A pipeline example that uses intergrates several interfaces to perform a first and second level analysis on a two-subject data set.
import numpy as np
import nipype.interfaces.io as nio # Data i/o
import nipype.interfaces.fsl as fsl # fsl
import nipype.interfaces.utility as util # utility
import nipype.pipeline.engine as pe # pypeline engine
import nipype.algorithms.modelgen as model # model generation
import os # system functions
#####################################################################
# Preliminaries
# Tell fsl to generate all output in compressed nifti format
print fsl.Info.version()
fsl.FSLCommand.set_default_output_type('NIFTI_GZ')
extract_ref = pe.Node(interface=fsl.ExtractROI(t_min=42,
t_size=1),
name = 'extractref')
# run FSL's bet
# bet my_structural my_betted_structural
in the provided data set, the nose is behind the head and causes problems for segmentation routines
nosestrip = pe.Node(interface=fsl.BET(frac=0.3),
name = 'nosestrip')
skullstrip = pe.Node(interface=fsl.BET(mask = True),
name = 'stripstruct')
refskullstrip = pe.Node(interface=fsl.BET(mask = True),
name = 'stripref')
coregister = pe.Node(interface=fsl.FLIRT(dof=6),
name = 'coregister')
# Preprocess functionals
motion_correct = pe.Node(interface=fsl.MCFLIRT(save_plots = True),
name='realign')
#iterfield = ['in_file'])
skull strip functional data
func_skullstrip = pe.Node(interface=fsl.BET(functional = True),
name='stripfunc')
#iterfield = ['in_file'])
Run FAST on T1 anatomical image to obtain CSF mask. Create mask for three tissue types.
getCSFmasks = pe.Node(interface=fsl.FAST(no_pve=True,segments=True),
name = 'segment')
Apply registration matrix to CSF segmentation mask.
applyReg2CSFmask = pe.Node(interface=fsl.ApplyXfm(apply_xfm=True),
name = 'applyreg2csfmask')
Threshold CSF segmentation mask from .90 to 1
threshCSFseg = pe.Node(interface = fsl.ImageMaths(op_string = ' -thr .90 -uthr 1 -bin '),
name = 'threshcsfsegmask')
Extract CSF timeseries
avgCSF = pe.Node(interface = fsl.ImageMeants(), name='extractcsfts')
def pickfirst(files):
return files[0]
Create the workflow
csffilter = pe.Workflow(name='csffilter')
csffilter.connect([(extract_ref, motion_correct,[('roi_file', 'ref_file')]),
(extract_ref, refskullstrip,[('roi_file', 'in_file')]),
(nosestrip, skullstrip, [('out_file','in_file')]),
(skullstrip, getCSFmasks,[('out_file','in_files')]),
(skullstrip, coregister,[('mask_file','in_file')]),
(refskullstrip, coregister,[('out_file','reference')]),
(motion_correct, func_skullstrip, [('out_file', 'in_file')]),
(getCSFmasks, applyReg2CSFmask,[(('tissue_class_files',pickfirst),'in_file')]),
(refskullstrip, applyReg2CSFmask,[('out_file','reference')]),
(coregister, applyReg2CSFmask,[('out_matrix_file','in_matrix_file')]),
(applyReg2CSFmask,threshCSFseg,[('out_file','in_file')]),
(func_skullstrip,avgCSF,[('out_file','in_file')]),
(threshCSFseg,avgCSF,[('out_file','mask')]),
])
modelfit = pe.Workflow(name='modelfit')
c. Use :class:`nipype.interfaces.spm.SpecifyModel` to generate
SPM-specific design information.
modelspec = pe.Node(interface=model.SpecifyModel(), name="modelspec")
d. Use :class:`nipype.interfaces.fsl.Level1Design` to generate a
run specific fsf file for analysis
level1design = pe.Node(interface=fsl.Level1Design(), name="fsfdesign")
e. Use :class:`nipype.interfaces.fsl.FEATModel` to generate a
run specific mat file for use by FILMGLS
modelgen = pe.Node(interface=fsl.FEATModel(), name='modelgen')
f. Use :class:`nipype.interfaces.fsl.FILMGLS` to estimate a model
specified by a mat file and a functional run
modelestimate = pe.Node(interface=fsl.FILMGLS(), name='modelestimate')
#iterfield = ['design_file','in_file'])
modelfit.connect([(modelspec,level1design,[('session_info','session_info')]),
(level1design,modelgen,[('fsf_files','fsf_file'),
('ev_files', 'ev_files')]),
(modelgen,modelestimate,[('design_file','design_file')]),
])
The nipype tutorial contains data for two subjects. Subject data is in two subdirectories, s1 and s2. Each subject directory contains four functional volumes: f3.nii, f5.nii, f7.nii, f10.nii. And one anatomical volume named struct.nii.
