interfaces.fsl.model

Cluster

Link to code

Wraps the executable command cluster.

Uses FSL cluster to perform clustering on statistical output

Examples

>>> cl = Cluster()
>>> cl.inputs.threshold = 2.3
>>> cl.inputs.in_file = 'zstat1.nii.gz'
>>> cl.inputs.out_localmax_txt_file = 'stats.txt'
>>> cl.inputs.use_mm = True
>>> cl.cmdline
'cluster --in=zstat1.nii.gz --olmax=stats.txt --thresh=2.3000000000 --mm'

Inputs:

[Mandatory]
threshold: (a float)
        threshold for input volume
        argument: ``--thresh=%.10f``
in_file: (an existing file name)
        input volume
        argument: ``--in=%s``

[Optional]
dlh: (a float)
        smoothness estimate = sqrt(det(Lambda))
        argument: ``--dlh=%.10f``
num_maxima: (an integer (int or long))
        no of local maxima to report
        argument: ``--num=%d``
cope_file: (a file name)
        cope volume
        argument: ``--cope=%s``
pthreshold: (a float)
        p-threshold for clusters
        argument: ``--pthresh=%.10f``
        requires: dlh, volume
out_index_file: (a boolean or a file name)
        output of cluster index (in size order)
        argument: ``--oindex=%s``
out_mean_file: (a boolean or a file name)
        filename for output of mean image
        argument: ``--omean=%s``
find_min: (a boolean, nipype default value: False)
        find minima instead of maxima
        argument: ``--min``
std_space_file: (a file name)
        filename for standard-space volume
        argument: ``--stdvol=%s``
out_localmax_txt_file: (a boolean or a file name)
        local maxima text file
        argument: ``--olmax=%s``
out_localmax_vol_file: (a boolean or a file name)
        output of local maxima volume
        argument: ``--olmaxim=%s``
fractional: (a boolean, nipype default value: False)
        interprets the threshold as a fraction of the robust range
        argument: ``--fractional``
out_max_file: (a boolean or a file name)
        filename for output of max image
        argument: ``--omax=%s``
out_pval_file: (a boolean or a file name)
        filename for image output of log pvals
        argument: ``--opvals=%s``
use_mm: (a boolean, nipype default value: False)
        use mm, not voxel, coordinates
        argument: ``--mm``
minclustersize: (a boolean, nipype default value: False)
        prints out minimum significant cluster size
        argument: ``--minclustersize``
no_table: (a boolean, nipype default value: False)
        suppresses printing of the table info
        argument: ``--no_table``
warpfield_file: (a file name)
        file contining warpfield
        argument: ``--warpvol=%s``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
volume: (an integer (int or long))
        number of voxels in the mask
        argument: ``--volume=%d``
connectivity: (an integer (int or long))
        the connectivity of voxels (default 26)
        argument: ``--connectivity=%d``
out_size_file: (a boolean or a file name)
        filename for output of size image
        argument: ``--osize=%s``
xfm_file: (a file name)
        filename for Linear: input->standard-space transform. Non-linear:
        input->highres transform
        argument: ``--xfm=%s``
peak_distance: (a float)
        minimum distance between local maxima/minima, in mm (default 0)
        argument: ``--peakdist=%.10f``
out_threshold_file: (a boolean or a file name)
        thresholded image
        argument: ``--othresh=%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type

Outputs:

threshold_file: (a file name)
        thresholded image
localmax_vol_file: (a file name)
        output of local maxima volume
size_file: (a file name)
        filename for output of size image
mean_file: (a file name)
        filename for output of mean image
max_file: (a file name)
        filename for output of max image
pval_file: (a file name)
        filename for image output of log pvals
localmax_txt_file: (a file name)
        local maxima text file
index_file: (a file name)
        output of cluster index (in size order)

References:

None

ContrastMgr

Link to code

Wraps the executable command contrast_mgr.

Use FSL contrast_mgr command to evaluate contrasts

In interface mode this file assumes that all the required inputs are in the same location. This has deprecated for FSL versions 5.0.7+ as the necessary corrections file is no longer generated by FILMGLS.

Inputs:

[Mandatory]
dof_file: (an existing file name)
        degrees of freedom
corrections: (an existing file name)
        statistical corrections used within FILM modelling
param_estimates: (a list of items which are an existing file name)
        Parameter estimates for each column of the design matrix
sigmasquareds: (an existing file name)
        summary of residuals, See Woolrich, et. al., 2001
tcon_file: (an existing file name)
        contrast file containing T-contrasts
        argument: ``%s``, position: -1

[Optional]
suffix: (a unicode string)
        suffix to put on the end of the cope filename before the contrast
        number, default is nothing
        argument: ``-suffix %s``
fcon_file: (an existing file name)
        contrast file containing F-contrasts
        argument: ``-f %s``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
contrast_num: (an integer >= 1)
        contrast number to start labeling copes from
        argument: ``-cope``

Outputs:

neffs: (a list of items which are an existing file name)
        neff file ?? for each contrast
varcopes: (a list of items which are an existing file name)
        Variance estimates for each contrast
tstats: (a list of items which are an existing file name)
        t-stat file for each contrast
zstats: (a list of items which are an existing file name)
        z-stat file for each contrast
fstats: (a list of items which are an existing file name)
        f-stat file for each contrast
zfstats: (a list of items which are an existing file name)
        z-stat file for each F contrast
copes: (a list of items which are an existing file name)
        Contrast estimates for each contrast

References:

None

DualRegression

Link to code

Wraps the executable command dual_regression.

