Module plottools.neurons
Draw sketches of neurons.
Axes member functions
neuron(): draw a sketch of a neuron.
Install/uninstall neurons functions
You usually do not need to call these functions. Upon loading the neurons
module, install_neurons() is called automatically.
install_neurons(): install functions of the neurons module in matplotlib.uninstall_neurons(): uninstall all code of the neurons module from matplotlib.
Expand source code
"""
Draw sketches of neurons.
## Axes member functions
- `neuron()`: draw a sketch of a neuron.
## Install/uninstall neurons functions
You usually do not need to call these functions. Upon loading the neurons
module, `install_neurons()` is called automatically.
- `install_neurons()`: install functions of the neurons module in matplotlib.
- `uninstall_neurons()`: uninstall all code of the neurons module from matplotlib.
"""
import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt
def neuron(ax, xy, r, label=None, xytarget=None, synapse='exc', adapt=0, xyinput=None,
fc='white', ec='black', lw=2, **kwargs):
""" Draw a sketch of a neuron.
The coordinate system should have equal distances in both directions
for the cell bodies being true circles. That is, you should call
```
ax.set_aspect('equal')
```
or
```
import plottools.axes
ax.set_ylim_equal()
```
Parameters
----------
ax: matplotlib axes
Axes into which to draw the neuron.
xy: tuple of floats
Coordinates of center of cell body in data coordinates.
r: float
Radius of cell body in data coordinates.
label: string or None
If not `None`, text to printed inside the cell body.
xytarget: tuple of floats, sequence of tuple of floats or None
If not `None`, list of x and y coordinates to which connections are drawn.
The coordinates are the centers of the target neurons in data coordinates
from which the radius `r` is subtracted.
synapse: 'exc', 'inh' or 'arr' or lis thereof
For each `xytarget` the type of synapse to be drawn: excitatory synapse (atriangle),
inhibitory synapse (a bar), or an arrow.
adapt: int
If not zero draw a self-inhibitory circular connection,
indicating an adaptation current.
1: draw self-inhibition to the side of the top-most connection to a target neuron.
2: draw self-inhibition opposite of outgoing connections to target neurons.
3: draw self-inhibition on top.
xyinput: tuple of floats or None
If not `None`, draw an input arrow originating from the given coordinates
(in data coordinates).
fc: matplotlib color specification
Fill color for the cell body.
ec: matplotlib color specification
Edge color for the cell body and line color for all the connections,
synapses and arrows.
lw: float
Line width for the cell body and line color for all the connections,
synapses and arrows.
**kwargs: key-word arguments
Arguments passed on to `ax.text()` for drawing the `label`.
"""
xy = np.asarray(xy)
arx = 0.5*r
ary = 0.4*arx
# inputs:
if xyinput is not None:
xyinput = np.asarray(xyinput)
rx = 1.2*r
d = xyinput - xy
dd = np.sqrt(np.dot(d, d))
theta = np.arctan2(d[1], d[0])
tt = mpl.transforms.Affine2D().rotate(theta)
tt += mpl.transforms.Affine2D().translate(xy[0], xy[1])
tt += ax.transData
# input arrow:
ax.plot((rx, dd), (0.0, 0.0), color=ec, lw=lw, clip_on=False, transform=tt)
ax.plot((rx+arx, rx, rx+arx), (ary, 0.0, -ary), color=ec, lw=lw,
clip_on=False, transform=tt)
# targets:
dm = None
t = None
if xytarget is not None:
if not isinstance(xytarget[0], (list, tuple)):
xytarget = [xytarget]
if not isinstance(synapse, (list, tuple)):
synapse = [synapse]*len(xytarget)
x = 0.5*r
y = 0.4*x
ym = None
dm = np.zeros((len(xytarget), 2))
for k, (xyt, syn) in enumerate(zip(xytarget, synapse)):
xyt = np.asarray(xyt)
d = xyt - xy
dm[k,:] = d
dd = np.sqrt(np.dot(d, d))
dd -= r
theta = np.arctan2(d[1], d[0])
