POP_CLUSTEDIT	Graphic user interface (GUI)-based function with editing and plotting options for visualizing and manipulating an input STUDY structure. Only component measures (e.g., dipole locations, scalp maps, spectra, ERPs, ERSPs, ITCs) that have been computed and saved in the study EEG datasets can be visualized. These can be computed during pre-clustering using the GUI-based function pop_preclust() or the equivalent command line functions eeg_createdata() and eeg_preclust(). To use dipole locations for clustering, they must first be stored in the EEG dataset structures using dipfit(). Supported cluster editing functions include new cluster creation, cluster merging, outlier rejection, and cluster renaming. Components can also be moved from one cluster to another or to the outlier cluster.
Usage:	>> STUDY = pop_clustedit(STUDY, ALLEEG, clusters);
Inputs:
ALLEEG	Top-level EEGLAB vector of loaded EEG structures for the dataset(s) in the STUDY. ALLEEG for a STUDY set is typically loaded using pop_loadstudy(), or in creating a new STUDY, using pop_createstudy().
STUDY	EEGLAB STUDY set comprising some or all of the EEG datasets in ALLEEG.
Optional inputs:
clusters	[integer vector] of cluster numbers. These clusters will be visualized and manipulated in the pop_clustedit graphic interface. There are restrictions on which clusters can be loaded together. The clusters must either originate from the same clustering (same pre_clustering() and subsequent pop_clust execution), or they must all be leaf clusters (i.e., clusters with no child clusters) {default: all leaf clusters}.
Outputs:
STUDY	The input STUDY set structure modified according to specified user edits, if any. Plotted cluster measure means (maps, ERSPs, etc.) are added to the STUDY structure after they are first plotted to allow quick replotting.
Graphic interface buttons:
"Select cluster to plot"	[list box] Displays available clusters to plot (format is 'cluster name (number of components)'). The presented clusters depend on the optional input variable 'clusters'. Selecting (clicking on) a cluster from the list will display the selected cluster components in the "Select component(s) to plot" list box. Use the plotting buttons below to plot selected measures of the selected cluster. Additional editing options (renaming the cluster, rejecting outliers, moving components to another cluster) are also available. The option 'All N cluster centroids' at the top of the list displays all the clusters in the list except the 'Notcluster', 'Outlier' and 'ParentCluster' clusters. Selecting this option will plot the cluster centroids (i.e. ERP, ERSP, ...) in a single figure.
"Select component(s) to plot"	[list box] Displays the ICA components of the currently selected cluster (in the "Select cluster to plot" list box). Each component has the format: 'subject name, component index'. Multiple components can be selected from the list. Use the plotting buttons below to plot different measures of the selected components on different figures. Selecting the "All components" option is equivalent to using the cluster plotting buttons. Additional editing options are reassigning the selected components to another cluster or moving them to the outlier cluster.
"Plot Cluster properties"	[button] Displays in one figure all the mean cluster measures (e.g., dipole locations, scalp maps, spectra, etc.) that were calculated and saved in the EEG datsets. If there is more than one condition, the ERP and the spectrum will have different colors for each condition. The ERSP and ITC plots will show only the first condition; clicking on the subplot will open a new figure with the different conditions displayed together. Uses the command line function std_propplot().
"Plot scalp maps"	[button] Displays the scalp maps of cluster components. If applied to a cluster, scalp maps of the cluster components are plotted along with the cluster mean scalp map in one figure. If "All # cluster centroids" option is selected, all cluster scalp map means are plotted in the same figure. If applied to components, displays the scalp maps of the specified cluster components in separate figures. Uses the command line functions std_plotmap() and std_plotcompmap().
"Plot ERSPs"	[button] Displays the cluster component ERSPs. If applied to a cluster, component ERSPs are plotted in one figure (per condition) with the cluster mean ERSP. If "All # cluster centroids" option is selected, plots all average ERSPs of the clusters in one figure per condition. If applied to components, display the ERSP images of specified cluster components in separate figures, using one figure for all conditions. Uses the command line functions std_plotersp() and std_plotcompersp().
"Plot ITCs"	[button] Same as "Plot ERSPs" but with ITC. Uses the command line functions std_plotitc() and std_plotcompitc().
"Plot dipoles"	[button] Displays the dipoles of the cluster components. If applied to a cluster, plots the cluster component dipoles (in blue) plus the average cluster dipole (in red). If "All # cluster centroids" option is selected, all cluster plots are displayed in one figure each cluster in a separate subplot. If applied to components, displays the ERSP images of the specified cluster. For specific components displays components dipole (in blue) plus the average cluster dipole (in Red) in separate figures. Uses the command line functions std_dipplot() and std_plotcompdip().
"Plot spectra"	[button] Displays the cluster component spectra. If applied to a cluster, displays component spectra plus the average cluster spectrum in bold. For a specific cluster, displays the cluster component spectra plus the average cluster spectrum (in bold) in one figure per condition. If the "All # cluster centroids" option is selected, displays the average spectrum of all clusters in the same figure, with spectrum for different conditions (if any) plotted in different colors. If applied to components, displays the spectrum of specified cluster components in separate figures using one figure for all conditions. Uses the command line functions std_plotspec() and std_plotcompspec().
"Plot ERPs"	[button] Same as "Plot spectra" but for ERPs. Uses the command line functions std_ploterp() and std_plotcomperp().
"Create new cluster"	[button] Creates a new empty cluster. Opens a popup window in which a name for the new cluster can be entered. If no name is given the default name is 'Cls #', where '#' is the next available cluster number. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. After the empty cluster is created, components can be moved into it using, 'Reassign selected component(s)' (see below). Uses the command line function std_createclust().
"Rename selected cluster"	[button] Renames a cluster using the selected (mnemonic) name. Opens a popup window in which a new name for the selected cluster can be entered. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. Uses the command line function std_renameclust().
"Reject outlier components"	[button] rejects outlier components to an outlier cluster. Opens a popup window to specify the outlier threshold. Move outlier components that are more than x standard deviations devs from the cluster centroid to an outlier cluster. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. Uses the command line function std_rejectoutliers().
"Merge clusters"	[button] Merges several clusters into one cluster. Opens a popup window in which the clusters to merge may be specified An optional name can be given to the merged cluster. If no name is given, the default name is 'Cls #', where '#' is the next available cluster number. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. Uses the command line function std_mergeclust().
"Remove selected outlier component(s)"	[button] Moves selected component(s) to the outlier cluster. The components that will be moved are the ones selected in the "Select component(s) to plot" list box. Opens a popup window in which a list of the selected component(s) is presented. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. Uses the command line function std_moveoutlier().
"Reassign selected component(s)"	[button] Moves selected component(s) from one cluster to another. The components that will reassign are the ones selected in the "Select component(s) to plot" list box. Opens a popup window in which a list of possible clusters to which to move the selected component(s) is presented. For changes to take place, press the popup window 'OK' button, else press the 'Cancel' button. Uses the command line function std_movecomp().
"Save STUDY set to disk"	[check box] Saves the STUDY set structure modified according to specified user edits to the disk. If no file name is entered will overwrite the current STUDY set file.
See also:	pop_preclust(), pop_clust.
Authors:	Arnaud Delorme, Hilit Serby, Scott Makeig, SCCN/INC/UCSD, October 11, 2004

See the matlab file pop_clustedit.m (may require other functions)

Back to functions