Wrapper Script for Dual Regression Workflow

Examples

>>> dual_regression = DualRegression()
>>> dual_regression.inputs.in_files = ["functional.nii", "functional2.nii", "functional3.nii"]
>>> dual_regression.inputs.group_IC_maps_4D = "allFA.nii"
>>> dual_regression.inputs.des_norm = False
>>> dual_regression.inputs.one_sample_group_mean = True
>>> dual_regression.inputs.n_perm = 10
>>> dual_regression.inputs.out_dir = "my_output_directory"
>>> dual_regression.cmdline
'dual_regression allFA.nii 0 -1 10 my_output_directory functional.nii functional2.nii functional3.nii'
>>> dual_regression.run() # doctest: +SKIP

Inputs:

[Mandatory]
n_perm: (an integer (int or long))
        Number of permutations for randomise; set to 1 for just raw tstat
        output, set to 0 to not run randomise at all.
        argument: ``%i``, position: 5
in_files: (a list of items which are an existing file name)
        List all subjects' preprocessed, standard-space 4D datasets
        argument: ``%s``, position: -1
group_IC_maps_4D: (an existing file name)
        4D image containing spatial IC maps (melodic_IC) from the whole-
        group ICA analysis
        argument: ``%s``, position: 1

[Optional]
con_file: (an existing file name)
        Design contrasts for final cross-subject modelling with randomise
        argument: ``%s``, position: 4
out_dir: (a directory name, nipype default value: output)
        This directory will be created to hold all output and logfiles
        argument: ``%s``, position: 6
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
design_file: (an existing file name)
        Design matrix for final cross-subject modelling with randomise
        argument: ``%s``, position: 3
one_sample_group_mean: (a boolean)
        perform 1-sample group-mean test instead of generic permutation test
        argument: ``-1``, position: 3
des_norm: (a boolean, nipype default value: True)
        Whether to variance-normalise the timecourses used as the stage-2
        regressors; True is default and recommended
        argument: ``%i``, position: 2
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type

Outputs:

out_dir: (an existing directory name)

References:

None

FEAT

Link to code

Wraps the executable command feat.

Uses FSL feat to calculate first level stats

Inputs:

[Mandatory]
fsf_file: (an existing file name)
        File specifying the feat design spec file
        argument: ``%s``, position: 0

[Optional]
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables

Outputs:

feat_dir: (an existing directory name)

References:

None

FEATModel

Link to code

Wraps the executable command feat_model.

Uses FSL feat_model to generate design.mat files

Inputs:

[Mandatory]
ev_files: (a list of items which are an existing file name)
        Event spec files generated by level1design
        argument: ``%s``, position: 1
fsf_file: (an existing file name)
        File specifying the feat design spec file
        argument: ``%s``, position: 0

[Optional]
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type

Outputs:

design_cov: (an existing file name)
        Graphical representation of design covariance
design_file: (an existing file name)
        Mat file containing ascii matrix for design
design_image: (an existing file name)
        Graphical representation of design matrix
con_file: (an existing file name)
        Contrast file containing contrast vectors
fcon_file: (a file name)
        Contrast file containing contrast vectors

References:

None

FEATRegister

Link to code

Register feat directories to a specific standard

Inputs:

[Mandatory]
reg_image: (an existing file name)
        image to register to (will be treated as standard)
feat_dirs: (a list of items which are an existing directory name)
        Lower level feat dirs

[Optional]
reg_dof: (an integer (int or long), nipype default value: 12)
        registration degrees of freedom

Outputs:

fsf_file: (an existing file name)
        FSL feat specification file

FILMGLS

Link to code

Wraps the executable command film_gls.