tt = mpl.transforms.Affine2D().rotate(theta)
tt += mpl.transforms.Affine2D().translate(xy[0], xy[1])
tt += ax.transData
if ym is None or ym < d[1]:
t = tt
ym = d[1]
# synapse:
if 'exc' in syn:
dd -= x+0.2*r
ax.plot((dd, dd+x, dd+x, dd), (0.0, -y, y, 0.0),
clip_on=False, color=ec, lw=lw, transform=tt)
elif 'inh' in syn:
dd -= 0.2*r
ax.plot((dd, dd), (-y, y), color=ec, lw=lw, clip_on=False, transform=tt)
elif 'arr' in syn:
dd -= 0.2*r
ax.plot((dd-arx, dd, dd-arx), (-ary, 0.0, ary),
clip_on=False, color=ec, lw=lw, transform=tt)
# axon:
ax.plot((r, dd), (0.0, 0.0), color=ec, lw=lw, clip_on=False, transform=tt)
# adaptation:
if adapt > 0:
ar = 1.5*r
if adapt == 1 and t is not None:
t = mpl.transforms.Affine2D().rotate_deg(-30) + t
elif adapt == 2 and dm is not None:
dm = np.mean(dm, axis=0)
theta = np.arctan2(-dm[1], -dm[0]) - 0.5*np.pi
t = mpl.transforms.Affine2D().rotate(theta)
t += mpl.transforms.Affine2D().translate(xy[0], xy[1])
t += ax.transData
else:
t = mpl.transforms.Affine2D().translate(xy[0], xy[1])
t += ax.transData
ax.add_patch(mpl.patches.Arc((0.0, ar), ar, ar, 0.0, -50.0, 220.0,
transform=t, clip_on=False, color=ec, lw=lw))
ax.plot([1.15*r, 1.2*r], [-0.35*r, 0.35*r], color=ec, lw=lw, clip_on=False,
transform=mpl.transforms.Affine2D().rotate_deg(120) + t)
# cell body:
ax.add_patch(mpl.patches.Circle((xy[0], xy[1]), r, ec='none', fc=fc, lw=0, clip_on=False))
ax.add_patch(mpl.patches.Circle((xy[0], xy[1]), r, ec=ec, fc='none', lw=lw, clip_on=False))
# label:
if label:
figw, _ = ax.get_figure().get_size_inches()
pw = ax.get_position().width * figw * 72.0
aw = np.diff(ax.get_xlim())[0]
fs = 'medium'
if 'fs' in kwargs:
fs = kwargs.pop('fs')
kwargs['fontsize'] = fs
if 'fontsize' in kwargs:
fs = kwargs['fontsize']
fp = mpl.font_manager.FontProperties(size=fs)
cfs = fp.get_size_in_points()
cf = cfs*aw/pw
ax.text(xy[0]-0.03*cf, xy[1]-0.08*cf, label, ha='center', va='center',
clip_on=False, **kwargs)
def install_neurons():
""" Install neurons functions on matplotlib axes.
This function is also called automatically upon importing the module.
See also
--------
uninstall_neurons()
"""
if not hasattr(mpl.axes.Axes, 'neuron'):
mpl.axes.Axes.neuron = neuron
def uninstall_neurons():
""" Uninstall neurons functions from matplotlib axes.
Call this code to disable anything that was installed by `install_neurons()`.
See also
--------
install_neurons()
"""
if hasattr(mpl.axes.Axes, 'neuron'):
delattr(mpl.axes.Axes, 'neuron')
install_neurons()
def demo():
""" Run a demonstration of the neurons module.
"""
fig, ax = plt.subplots()
fig.suptitle('plottools.neurons')
ax.set_aspect('equal')
ax.set_xlim(0, 10)
neuronstyle = dict(ec='gray', lw=1.0, fs='medium')
neuronA = dict(fc='blue', **neuronstyle)
neuronB = dict(fc='green', **neuronstyle)
neuronT = dict(fc='orange', **neuronstyle)
neuronS = dict(fc='yellow', **neuronstyle)
neuronR = dict(fc='red', **neuronstyle)
r = 0.6 # neuron radius
d = 3.5 # short distance between neurons
dd = 1.3*d # long distance between neurons
ad = 0.7*d # input arrow length
x = 2.5 # position of first neuron
y = 0.35*d # y-position of two neurons
yt = 1.8*y # y-position of three neurons
nt = (x+dd, yt)
ns = (x+dd, 0)
nr = (x+dd, -yt)
ax.neuron((x, y), r, 'L', [nt, ns, nr],
['exc', 'exc', 'inh'], 3, (x-ad, y), **neuronA)
ax.neuron((x, -y), r, 'R', [nt, ns, nr],
['inh', 'exc', 'exc'], 3, (x-ad, -y), **neuronB)
ax.neuron(nt, r, '$-$', (x+dd+d, yt), 'exc', adapt=1, **neuronT)
ax.neuron(ns, r, '$+$', (x+dd+d, 0), 'exc', adapt=0, **neuronS)
ax.neuron(nr, r, '$-$', (x+dd+d, -yt), 'exc', adapt=1, **neuronR)
plt.show()
if __name__ == "__main__":
demo()
Functions
def neuron(ax, xy, r, label=None, xytarget=None, synapse='exc', adapt=0, xyinput=None, fc='white', ec='black', lw=2, **kwargs)-
Draw a sketch of a neuron.
