Resource competition in growing neurites (Hjorth et al 2014)

 Download zip file 
Help downloading and running models
Computer model of neurite outgrowth in a simplified neuron. A growth limiting resource is produced in the soma, transported through the neurites and consumed at the growth cones.
1 . Hjorth JJ, van Pelt J, Mansvelder HD, van Ooyen A (2014) Competitive dynamics during resource-driven neurite outgrowth. PLoS One 9:e86741 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Neuron or other electrically excitable cell;
Brain Region(s)/Organism:
Cell Type(s):
Gap Junctions:
Simulation Environment: Python;
Model Concept(s): Simplified Models; Development;
Implementer(s): Hjorth, Johannes [hjorth at];
# Keeps track of the simulation time

from time import time

class Clock():
  Simulation clock

  def __init__(self,dt=1.0,start=0.0,end=20.0):
    self._dt = float(dt)
    self._t = float(start)
    self._start = float(start)
    self._end = float(end)

  def reset(self):
    self._t = float(self._start)

  def tick(self):
    self._t += self._dt

  def set_dt(self,dt):
    self._dt = float(dt)

  def get_dt(self):
    return self._dt

  dt = property(fget=get_dt,fset=set_dt)

  def set_start(self,start):
    self._start = float(start)

  def get_start(self):
    return self._start

  start = property(fget=get_start,fset=set_start)

  def set_end(self,end):
    self._end = float(end)

  def get_end(self):
    return self._end

  end = property(fget=get_end,fset=set_end)

  def done(self):
    return (self._t >= self._end)

  def curTime(self):
    return self._t


Loading data, please wait...