API reference¶

`lyman.frontend`: Front-end interface¶

Forward facing lyman tools with information about ecosystem.

Interface functions¶

`frontend.info`([experiment, model, lyman_dir])	Load information from various files to control analyses.
`frontend.subjects`([subject_arg, sessions, …])	Find a list of subjects in a variety of ways.
`frontend.execute`(wf, args, info)	Main interface for (probably) executing a nipype workflow.

Information classes¶

`frontend.LymanInfo`	Combination of all information classes.
`frontend.ProjectInfo`	General information common to a project.
`frontend.ExperimentInfo`	More specific experiment-level information.
`frontend.ModelInfo`	Model-specific level of information about a specific model.

`lyman.glm`: General linear modeling¶

HRF models¶

`glm.HRFModel`	Abstract base class for HRF models used in design construction.
`glm.GammaHRF`([res, duration, pos_shape, …])	Double gamma variate model of cannonical HRF.
`glm.GammaBasis`([time_derivative, …])	Basis set for HRF based on Gamma variate model.
`glm.FIRBasis`(n[, offset, suffix])	Finite Impulse Response basis model.
`glm.IdentityHRF`	Model that does not alter input during transform; useful for testing.

Design construction¶

`glm.build_design_matrix`([conditions, …])	Use design information to build a matrix for a BOLD time series GLM.
`glm.condition_to_regressors`(name, condition, …)	Generate design matrix columns from information about event occurrence.
`glm.contrast_matrix`(contrast, design_matrix)	Return a contrast matrix that is valid for a given design matrix.

Model estimation¶

`glm.prewhiten_image_data`(ts_img, mask_img, X)	Estimate autocorrelation and transform data and design for OLS.
`glm.estimate_residual_autocorrelation`(Y, X)	Fit OLS model and estimate residual autocorrelation with regularization.
`glm.iterative_ols_fit`(Y, X)	Fit a linear model using ordinary least squares in each voxel.
`glm.iterative_contrast_estimation`(B, SS, …)	Compute contrast parameter and variance estimates in each voxel.
`glm.contrast_fixed_effects`(G, V)	Compute higher-order fixed effects parameters.

Temporal filtering¶

`glm.highpass_filter_matrix`(n_tp, cutoff[, tr])	Return an array to implement a gaussian running line filter.
`glm.highpass_filter`(data, cutoff[, tr, copy])	Highpass filter data with gaussian running line filter.

`lyman.signals`: Signal processing¶

Time series transformation and diagnostics¶

`signals.percent_change`(data[, axis])	Convert data to percent signal change over specified axis.
`signals.detrend`(data[, axis, replace_mean])	Linearly detrend on an axis, optionally replacing the original mean.
`signals.pca_transform`(data[, keep, whiten])	Transform data matrix using PCA.
`signals.identify_noisy_voxels`(ts_img, mask_img)	Create a mask of voxels that are unusually noisy given neighbors.
`signals.cv`(data[, axis, detrend, mask, …])	Compute the temporal coefficient of variation.

Spatial filtering¶

`signals.smooth_volume`(data_img, fwhm[, …])	Filter volume data with an isotropic gaussian kernel.
`signals.smooth_segmentation`(data_img, …[, …])	Filter each compartment of a segmentation with an isotropic gaussian.
`signals.smooth_surface`(data_img, vert_img, …)	Smooth cortical voxels with Gaussian weighted surface distances.
`signals.smoothing_matrix`(measure, vertids, fwhm)	Define a matrix to smooth voxels using surface geometry.
`signals.voxel_sigmas`(fwhm, img)	Convert isotropic fwhm in mm to an array of voxelwise sigmas.

`lyman.surface`: Surface mesh operations¶

Geodesic distance measurement¶

surface.SurfaceMeasure(verts, faces)

Object for computing distance along a surface mesh.

Data representation conversion¶

surface.vol_to_surf(data_img, subject, hemi)

Sample data from a volume image onto a surface mesh.

`lyman.utils`: Utilities¶

Custom Nipype interfaces¶

`utils.LymanInterface`(**inputs)	Enhanced Interface object that custom interface should inherit from.
`utils.SaveInfo`(**inputs)

Data representation conversion¶

`utils.image_to_matrix`(img, mask_img[, use])	Extract image data from voxels using segmentation mask.
`utils.matrix_to_image`(data, mask_img[, …])	Convert a vector or matrix of data into a nibabel image.
`utils.check_mask`(mask, data)	Check the dtype and shape of a mask array.

`lyman.visualization`: Data visualization¶

Classes for image representation¶

`visualizations.Mosaic`(anat[, stat, mask, …])
`visualizations.CarpetPlot`(data, seg[, …])

Model visualization¶

`visualizations.plot_design_matrix`(X[, title])	Show the design matrix as a transposed heatmap.
`visualizations.plot_nuisance_variables`(X[, …])	Show the timeseries of each nuisance variable, by source.