Use FSL film_gls command to fit a design matrix to voxel timeseries

Examples

Initialize with no options, assigning them when calling run:

>>> from nipype.interfaces import fsl
>>> fgls = fsl.FILMGLS()
>>> res = fgls.run('in_file', 'design_file', 'thresh', rn='stats') #doctest: +SKIP

Assign options through the inputs attribute:

>>> fgls = fsl.FILMGLS()
>>> fgls.inputs.in_file = 'functional.nii'
>>> fgls.inputs.design_file = 'design.mat'
>>> fgls.inputs.threshold = 10
>>> fgls.inputs.results_dir = 'stats'
>>> res = fgls.run() #doctest: +SKIP

Specify options when creating an instance:

>>> fgls = fsl.FILMGLS(in_file='functional.nii', design_file='design.mat', threshold=10, results_dir='stats')
>>> res = fgls.run() #doctest: +SKIP

Inputs:

[Mandatory]
in_file: (an existing file name)
        input data file
        argument: ``%s``, position: -3

[Optional]
output_pwdata: (a boolean)
        output prewhitened data and average design matrix
        argument: ``-output_pwdata``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
smooth_autocorr: (a boolean)
        Smooth auto corr estimates
        argument: ``-sa``
results_dir: (a directory name, nipype default value: results)
        directory to store results in
        argument: ``-rn %s``
fit_armodel: (a boolean)
        fits autoregressive model - default is to use tukey with
        M=sqrt(numvols)
        argument: ``-ar``
        mutually_exclusive: autocorr_estimate_only, fit_armodel,
          tukey_window, multitaper_product, use_pava, autocorr_noestimate
design_file: (an existing file name)
        design matrix file
        argument: ``%s``, position: -2
brightness_threshold: (an integer >= 0)
        susan brightness threshold, otherwise it is estimated
        argument: ``-epith %d``
mask_size: (an integer (int or long))
        susan mask size
        argument: ``-ms %d``
tukey_window: (an integer (int or long))
        tukey window size to estimate autocorr
        argument: ``-tukey %d``
        mutually_exclusive: autocorr_estimate_only, fit_armodel,
          tukey_window, multitaper_product, use_pava, autocorr_noestimate
full_data: (a boolean)
        output full data
        argument: ``-v``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
multitaper_product: (an integer (int or long))
        multitapering with slepian tapers and num is the time-bandwidth
        product
        argument: ``-mt %d``
        mutually_exclusive: autocorr_estimate_only, fit_armodel,
          tukey_window, multitaper_product, use_pava, autocorr_noestimate
autocorr_estimate_only: (a boolean)
        perform autocorrelation estimatation only
        argument: ``-ac``
        mutually_exclusive: autocorr_estimate_only, fit_armodel,
          tukey_window, multitaper_product, use_pava, autocorr_noestimate
threshold: (a floating point number >= 0.0, nipype default value:
          1000.0)
        threshold
        argument: ``%f``, position: -1
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
use_pava: (a boolean)
        estimates autocorr using PAVA
        argument: ``-pava``
autocorr_noestimate: (a boolean)
        do not estimate autocorrs
        argument: ``-noest``
        mutually_exclusive: autocorr_estimate_only, fit_armodel,
          tukey_window, multitaper_product, use_pava, autocorr_noestimate

Outputs:

corrections: (an existing file name)
        statistical corrections used within FILM modeling
results_dir: (an existing directory name)
        directory storing model estimation output
param_estimates: (a list of items which are an existing file name)
        Parameter estimates for each column of the design matrix
sigmasquareds: (an existing file name)
        summary of residuals, See Woolrich, et. al., 2001
dof_file: (an existing file name)
        degrees of freedom
logfile: (an existing file name)
        FILM run logfile
thresholdac: (an existing file name)
        The FILM autocorrelation parameters
residual4d: (an existing file name)
        Model fit residual mean-squared error for each time point

References:

None

FLAMEO

Link to code

Wraps the executable command flameo.

Use FSL flameo command to perform higher level model fits

Examples

Initialize FLAMEO with no options, assigning them when calling run:

>>> from nipype.interfaces import fsl
>>> flameo = fsl.FLAMEO()
>>> flameo.inputs.cope_file = 'cope.nii.gz'
>>> flameo.inputs.var_cope_file = 'varcope.nii.gz'
>>> flameo.inputs.cov_split_file = 'cov_split.mat'
>>> flameo.inputs.design_file = 'design.mat'
>>> flameo.inputs.t_con_file = 'design.con'
>>> flameo.inputs.mask_file = 'mask.nii'
>>> flameo.inputs.run_mode = 'fe'
>>> flameo.cmdline
'flameo --copefile=cope.nii.gz --covsplitfile=cov_split.mat --designfile=design.mat --ld=stats --maskfile=mask.nii --runmode=fe --tcontrastsfile=design.con --varcopefile=varcope.nii.gz'

Inputs:

[Mandatory]
cope_file: (an existing file name)
        cope regressor data file
        argument: ``--copefile=%s``
design_file: (an existing file name)
        design matrix file
        argument: ``--designfile=%s``
cov_split_file: (an existing file name)
        ascii matrix specifying the groups the covariance is split into
        argument: ``--covsplitfile=%s``
t_con_file: (an existing file name)
        ascii matrix specifying t-contrasts
        argument: ``--tcontrastsfile=%s``
mask_file: (an existing file name)
        mask file
        argument: ``--maskfile=%s``
run_mode: (u'fe' or u'ols' or u'flame1' or u'flame12')
        inference to perform
        argument: ``--runmode=%s``

