Functions¶
Individual Bold File Functions¶
-
atlascorr.atlas_correlations.get_files(img)[source]¶ Find the brainmask and confound files given the bold file.
Parameters: - img : str
full path and name of the bold file
Returns: - confound : str
full path and name of the confounds file
- brainmask : str
full path and name of the brainmask file
-
atlascorr.atlas_correlations.proc_confounds(confounds, confound_file)[source]¶ Filter confounds file to selected confounds & replaces “n/a”s in confounds file with the mean.
Parameters: - confounds : list
list of confounds to include in the model
- confounds_file : str
full path and name of the confounds file
Returns: - confounds_df : pandas.core.frame.DataFrame
dataframe containing the selected confounds
-
atlascorr.atlas_correlations.make_corr_matrix(ts_matrix)[source]¶ Make a symmetric pearson’s r->z transforme correlation matrix.
Parameters: - ts_matrix : numpy.ndarray
2D numpy array with each column representing an atlas region and each row representing a volume (time point)
Returns: - zcorr_matrix : numpy.ndarray
2D symmetric matrix measuring region-region correlations main diagnal is all zeros
-
atlascorr.atlas_correlations.write_out_corr_matrix(corr_matrix, atlas_lut, img, output_dir)[source]¶ Write out a symmetric correlation matrix using BIDS naming conventions
Parameters: - corr_matrix : numpy.ndarray
2D symmetric matrix measuring region-region correlations main diagnal is all zeros
- atlas_lut : str
full path and name to atlas lookup tsv with two columns (regions and index)
- img : str
full path and name of the bold file
- output_dir : str
full path to the base directory where all correlation matrices will be written out to.
Returns: - dst : str
full path and name of the correlation matrix
Group Level Functions¶
-
atlascorr.atlas_correlations.proc_matrix(matrix_tsv)[source]¶ Vectorize symmetric correlation matrix so that each unique region-region correlation gets a column.
Parameters: - matrix_tsv : str
full path and name of the correlation matrix
Returns: - flat_df : pandas.core.frame.DataFrame
a flat dataframe that is one entry and has as many columns as there are unique region-region pairs
-
atlascorr.atlas_correlations.merge_dfs(dfs)[source]¶ Merge a list of dataframes where each contains one row showing all unique region-region pairs.
Parameters: - dfs : list
list of dataframes where each contains one row showing all unique region-region pairs
Returns: - out_df : pandas.core.frame.DataFrame
merged dataframe where each row represents a unique scan