Below we set some variables to inform the datasource about the layout of our data. We specify the location of the data, the subject sub-directories and a dictionary that maps each run to a mnemonic (or field) for the run type (struct or func). These fields become the output fields of the datasource node in the pipeline.
In the example below, run ‘f3’ is of type ‘func’ and gets mapped to a nifti filename through a template ‘%s.nii’. So ‘f3’ would become ‘f3.nii’.
# Specify the location of the data.
data_dir = os.path.abspath('data')
# Specify the subject directories
subject_list = ['s1']
# Map field names to individual subject runs.
info = dict(func=[['subject_id', ['f3',]]], #'f5','f7','f10']]],
struct=[['subject_id','struct']])
infosource = pe.Node(interface=util.IdentityInterface(fields=['subject_id']),
name="infosource")
Here we set up iteration over all the subjects. The following line is a particular example of the flexibility of the system. The datasource attribute iterables tells the pipeline engine that it should repeat the analysis on each of the items in the subject_list. In the current example, the entire first level preprocessing and estimation will be repeated for each subject contained in subject_list.
infosource.iterables = ('subject_id', subject_list)
Now we create a nipype.interfaces.io.DataSource object and fill in the information from above about the layout of our data. The nipype.pipeline.NodeWrapper module wraps the interface object and provides additional housekeeping and pipeline specific functionality.
datasource = pe.Node(interface=nio.DataGrabber(infields=['subject_id'],
outfields=['func', 'struct']),
name = 'datasource')
datasource.inputs.base_directory = data_dir
datasource.inputs.template = '%s/%s.nii'
datasource.inputs.template_args = info
a. Setup a function that returns subject-specific information about
the experimental paradigm. This is used by the
:class:`nipype.modelgen.SpecifyModel` to create the
information necessary to generate an SPM design matrix. In this
tutorial, the same paradigm was used for every participant. Other
examples of this function are available in the `doc/examples`
folder. Note: Python knowledge required here.
def subjectinfo(meantsfile):
import numpy as np
from nipype.interfaces.base import Bunch
ts = np.loadtxt(meantsfile)
output = [Bunch(regressor_names=['MeanIntensity'],
regressors=[ts.tolist()])]
return output
hpcutoff = np.inf
TR = 3.
modelfit.inputs.modelspec.input_units = 'secs'
modelfit.inputs.modelspec.time_repetition = TR
modelfit.inputs.modelspec.high_pass_filter_cutoff = hpcutoff
modelfit.inputs.fsfdesign.interscan_interval = TR
modelfit.inputs.fsfdesign.bases = {'none': None}
modelfit.inputs.fsfdesign.model_serial_correlations = False
modelfit.inputs.modelestimate.autocorr_noestimate = True
Band pass filter the data to remove frequencies below .1 Hz
bandPassFilterData = pe.Node(interface=fsl.ImageMaths(op_string = ' -bptf 128 12.5 '),
name='bandpassfiltermcdata_fslmaths')
l1pipeline = pe.Workflow(name= "resting")
l1pipeline.base_dir = os.path.abspath('./fslresting/workingdir')
l1pipeline.connect([(infosource, datasource, [('subject_id', 'subject_id')]),
(datasource, csffilter, [('struct','nosestrip.in_file'),
('func', 'realign.in_file'),
#(('func', pickfirst), 'extractref.in_file'),
('func', 'extractref.in_file'),
]),
(csffilter, modelfit, [('stripfunc.out_file', 'modelspec.functional_runs'),
('realign.par_file', 'modelspec.realignment_parameters'),
(('extractcsfts.out_file', subjectinfo),'modelspec.subject_info'),
('stripfunc.out_file', 'modelestimate.in_file')
]),
(modelfit, bandPassFilterData, [('modelestimate.residual4d', 'in_file')]),
])
if __name__ == '__main__':
l1pipeline.run()
l1pipeline.write_graph()
Example source code
You can download the full source code of this example. This same script is also included in the Nipype source distribution under the examples directory.