[Optional]
sigma_dofs: (an integer (int or long))
        sigma (in mm) to use for Gaussian smoothing the DOFs in FLAME 2.
        Default is 1mm, -1 indicates no smoothing
        argument: ``--sigma_dofs=%d``
f_con_file: (an existing file name)
        ascii matrix specifying f-contrasts
        argument: ``--fcontrastsfile=%s``
burnin: (an integer (int or long))
        number of jumps at start of mcmc to be discarded
        argument: ``--burnin=%d``
outlier_iter: (an integer (int or long))
        Number of max iterations to use when inferring outliers. Default is
        12.
        argument: ``--ioni=%d``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
sample_every: (an integer (int or long))
        number of jumps for each sample
        argument: ``--sampleevery=%d``
n_jumps: (an integer (int or long))
        number of jumps made by mcmc
        argument: ``--njumps=%d``
dof_var_cope_file: (an existing file name)
        dof data file for varcope data
        argument: ``--dofvarcopefile=%s``
no_pe_outputs: (a boolean)
        do not output pe files
        argument: ``--nopeoutput``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
log_dir: (a directory name, nipype default value: stats)
        argument: ``--ld=%s``
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
infer_outliers: (a boolean)
        infer outliers - not for fe
        argument: ``--inferoutliers``
var_cope_file: (an existing file name)
        varcope weightings data file
        argument: ``--varcopefile=%s``
fix_mean: (a boolean)
        fix mean for tfit
        argument: ``--fixmean``

Outputs:

mrefvars: (a list of items which are an existing file name)
        mean random effect variances for each contrast
tstats: (a list of items which are an existing file name)
        t-stat file for each contrast
pes: (a list of items which are an existing file name)
        Parameter estimates for each column of the design matrix for each
        voxel
stats_dir: (a directory name)
        directory storing model estimation output
zstats: (a list of items which are an existing file name)
        z-stat file for each contrast
weights: (a list of items which are an existing file name)
        weights file for each contrast
tdof: (a list of items which are an existing file name)
        temporal dof file for each contrast
var_copes: (a list of items which are an existing file name)
        Variance estimates for each contrast
copes: (a list of items which are an existing file name)
        Contrast estimates for each contrast
fstats: (a list of items which are an existing file name)
        f-stat file for each contrast
zfstats: (a list of items which are an existing file name)
        z stat file for each f contrast
res4d: (a list of items which are an existing file name)
        Model fit residual mean-squared error for each time point

References:

None None

GLM

Link to code

Wraps the executable command fsl_glm.

FSL GLM:

Example

>>> import nipype.interfaces.fsl as fsl
>>> glm = fsl.GLM(in_file='functional.nii', design='maps.nii', output_type='NIFTI')
>>> glm.cmdline
'fsl_glm -i functional.nii -d maps.nii -o functional_glm.nii'

Inputs:

[Mandatory]
design: (an existing file name)
        file name of the GLM design matrix (text time courses for temporal
        regression or an image file for spatial regression)
        argument: ``-d %s``, position: 2
in_file: (an existing file name)
        input file name (text matrix or 3D/4D image file)
        argument: ``-i %s``, position: 1

[Optional]
out_sigsq_name: (a file name)
        output file name for residual noise variance sigma-square
        argument: ``--out_sigsq=%s``
dof: (an integer (int or long))
        set degrees of freedom explicitly
        argument: ``--dof=%d``
out_pf_name: (a file name)
        output file name for p-value for full model fit
        argument: ``--out_pf=%s``
out_t_name: (a file name)
        output file name for t-stats (either as txt or image
        argument: ``--out_t=%s``
out_cope: (a file name)
        output file name for COPE (either as txt or image
        argument: ``--out_cope=%s``
mask: (an existing file name)
        mask image file name if input is image
        argument: ``-m %s``
out_res_name: (a file name)
        output file name for residuals
        argument: ``--out_res=%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
out_vnscales_name: (a file name)
        output file name for scaling factors for variance normalisation
        argument: ``--out_vnscales=%s``
out_varcb_name: (a file name)
        output file name for variance of COPEs
        argument: ``--out_varcb=%s``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
out_z_name: (a file name)
        output file name for Z-stats (either as txt or image
        argument: ``--out_z=%s``
var_norm: (a boolean)
        perform MELODIC variance-normalisation on data
        argument: ``--vn``
out_f_name: (a file name)
        output file name for F-value of full model fit
        argument: ``--out_f=%s``
out_data_name: (a file name)
        output file name for pre-processed data
        argument: ``--out_data=%s``
contrasts: (an existing file name)
        matrix of t-statics contrasts
        argument: ``-c %s``
out_file: (a file name)
        filename for GLM parameter estimates (GLM betas)
        argument: ``-o %s``, position: 3
dat_norm: (a boolean)
        switch on normalization of the data time series to unit std
        deviation
        argument: ``--dat_norm``
demean: (a boolean)
        switch on demeaining of design and data
        argument: ``--demean``
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
des_norm: (a boolean)
        switch on normalization of the design matrix columns to unit std
        deviation
        argument: ``--des_norm``
out_p_name: (a file name)
        output file name for p-values of Z-stats (either as text file or
        image)
        argument: ``--out_p=%s``

