- qx_mri.general.general_extract_glm_volumes(flist, outf, effects, frames, saveoption, values, verbose, txtf)#
general_extract_glm_volumes(flist, outf, effects, frames, saveoption, values, verbose, txtf)
For sessions specified in the session list it extracts the GLM estimates of the effects of interests and saves them in the specified file.
- --flist (str):
A path to the list file or a well structured string of files or sessions to process. For each session id in the list, there has to be a glm: file listed that is a result of GLM analyses (a 'Bcoeff' file).
- --outf (str, default ''):
Root file name for the results. If empty, the flist name is used.
- --effects (cell array | str, default ''):
A cell array of strings or a comma separated list of effects of interest. If empty all effects but Baseline and Trend are extracted.
- --frames (int, default ''):
Frame indeces to extract. If empty, all frames are extracted.
- --saveoption (str, default 'by_session'):
Whether to save the extracted estimates in a single file organized 'by_session', 'by_effect', or in separate files for each effect ('effect_files').
- --values (str, default 'raw'):
What kind of values to save: 'raw' or 'psc'.
- --verbose (bool, default false):
Whether to report on the progress or not.
- --txtf (str, default ''):
An optional designator in what text file to also output the data. Only saved if an option is provided and the input is ptseries. Valid options are 'long' to save the data in long format or empty to skip saving data in a text file.
- --out (nimage)
A nimage object with the extracted glm volumes.
The function is used to extract GLM estimates for the effects of interest for all the specified sessions and save them in a single file (or one file per effect of interest). This files can then be used for more focused analyses, such as second-level statistical testing using PALM.
To extract the effects of interest, the function calls the img_extract_glm_estimates nimage method.
The underlying method extracts the effects of interest by removing those frames that relate to irrelevant effects. The order of the effects in the resulting files will be the same as in the original GLM files when saved organized 'by_session' and not as specified in the call to the function. When the results are organized 'by_effect', the order of estimates will be the same as in the effects variable. To be sure in what order the data is present in the resulting file, please consult the 'list' structure present in the extracted file, that for each frame specifies the session, effect and frame the estimate belongs to.
Additionally, the code does not check for missing estimates. If an estimate is not present in the file, no warning or error will be generated. So do check the list structure that all the data is there.
qunex general_extract_glm_volumes \ --flist='wm-glm.list' \ --outf='wm-encoding-delay' \ --effects='encoding,delay'