Python API

Python API#

Reading a file#

The most useful features of this package it to read imaging data which can be done as follow

import mps
# Object containing the frames, time stamps and metadata
data = mps.MPS("file.nd2")

The package supports numerous types of formats including .nd2, .czi and .tif.

Accessing info about the file#

You can print some general info about the dataset by accessing the .info attribute

>>> print(data.info)
{'num_frames': 591, 'dt': 10.007577876721394, 'time_unit': 'ms', 'um_per_pixel': 2.589535590385804, 'size_x': 415, 'size_y': 154}

Most of fields here are self-explanatory. num_frames is the number of frames in the image stack, dt is the average time difference between two successive frames, time_unit is the unit of time (either 'ms' or 's'), um_per_pixel is a factor to convert from pixel unit to micro meters. Finally size_x and size_y is the number of pixels in the x- and y direction.

There is also another attribute called .metadata that contains all the metadata that is possible to extract from the file

>>> print(data.metadata)

{'ImageTextInfoLV': {'SLxImageTextInfo': {'TextInfoItem_0': '', 'TextInfoItem_1': '', 'TextInfoItem_2': '', 'TextInfoItem_3': '', 'TextInfoItem_4': '', 'TextInfoItem_5': 'Metadata:\r\nDimensions: T(591) x λ(1)\r\nCamera Name: Flash4.0, SN:001336\r\nNumerical Aperture: 0.3\r\nRefractive Index: 1\r\n Name: Red_VC\r\n Component Count: 1\r\n Modality: Widefield Fluorescence\r\n Camera Settings:   Exposure: 10 ms\r\n  Binning: 4x4\r\n  Scan Mode: Fast\r\n Microscope Settings:
...

Imaging data#

data.frames gives you access to the frames which are numpy arrays. You can verify that the shape of the frames matches the sizes from the info dictionary

>>> print(data.frames.shape)
(415, 154, 591)

You can also check the data type

>>> print(data.frames.dtype)
dtype('uint16')

The time stamps are stored as an attribute called .time_stamps

>>> print(data.time_stamps.shape)
(591, )

Finally, in some cases pacing amplitude is also available in the metadata, and this can be accessed from the attribute .pacing

>>> print(data.pacing.shape)
(591, )

If pacing information is not available then data.pacing will only contain zeros.

Analyzing imaging data#

To convert the data into a trace you can use the library ap_features

import ap_features as apf

# Compute the average over all pixels
y = mps.average.get_average_all(data.frames)
# Convert it to an apf.Beats and compute features
# using the ap_features package
trace = apf.Beats(y=y, t=data.time_stamps, pacing=data.pacing)

See the documentation of ap_features to learn more about how to analyze a trace.

Contents