Outputs:

out_res: (a list of items which are an existing file name)
        output file name for residuals
out_varcb: (a list of items which are an existing file name)
        output file name for variance of COPEs
out_f: (a list of items which are an existing file name)
        output file name for F-value of full model fit
out_file: (an existing file name)
        file name of GLM parameters (if generated)
out_vnscales: (a list of items which are an existing file name)
        output file name for scaling factors for variance normalisation
out_cope: (a list of items which are an existing file name)
        output file name for COPEs (either as text file or image)
out_sigsq: (a list of items which are an existing file name)
        output file name for residual noise variance sigma-square
out_t: (a list of items which are an existing file name)
        output file name for t-stats (either as text file or image)
out_pf: (a list of items which are an existing file name)
        output file name for p-value for full model fit
out_p: (a list of items which are an existing file name)
        output file name for p-values of Z-stats (either as text file or
        image)
out_data: (a list of items which are an existing file name)
        output file for preprocessed data
out_z: (a list of items which are an existing file name)
        output file name for COPEs (either as text file or image)

References:

None

L2Model

Link to code

Generate subject specific second level model

Examples

>>> from nipype.interfaces.fsl import L2Model
>>> model = L2Model(num_copes=3) # 3 sessions

Inputs:

[Mandatory]
num_copes: (an integer >= 1)
        number of copes to be combined

Outputs:

design_grp: (an existing file name)
        design group file
design_con: (an existing file name)
        design contrast file
design_mat: (an existing file name)
        design matrix file

Level1Design

Link to code

Generate FEAT specific files

Examples

>>> level1design = Level1Design()
>>> level1design.inputs.interscan_interval = 2.5
>>> level1design.inputs.bases = {'dgamma':{'derivs': False}}
>>> level1design.inputs.session_info = 'session_info.npz'
>>> level1design.run() # doctest: +SKIP

Inputs:

[Mandatory]
interscan_interval: (a float)
        Interscan interval (in secs)
model_serial_correlations: (a boolean)
        Option to model serial correlations using an autoregressive
        estimator (order 1). Setting this option is only useful in the
        context of the fsf file. If you set this to False, you need to
        repeat this option for FILMGLS by setting autocorr_noestimate to
        True
bases: (a dictionary with keys which are u'dgamma' and with values
          which are a dictionary with keys which are u'derivs' and with
          values which are a boolean or a dictionary with keys which are
          u'gamma' and with values which are a dictionary with keys which
          are u'derivs' or u'gammasigma' or u'gammadelay' and with values
          which are any value or a dictionary with keys which are u'custom'
          and with values which are a dictionary with keys which are
          u'bfcustompath' and with values which are a unicode string or a
          dictionary with keys which are u'none' and with values which are a
          dictionary with keys which are any value and with values which are
          any value or a dictionary with keys which are u'none' and with
          values which are None)
        name of basis function and options e.g., {'dgamma': {'derivs':
        True}}
session_info: (any value)
        Session specific information generated by ``modelgen.SpecifyModel``

[Optional]
orthogonalization: (a dictionary with keys which are an integer (int
          or long) and with values which are a dictionary with keys which
          are an integer (int or long) and with values which are a boolean
          or an integer (int or long), nipype default value: {})
        which regressors to make orthogonal e.g., {1: {0:0,1:0,2:0}, 2:
        {0:1,1:1,2:0}} to make the second regressor in a 2-regressor model
        orthogonal to the first.
contrasts: (a list of items which are a tuple of the form: (a unicode
          string, u'T', a list of items which are a unicode string, a list
          of items which are a float) or a tuple of the form: (a unicode
          string, u'T', a list of items which are a unicode string, a list
          of items which are a float, a list of items which are a float) or
          a tuple of the form: (a unicode string, u'F', a list of items
          which are a tuple of the form: (a unicode string, u'T', a list of
          items which are a unicode string, a list of items which are a
          float) or a tuple of the form: (a unicode string, u'T', a list of
          items which are a unicode string, a list of items which are a
          float, a list of items which are a float)))
        List of contrasts with each contrast being a list of the form -
        [('name', 'stat', [condition list], [weight list], [session list])].
        if session list is None or not provided, all sessions are used. For
        F contrasts, the condition list should contain previously defined
        T-contrasts.

Outputs:

ev_files: (a list of items which are a list of items which are an
          existing file name)
        condition information files
fsf_files: (a list of items which are an existing file name)
        FSL feat specification files

MELODIC

Link to code

Wraps the executable command melodic.

Multivariate Exploratory Linear Optimised Decomposition into Independent Components

Examples

>>> melodic_setup = MELODIC()
>>> melodic_setup.inputs.approach = 'tica'
>>> melodic_setup.inputs.in_files = ['functional.nii', 'functional2.nii', 'functional3.nii']
>>> melodic_setup.inputs.no_bet = True
>>> melodic_setup.inputs.bg_threshold = 10
>>> melodic_setup.inputs.tr_sec = 1.5
>>> melodic_setup.inputs.mm_thresh = 0.5
>>> melodic_setup.inputs.out_stats = True
>>> melodic_setup.inputs.t_des = 'timeDesign.mat'
>>> melodic_setup.inputs.t_con = 'timeDesign.con'
>>> melodic_setup.inputs.s_des = 'subjectDesign.mat'
>>> melodic_setup.inputs.s_con = 'subjectDesign.con'
>>> melodic_setup.inputs.out_dir = 'groupICA.out'
>>> melodic_setup.cmdline
'melodic -i functional.nii,functional2.nii,functional3.nii -a tica --bgthreshold=10.000000 --mmthresh=0.500000 --nobet -o groupICA.out --Ostats --Scon=subjectDesign.con --Sdes=subjectDesign.mat --Tcon=timeDesign.con --Tdes=timeDesign.mat --tr=1.500000'
>>> melodic_setup.run() # doctest: +SKIP

Inputs:

[Mandatory]
in_files: (a list of items which are an existing file name)
        input file names (either single file name or a list)
        argument: ``-i %s``, position: 0

[Optional]
dim_est: (a unicode string)
        use specific dim. estimation technique: lap, bic, mdl, aic, mean
        (default: lap)
        argument: ``--dimest=%s``
out_dir: (a directory name)
        output directory name
        argument: ``-o %s``
pbsc: (a boolean)
        switch off conversion to percent BOLD signal change
        argument: ``--pbsc``
smode: (an existing file name)
        matrix of session modes for report generation
        argument: ``--smode=%s``
out_orig: (a boolean)
        output the original ICs
        argument: ``--Oorig``
sep_whiten: (a boolean)
        switch on separate whitening
        argument: ``--sep_whiten``
t_des: (an existing file name)
        design matrix across time-domain
        argument: ``--Tdes=%s``
maxit: (an integer (int or long))
        maximum number of iterations before restart
        argument: ``--maxit=%d``
report_maps: (a unicode string)
        control string for spatial map images (see slicer)
        argument: ``--report_maps=%s``
t_con: (an existing file name)
        t-contrast matrix across time-domain
        argument: ``--Tcon=%s``
num_ICs: (an integer (int or long))
        number of IC's to extract (for deflation approach)
        argument: ``-n %d``
no_mm: (a boolean)
        switch off mixture modelling on IC maps
        argument: ``--no_mm``
log_power: (a boolean)
        calculate log of power for frequency spectrum
        argument: ``--logPower``
var_norm: (a boolean)
        switch off variance normalization
        argument: ``--vn``
cov_weight: (a float)
        voxel-wise weights for the covariance matrix (e.g. segmentation
        information)
        argument: ``--covarweight=%f``
tr_sec: (a float)
        TR in seconds
        argument: ``--tr=%f``
mix: (an existing file name)
        mixing matrix for mixture modelling / filtering
        argument: ``--mix=%s``
migpN: (an integer (int or long))
        number of internal Eigenmaps
        argument: ``--migpN %d``
out_all: (a boolean)
        output everything
        argument: ``--Oall``
epsilonS: (a float)
        minimum error change for rank-1 approximation in TICA
        argument: ``--epsS=%f``
migp: (a boolean)
        switch on MIGP data reduction
        argument: ``--migp``
migp_factor: (an integer (int or long))
        Internal Factor of mem-threshold relative to number of Eigenmaps
        (default: 2)
        argument: ``--migp_factor %d``
approach: (a unicode string)
        approach for decomposition, 2D: defl, symm (default), 3D: tica
        (default), concat
        argument: ``-a %s``
update_mask: (a boolean)
        switch off mask updating
        argument: ``--update_mask``
no_mask: (a boolean)
        switch off masking
        argument: ``--nomask``
out_mean: (a boolean)
        output mean volume
        argument: ``--Omean``
bg_threshold: (a float)
        brain/non-brain threshold used to mask non-brain voxels, as a
        percentage (only if --nobet selected)
        argument: ``--bgthreshold=%f``
out_unmix: (a boolean)
        output unmixing matrix
        argument: ``--Ounmix``
sep_vn: (a boolean)
        switch off joined variance normalization
        argument: ``--sep_vn``
epsilon: (a float)
        minimum error change
        argument: ``--eps=%f``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
rem_cmp: (a list of items which are an integer (int or long))
        component numbers to remove
        argument: ``-f %d``
remove_deriv: (a boolean)
        removes every second entry in paradigm file (EV derivatives)
        argument: ``--remove_deriv``
bg_image: (an existing file name)
        specify background image for report (default: mean image)
        argument: ``--bgimage=%s``
non_linearity: (a unicode string)
        nonlinearity: gauss, tanh, pow3, pow4
        argument: ``--nl=%s``
out_pca: (a boolean)
        output PCA results
        argument: ``--Opca``
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
report: (a boolean)
        generate Melodic web report
        argument: ``--report``
s_con: (an existing file name)
        t-contrast matrix across subject-domain
        argument: ``--Scon=%s``
migp_shuffle: (a boolean)
        randomise MIGP file order (default: TRUE)
        argument: ``--migp_shuffle``
dim: (an integer (int or long))
        dimensionality reduction into #num dimensions (default: automatic
        estimation)
        argument: ``-d %d``
s_des: (an existing file name)
        design matrix across subject-domain
        argument: ``--Sdes=%s``
no_bet: (a boolean)
        switch off BET
        argument: ``--nobet``
mask: (an existing file name)
        file name of mask for thresholding
        argument: ``-m %s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
max_restart: (an integer (int or long))
        maximum number of restarts
        argument: ``--maxrestart=%d``
mm_thresh: (a float)
        threshold for Mixture Model based inference
        argument: ``--mmthresh=%f``
ICs: (an existing file name)
        filename of the IC components file for mixture modelling
        argument: ``--ICs=%s``
out_stats: (a boolean)
        output thresholded maps and probability maps
        argument: ``--Ostats``
out_white: (a boolean)
        output whitening/dewhitening matrices
        argument: ``--Owhite``