The coordinate system should have equal distances in both directions for the cell bodies being true circles. That is, you should call
ax.set_aspect('equal')or
import plottools.axes ax.set_ylim_equal()Parameters
ax:matplotlib axes- Axes into which to draw the neuron.
xy:tupleoffloats- Coordinates of center of cell body in data coordinates.
r:float- Radius of cell body in data coordinates.
label:stringorNone- If not
None, text to printed inside the cell body. xytarget:tupleoffloats, sequenceoftupleoffloatsorNone- If not
None, list of x and y coordinates to which connections are drawn. The coordinates are the centers of the target neurons in data coordinates from which the radiusris subtracted. synapse:'exc', 'inh'or'arr'orlis thereof- For each
xytargetthe type of synapse to be drawn: excitatory synapse (atriangle), inhibitory synapse (a bar), or an arrow. adapt:int- If not zero draw a self-inhibitory circular connection, indicating an adaptation current. 1: draw self-inhibition to the side of the top-most connection to a target neuron. 2: draw self-inhibition opposite of outgoing connections to target neurons. 3: draw self-inhibition on top.
xyinput:tupleoffloatsorNone- If not
None, draw an input arrow originating from the given coordinates (in data coordinates). fc:matplotlib color specification- Fill color for the cell body.
ec:matplotlib color specification- Edge color for the cell body and line color for all the connections, synapses and arrows.
lw:float- Line width for the cell body and line color for all the connections, synapses and arrows.
**kwargs:key-word arguments- Arguments passed on to
ax.text()for drawing thelabel.
Expand source code
def neuron(ax, xy, r, label=None, xytarget=None, synapse='exc', adapt=0, xyinput=None, fc='white', ec='black', lw=2, **kwargs): """ Draw a sketch of a neuron. The coordinate system should have equal distances in both directions for the cell bodies being true circles. That is, you should call ``` ax.set_aspect('equal') ``` or ``` import plottools.axes ax.set_ylim_equal() ``` Parameters ---------- ax: matplotlib axes Axes into which to draw the neuron. xy: tuple of floats Coordinates of center of cell body in data coordinates. r: float Radius of cell body in data coordinates. label: string or None If not `None`, text to printed inside the cell body. xytarget: tuple of floats, sequence of tuple of floats or None If not `None`, list of x and y coordinates to which connections are drawn. The coordinates are the centers of the target neurons in data coordinates from which the radius `r` is subtracted. synapse: 'exc', 'inh' or 'arr' or lis thereof For each `xytarget` the type of synapse to be drawn: excitatory synapse (atriangle), inhibitory synapse (a bar), or an arrow. adapt: int If not zero draw a self-inhibitory circular connection, indicating an adaptation current. 1: draw self-inhibition to the side of the top-most connection to a target neuron. 2: draw self-inhibition opposite of outgoing connections to target neurons. 3: draw self-inhibition on top. xyinput: tuple of floats or None If not `None`, draw an input arrow originating from the given coordinates (in data coordinates). fc: matplotlib color specification Fill color for the cell body. ec: matplotlib color specification Edge color for the cell body and line color for all the connections, synapses and arrows. lw: float Line width for the cell body and line color for all the connections, synapses and arrows. **kwargs: key-word arguments Arguments passed on to `ax.text()` for drawing the `label`. """ xy = np.asarray(xy) arx = 0.5*r ary = 0.4*arx # inputs: if xyinput is not None: xyinput = np.asarray(xyinput) rx = 1.2*r d = xyinput - xy dd = np.sqrt(np.dot(d, d)) theta = np.arctan2(d[1], d[0]) tt = mpl.transforms.Affine2D().rotate(theta) tt += mpl.transforms.Affine2D().translate(xy[0], xy[1]) tt += ax.transData # input arrow: ax.plot((rx, dd), (0.0, 0.0), color=ec, lw=lw, clip_on=False, transform=tt) ax.plot((rx+arx, rx, rx+arx), (ary, 0.0, -ary), color=ec, lw=lw, clip_on=False, transform=tt) # targets: dm = None t = None if xytarget is not None: if not isinstance(xytarget[0], (list, tuple)): xytarget = [xytarget] if not isinstance(synapse, (list, tuple)): synapse = [synapse]*len(xytarget) x = 0.5*r y = 0.4*x ym = None dm = np.zeros((len(xytarget), 2)) for k, (xyt, syn) in enumerate(zip(xytarget, synapse)): xyt = np.asarray(xyt) d = xyt - xy dm[k,:] = d dd = np.sqrt(np.dot(d, d)) dd -= r theta = np.arctan2(d[1], d[0]) tt = mpl.transforms.Affine2D().rotate(theta) tt += mpl.transforms.Affine2D().translate(xy[0], xy[1]) tt += ax.transData if ym is None or ym < d[1]: t = tt ym = d[1] # synapse: if 'exc' in syn: dd -= x+0.2*r ax.plot((dd, dd+x, dd+x, dd), (0.0, -y, y, 0.0), clip_on=False, color=ec, lw=lw, transform=tt) elif 'inh' in syn: dd -= 0.2*r ax.plot((dd, dd), (-y, y), color=ec, lw=lw, clip_on=False, transform=tt) elif 'arr' in syn: dd -= 0.2*r ax.plot((dd-arx, dd, dd-arx), (-ary, 0.0, ary), clip_on=False, color=ec, lw=lw, transform=tt) # axon: ax.plot((r, dd), (0.0, 0.0), color=ec, lw=lw, clip_on=False, transform=tt) # adaptation: if adapt > 0: ar = 1.5*r if adapt == 1 and t is not None: t = mpl.transforms.Affine2D().rotate_deg(-30) + t elif adapt == 2 and dm is not None: dm = np.mean(dm, axis=0) theta = np.arctan2(-dm[1], -dm[0]) - 0.5*np.pi t = mpl.transforms.Affine2D().rotate(theta) t += mpl.transforms.Affine2D().translate(xy[0], xy[1]) t += ax.transData else: t = mpl.transforms.Affine2D().translate(xy[0], xy[1]) t += ax.transData ax.add_patch(mpl.patches.Arc((0.0, ar), ar, ar, 0.0, -50.0, 220.0, transform=t, clip_on=False, color=ec, lw=lw)) ax.plot([1.15*r, 1.2*r], [-0.35*r, 0.35*r], color=ec, lw=lw, clip_on=False, transform=mpl.transforms.Affine2D().rotate_deg(120) + t) # cell body: ax.add_patch(mpl.patches.Circle((xy[0], xy[1]), r, ec='none', fc=fc, lw=0, clip_on=False)) ax.add_patch(mpl.patches.Circle((xy[0], xy[1]), r, ec=ec, fc='none', lw=lw, clip_on=False)) # label: if label: figw, _ = ax.get_figure().get_size_inches() pw = ax.get_position().width * figw * 72.0 aw = np.diff(ax.get_xlim())[0] fs = 'medium' if 'fs' in kwargs: fs = kwargs.pop('fs') kwargs['fontsize'] = fs if 'fontsize' in kwargs: fs = kwargs['fontsize'] fp = mpl.font_manager.FontProperties(size=fs) cfs = fp.get_size_in_points() cf = cfs*aw/pw ax.text(xy[0]-0.03*cf, xy[1]-0.08*cf, label, ha='center', va='center', clip_on=False, **kwargs) def install_neurons()-
Install neurons functions on matplotlib axes.
This function is also called automatically upon importing the module.
See Also
Expand source code
def install_neurons(): """ Install neurons functions on matplotlib axes. This function is also called automatically upon importing the module. See also -------- uninstall_neurons() """ if not hasattr(mpl.axes.Axes, 'neuron'): mpl.axes.Axes.neuron = neuron def uninstall_neurons()-
Uninstall neurons functions from matplotlib axes.
Call this code to disable anything that was installed by
install_neurons().See Also
Expand source code
def uninstall_neurons(): """ Uninstall neurons functions from matplotlib axes. Call this code to disable anything that was installed by `install_neurons()`. See also -------- install_neurons() """ if hasattr(mpl.axes.Axes, 'neuron'): delattr(mpl.axes.Axes, 'neuron') def demo()-
Run a demonstration of the neurons module.
Expand source code
def demo(): """ Run a demonstration of the neurons module. """ fig, ax = plt.subplots() fig.suptitle('plottools.neurons') ax.set_aspect('equal') ax.set_xlim(0, 10) neuronstyle = dict(ec='gray', lw=1.0, fs='medium') neuronA = dict(fc='blue', **neuronstyle) neuronB = dict(fc='green', **neuronstyle) neuronT = dict(fc='orange', **neuronstyle) neuronS = dict(fc='yellow', **neuronstyle) neuronR = dict(fc='red', **neuronstyle) r = 0.6 # neuron radius d = 3.5 # short distance between neurons dd = 1.3*d # long distance between neurons ad = 0.7*d # input arrow length x = 2.5 # position of first neuron y = 0.35*d # y-position of two neurons yt = 1.8*y # y-position of three neurons nt = (x+dd, yt) ns = (x+dd, 0) nr = (x+dd, -yt) ax.neuron((x, y), r, 'L', [nt, ns, nr], ['exc', 'exc', 'inh'], 3, (x-ad, y), **neuronA) ax.neuron((x, -y), r, 'R', [nt, ns, nr], ['inh', 'exc', 'exc'], 3, (x-ad, -y), **neuronB) ax.neuron(nt, r, '$-$', (x+dd+d, yt), 'exc', adapt=1, **neuronT) ax.neuron(ns, r, '$+$', (x+dd+d, 0), 'exc', adapt=0, **neuronS) ax.neuron(nr, r, '$-$', (x+dd+d, -yt), 'exc', adapt=1, **neuronR) plt.show()