Regensburg 2019 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

BP: Fachverband Biologische Physik

BP 28: Statistical physics of biological systems II (joint session BP/DY)

BP 28.5: Vortrag

Donnerstag, 4. April 2019, 16:00–16:15, H11

Immune Repertoire Dynamics out of Steady State — •Mario Udo Gaimann1, Jonathan Desponds2, and Andreas Mayer31Ludwig-Maximilians-Universität München, Faculty of Physics, Munich, Germany — 2University of California San Diego, Department of Physics, La Jolla, CA, USA — 3Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA

Over the last ten years high-throughput sequencing of lymphocyte receptor repertoires has provided an increasingly precise view of how immune defenses are organized. A highly reproducible finding of these sequencing efforts has been that the clone sizes of lymphocytes which share the same receptor are heavy-tail distributed. Here, we present a simple neutral birth-death model kept out of steady-state by the arrival of new clones in which competition between cells for a global resource couples the birth rate to the total size of the immune repertoire. We show that this model produces transient, but long-lived power-law scaling of clone sizes through a rich-get-richer mechanism resembling preferential attachment. Our model predicts an onset of power-law scaling early in life, which should persist throughout the human lifespan in the biologically relevant parameter regime. We verify both predictions by reanalyzing previously published T cell receptor sequencing data from a human aging study. Furthermore, we demonstrate that our mechanism is robust to relaxations of the model assumptions including when competition is based on the lymphocyte receptor specificity. Overall, our work suggests that early life has a strong influence on the long-term structure of the immune repertoire.

100% | Mobil-Ansicht | English Version | Kontakt/Impressum/Datenschutz
DPG-Physik > DPG-Verhandlungen > 2019 > Regensburg