Outputs:

report_dir: (an existing directory name)
out_dir: (an existing directory name)

References:

None

MultipleRegressDesign

Link to code

Generate multiple regression design

Note

FSL does not demean columns for higher level analysis.

Please see FSL documentation for more details on model specification for higher level analysis.

Examples

>>> from nipype.interfaces.fsl import MultipleRegressDesign
>>> model = MultipleRegressDesign()
>>> model.inputs.contrasts = [['group mean', 'T',['reg1'],[1]]]
>>> model.inputs.regressors = dict(reg1=[1, 1, 1], reg2=[2.,-4, 3])
>>> model.run() # doctest: +SKIP

Inputs:

[Mandatory]
contrasts: (a list of items which are a tuple of the form: (a unicode
          string, u'T', a list of items which are a unicode string, a list
          of items which are a float) or a tuple of the form: (a unicode
          string, u'F', a list of items which are a tuple of the form: (a
          unicode string, u'T', a list of items which are a unicode string,
          a list of items which are a float)))
        List of contrasts with each contrast being a list of the form -
        [('name', 'stat', [condition list], [weight list])]. if session list
        is None or not provided, all sessions are used. For F contrasts, the
        condition list should contain previously defined T-contrasts without
        any weight list.
regressors: (a dictionary with keys which are a unicode string and
          with values which are a list of items which are a float)
        dictionary containing named lists of regressors

[Optional]
groups: (a list of items which are an integer (int or long))
        list of group identifiers (defaults to single group)

Outputs:

design_fts: (an existing file name)
        design f-contrast file
design_grp: (an existing file name)
        design group file
design_mat: (an existing file name)
        design matrix file
design_con: (an existing file name)
        design t-contrast file

Randomise

Link to code

Wraps the executable command randomise.

FSL Randomise: feeds the 4D projected FA data into GLM modelling and thresholding in order to find voxels which correlate with your model

Example

>>> import nipype.interfaces.fsl as fsl
>>> rand = fsl.Randomise(in_file='allFA.nii', mask = 'mask.nii', tcon='design.con', design_mat='design.mat')
>>> rand.cmdline
'randomise -i allFA.nii -o "randomise" -d design.mat -t design.con -m mask.nii'

Inputs:

[Mandatory]
in_file: (an existing file name)
        4D input file
        argument: ``-i %s``, position: 0

