BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Clustering of cell surface receptors: Simulating the mesoscale bet
 ween reaction-diffusion and atomistic scales - Jun Allard (University of C
 alifornia\, Irvine)
DTSTART:20160623T080000Z
DTEND:20160623T084500Z
UID:TALK66551@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>Co-author: Omer Dushek (Oxford) <br></span> <br>Many bio
 logical molecules\, including cell surface receptors\, form  densely-packe
 d clusters that are weakly bound\, mechanically soft\, and have  volumes o
 n the same order as the volumes of the proteins they interact with.  Preve
 nting the formation of clusters dramatically attenuates proper cell  funct
 ion in many examples (including T cell activation and allergen activation 
  in Mast cells)\, but for unknown reason. Therefore\, receptor clusters in
 volve  biology hidden at the mesoscale between individual protein structur
 e (~0.1nm)  and the cell-scale signaling pathways of populations of diffus
 ing protein  (~1000nm). In some parameter regimes\, clusters comprise 10-1
 00 molecules tied to  fixed locations on the cell surface by molecular tet
 hers. The Dushek Lab is  developing an in vitro setup that mimics this reg
 ime\, and find that the time  courses of binding and enzymatic reactions a
 re non-trivial and cannot be fit to  simple ODE models. On the other hand\
 , fitting to explicitly spatial simulatio ns  with volume exclusion is pro
 hibitively slow. Here we present a fast algorithm  for tethered reactions 
 with volume exclusion. The algorithm exploits\, first\, the  spatially-fix
 ed tethers\, allowing us to construct a single nearest-neighbor  tree\, an
 d\, second\, a separation of timescales between the fast diffusion of  mol
 ecular domains and slow binding and catalytic reactions. This allows use o
 f a  hybrid Metropolis-Gillespie algorithm: on the fast timescale of domai
 n motion\,  efficient equilibrium algorithms that include volume exclusion
  provide the  effective concentrations for the slow timescale of binding a
 nd catalysis\, which  are simulated using a maximally-fast next-event algo
 rithm. Crucially\, we employ  dynamic connected-set-discovery subroutines 
 to simulate the minimal subset of  molecules each time step. The algorithm
  has computational time scaling  approximately with the number of molecule
 s and can reproduce the non-trivial  time courses observed experimentally.
  <br> <br>Related Links <ul> <li><a target="_blank" rel="nofollow">http://
 allardlab.com</a> - Jun Allard&#39\;s  website&nbsp\;</li></ul>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR