9.8.2. Massively univariate analysis of a calculation task from the Localizer datasetΒΆ

This example shows how to use the Localizer dataset in a basic analysis. A standard Anova is performed (massively univariate F-test) and the resulting Bonferroni-corrected p-values are plotted. We use a calculation task and 20 subjects out of the 94 available.

The Localizer dataset contains many contrasts and subject-related variates. The user can refer to the plot_localizer_mass_univariate_methods.py example to see how to use these.

# Author: Virgile Fritsch, <virgile.fritsch@inria.fr>, May. 2014
import numpy as np
import matplotlib.pyplot as plt
from nilearn import datasets
from nilearn.input_data import NiftiMasker
from nilearn.image import get_data

Load Localizer contrast

Out:

/home/nicolas/anaconda3/envs/nilearn/lib/python3.8/site-packages/numpy/lib/npyio.py:2405: VisibleDeprecationWarning: Reading unicode strings without specifying the encoding argument is deprecated. Set the encoding, use None for the system default.
  output = genfromtxt(fname, **kwargs)

Mask data

nifti_masker = NiftiMasker(
    smoothing_fwhm=5,
    memory='nilearn_cache', memory_level=1)  # cache options
cmap_filenames = localizer_dataset.cmaps
fmri_masked = nifti_masker.fit_transform(cmap_filenames)

Anova (parametric F-scores)

Out:

/home/nicolas/anaconda3/envs/nilearn/lib/python3.8/site-packages/sklearn/utils/validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
  return f(*args, **kwargs)

Visualization

from nilearn.plotting import plot_stat_map, show

# Various plotting parameters
z_slice = 45  # plotted slice

threshold = - np.log10(0.1)  # 10% corrected

# Plot Anova p-values
fig = plt.figure(figsize=(5, 6), facecolor='w')
display = plot_stat_map(neg_log_pvals_anova_unmasked,
                        threshold=threshold,
                        display_mode='z', cut_coords=[z_slice],
                        figure=fig)

masked_pvals = np.ma.masked_less(get_data(neg_log_pvals_anova_unmasked),
                                 threshold)

title = ('Negative $\\log_{10}$ p-values'
         '\n(Parametric + Bonferroni correction)'
         '\n%d detections' % (~masked_pvals.mask).sum())

display.title(title, y=1.1, alpha=0.8)

show()
plot localizer simple analysis

Total running time of the script: ( 0 minutes 3.100 seconds)

Gallery generated by Sphinx-Gallery