[Optional]
one_sample_group_mean: (a boolean)
        perform 1-sample group-mean test instead of generic permutation test
        argument: ``-1``
vox_p_values: (a boolean)
        output voxelwise (corrected and uncorrected) p-value images
        argument: ``-x``
show_total_perms: (a boolean)
        print out how many unique permutations would be generated and exit
        argument: ``-q``
tfce2D: (a boolean)
        carry out Threshold-Free Cluster Enhancement with 2D optimisation
        argument: ``--T2``
f_cm_thresh: (a float)
        carry out f cluster-mass thresholding
        argument: ``-S %.2f``
p_vec_n_dist_files: (a boolean)
        output permutation vector and null distribution text files
        argument: ``-P``
cm_thresh: (a float)
        carry out cluster-mass-based thresholding
        argument: ``-C %.1f``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
f_only: (a boolean)
        calculate f-statistics only
        argument: ``--f_only``
tfce_H: (a float)
        TFCE height parameter (default=2)
        argument: ``--tfce_H=%.2f``
tfce: (a boolean)
        carry out Threshold-Free Cluster Enhancement
        argument: ``-T``
tfce_E: (a float)
        TFCE extent parameter (default=0.5)
        argument: ``--tfce_E=%.2f``
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
tfce_C: (a float)
        TFCE connectivity (6 or 26; default=6)
        argument: ``--tfce_C=%.2f``
raw_stats_imgs: (a boolean)
        output raw ( unpermuted ) statistic images
        argument: ``-R``
f_c_thresh: (a float)
        carry out f cluster thresholding
        argument: ``-F %.2f``
var_smooth: (an integer (int or long))
        use variance smoothing (std is in mm)
        argument: ``-v %d``
design_mat: (an existing file name)
        design matrix file
        argument: ``-d %s``, position: 2
tcon: (an existing file name)
        t contrasts file
        argument: ``-t %s``, position: 3
show_info_parallel_mode: (a boolean)
        print out information required for parallel mode and exit
        argument: ``-Q``
base_name: (a unicode string, nipype default value: randomise)
        the rootname that all generated files will have
        argument: ``-o "%s"``, position: 1
num_perm: (an integer (int or long))
        number of permutations (default 5000, set to 0 for exhaustive)
        argument: ``-n %d``
mask: (an existing file name)
        mask image
        argument: ``-m %s``
c_thresh: (a float)
        carry out cluster-based thresholding
        argument: ``-c %.1f``
x_block_labels: (an existing file name)
        exchangeability block labels file
        argument: ``-e %s``
seed: (an integer (int or long))
        specific integer seed for random number generator
        argument: ``--seed=%d``
demean: (a boolean)
        demean data temporally before model fitting
        argument: ``-D``
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type
fcon: (an existing file name)
        f contrasts file
        argument: ``-f %s``

Outputs:

tstat_files: (a list of items which are an existing file name)
        t contrast raw statistic
t_p_files: (a list of items which are an existing file name)
        f contrast uncorrected p values files
f_p_files: (a list of items which are an existing file name)
        f contrast uncorrected p values files
f_corrected_p_files: (a list of items which are an existing file
          name)
        f contrast FWE (Family-wise error) corrected p values files
t_corrected_p_files: (a list of items which are an existing file
          name)
        t contrast FWE (Family-wise error) corrected p values files
fstat_files: (a list of items which are an existing file name)
        f contrast raw statistic

References:

None

SMM

Link to code

Wraps the executable command mm --ld=logdir.

Spatial Mixture Modelling. For more detail on the spatial mixture modelling see Mixture Models with Adaptive Spatial Regularisation for Segmentation with an Application to FMRI Data; Woolrich, M., Behrens, T., Beckmann, C., and Smith, S.; IEEE Trans. Medical Imaging, 24(1):1-11, 2005.

Inputs:

[Mandatory]
spatial_data_file: (an existing file name)
        statistics spatial map
        argument: ``--sdf="%s"``, position: 0
mask: (an existing file name)
        mask file
        argument: ``--mask="%s"``, position: 1

[Optional]
no_deactivation_class: (a boolean)
        enforces no deactivation class
        argument: ``--zfstatmode``, position: 2
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type

Outputs:

activation_p_map: (an existing file name)
deactivation_p_map: (an existing file name)
null_p_map: (an existing file name)

References:

None

SmoothEstimate

Link to code

Wraps the executable command smoothest.

Estimates the smoothness of an image

Examples

>>> est = SmoothEstimate()
>>> est.inputs.zstat_file = 'zstat1.nii.gz'
>>> est.inputs.mask_file = 'mask.nii'
>>> est.cmdline
'smoothest --mask=mask.nii --zstat=zstat1.nii.gz'

Inputs:

[Mandatory]
mask_file: (an existing file name)
        brain mask volume
        argument: ``--mask=%s``
dof: (an integer (int or long))
        number of degrees of freedom
        argument: ``--dof=%d``
        mutually_exclusive: zstat_file

[Optional]
residual_fit_file: (an existing file name)
        residual-fit image file
        argument: ``--res=%s``
        requires: dof
args: (a unicode string)
        Additional parameters to the command
        argument: ``%s``
environ: (a dictionary with keys which are a newbytes or None or a
          newstr or None and with values which are a newbytes or None or a
          newstr or None, nipype default value: {})
        Environment variables
zstat_file: (an existing file name)
        zstat image file
        argument: ``--zstat=%s``
        mutually_exclusive: dof
output_type: (u'NIFTI_PAIR' or u'NIFTI_PAIR_GZ' or u'NIFTI_GZ' or
          u'NIFTI')
        FSL output type

Outputs:

volume: (an integer (int or long))
        number of voxels in mask
resels: (a float)
        number of resels
dlh: (a float)
        smoothness estimate sqrt(det(Lambda))

References:

None

load_template()

Link to code

Load a template from the model_templates directory

Parameters

name : str
The name of the file to load

Returns

template : string.Template