures. These structures are called “N-domains” and it was proposed that N-domains borders are replication origins.

To test bioinformatics predictions we controlled whether N-domains borders are replicated earlier than surrounding regions. Using massive sequencing, we have determined a high-resolution replication timing profile of the entire human genome. Our data show that N-domain borders replicate earlier than surrounding regions, thus borders contain replication origins. Automatic detection of early replication peaks also led us to identify several thousands of segments whose temporal profile of replication has the same "U" shape as N-domains. The slope of this profile suggests that replication progresses away from U-domain borders faster than expected for a single fork. Moreover DNA combing revealed several initiation events within U-domains. Thus we propose a “Domino activation model” where U-domains borders are early master origins conditioning the firing of secondary origins inside U-domains by the approach of a center-oriented fork, explaining why apparent speeds are faster than observed for single forks.

Dans la même rubrique :