emodelrunner.synplas_analysis

Analysis tools for synapse plasticity.

Functions

epsp_slope(vtrace)

Returns slope between points at [0.3, 0.75] * peak of given trace.

get_epsp_vector(t, v, spikes, window)

Extract EPSPs at time spikes from voltage trace v.

Classes

Experiment(data[, c01duration, c02duration, ...])

A full STDP induction experiment.
class emodelrunner.synplas_analysis.Experiment(data, c01duration=40.0, c02duration=40.0, period=10.0, c01period=None, c02period=None)[source]

Bases: object

A full STDP induction experiment.

The experiment consists of two connectivity tests (C01 and C02), separed by the induction protocol.

t

Time vector (ms)

Type:

numpy.ndarray

v

Soma voltage trace (ms)

Type:

numpy.ndarray

spikes

Pre-synaptic spike times (ms)

Type:

numpy.ndarray

duration

Duration of C01 and C02 phases (ms)

Type:

dict

period

The Cx period (ms)

Type:

float

c01period

The C01 period (ms)

Type:

float

c02period

The C02 period (ms)

Type:

float

epspwindow

Size of EPS detection window (ms)

Type:

float

cxs

Names of the Cx tests. These names are used as keys in the dictionaries (cxspikes, duration, epsp, cxtrace)

Type:

list

cxspikes

Pre-synaptic spikes times (ms) occuring during each test (C01 & C02)

Type:

dict

compute_epsp_interval(interval)[source]

Compute mean EPSP amplitude at regular intervals.

Parameters:

interval (float) – The interval in minutes.

Returns:

A dictionary of interval statistics for each Cx in Experiment.

Return type:

dict

compute_epsp_ratio(n, method='amplitude', full=False)[source]

Compute mean EPSP change.

Parameters:

  • n (int) – Number of sweeps in Cx to be considered for mean EPSP calculation

  • method (str) – Method used to compute EPSP ratio (amplitude or slope)

  • full (bool) – whether to return the mean value and std of epsp before and after cell stimulus in addition to the epsp ratio

Returns:

if full is False, returns

  • ratio_epsp (float): Mean EPSP change

if full is True, returns a tuple containing

  • epsp_before (float): mean value of EPSP or slope of mean EPSP trace in test C01

  • epsp_after (float): mean value of EPSP or slope of mean EPSP trace in test C02

  • ratio_epsp (float): Mean EPSP change

  • epsp_before_std (float): std of EPSP in test C01 if method is amplitude, else 0

  • epsp_after_std (float): std of EPSP in test C02 if method is amplitude, else 0

Raises:

ValueError – if method is neither ‘amplitude’ nor ‘slope’

property cxtrace

Returns time and interpolated trace of each EPSP as dict.

property epsp

Returns dict containing EPSP vectors for each test (mV).

normalize_time(t)[source]

Normalize time vector t.

Convert milliseconds to minutes and shift t = 0 to beginning of induction phase.

Parameters:

t (numpy.ndarray) – Time vector (ms)

Results:

tnorm (numpy.ndarray): Normalized time vector (mn)

emodelrunner.synplas_analysis.epsp_slope(vtrace)[source]

Returns slope between points at [0.3, 0.75] * peak of given trace.

Attention! This function expects that the trace is sampled (or interpolated) at a single timestep, since time is not taken into account during calculations.

Parameters:

vtrace (numpy.ndarray) – interpolated voltage trace of one EPSP (mV)

Returns:

slope of the EPSP (mV)

Return type:

float

emodelrunner.synplas_analysis.get_epsp_vector(t, v, spikes, window)[source]

Extract EPSPs at time spikes from voltage trace v.

Parameters:

  • t (numpy.ndarray) – Time vector (ms).

  • v (numpy.ndarray) – Soma voltage trace. Must have the same size of t (mV).

  • spikes (numpy.ndarray) – Time of presynaptic spikes. Every value in spikes must be in the interval [min(t), max(t)] (ms).

  • window (float) – Size of EPS detection window in ms.

Returns:

Vector of EPSPs (mV).

Return type:

numpy.ndarray

Raises:

  • RuntimeError – if a postsynaptic cell spike is found during a connectivity test

  • RuntimeError – if the detection window can have multiple EPSPs in it