def renameItem(self):
"""
Give a new display name to the currently selected data item
"""
item = self.getSanitizedItemSelectionData()
if item is None:
return
if (item[0] is DATA_SET_ITEM):
data = DataStore.get(item[1])
dlg = EditNameDialog(self.Parent, data.displayname)
if dlg.ShowModal() == wx.ID_OK:
data.displayname = dlg.Text
dlg.Destroy()
if (item[0] is FIGURE_SET_ITEM):
figure = FigureStore.get(item[1])
dlg = EditNameDialog(self.Parent, figure.name)
if dlg.ShowModal() == wx.ID_OK:
figure.name = dlg.Text
dlg.Destroy()
item = self.tree.GetSelection()
item.SetText(dlg.Text)
self.tree.RefreshSelected()
item.SetHilight(False)
item.SetHilight(True)
def scatterplot2D(subplot, figure, dims):
# set default plot options if necessary
opts = subplot.opts
if len(opts) == 0:
opts['xRange'] = ()
opts['xRangeAuto'] = True
opts['yRange'] = ()
opts['yRangeAuto'] = True
opts['xTransform'] = ''
opts['yTransform'] = ''
opts['transformAuto'] = True
# Set axes transforms
if (opts['transformAuto']):
fcData = DataStore.get(subplot.dataIndex)
tf = ('xTransform', 'yTransform')
for i, dim in enumerate(dims):
t = fcData.getDefaultTransform(dim)
if (t is not None) and ('log' in t.lower()):
opts[tf[i]] = 'log'
else:
opts[tf[i]] = 'linear'
if opts['xRangeAuto']:
opts['xRange'] = (1, np.max(subplot.Data[:,dims[0]])*1.5)
if opts['yRangeAuto']:
opts['yRange'] = (1, np.max(subplot.Data[:,dims[1]])*1.5)
# create the subplot and set its attributes
subplot.axes = figure.add_subplot(subplot.mnp, xlim=opts['xRange'],
ylim=opts['yRange'], autoscale_on=False)
subplot.axes.set_xscale(opts['xTransform'], nonposx='clip')
subplot.axes.set_yscale(opts['yTransform'], nonposy='clip')
subplot.axes.set_xlabel(subplot.Labels[dims[0]])
subplot.axes.set_ylabel(subplot.Labels[dims[1]])
subplot.axes.set_title(subplot.Title)
# draw the supplied FACS data
if (not subplot.isDataClustered()):
subplot.axes.plot(subplot.Data[:,dims[0]],
subplot.Data[:,dims[1]],
'.', ms=1, color='black')
else:
data = separate(subplot.Data, subplot.Clustering)
for i in range(len(data)):
xs = data[i][:,dims[0]]
ys = data[i][:,dims[1]]
subplot.axes.plot(xs, ys, '.', ms=1, color=methods.plotColors[i])
def saveState(dir, filename):
"""
Save a representation of the system state: All the loaded data sets,
their clusterings, any transformations or analyses (future),
and all plots.
The state is saved in JSON format based on a dict of the following form:
data: list of IDs
binfile: the filename of the binary file used to store all the actual data
data-dID: dict of settings belonging to a FacsData instance
clust-dID-cID: a dict of attributes belonging to a clustering
figures: list of figureID strings
fig-ID: A dict for each figure keyed on the ID. The subplot attribute here
is replaced with a list of fig-ID-p-ID strings for locating subplot dicts
fig-ID-p-ID: A dict for each subplot in each figure keyed on fig ID and plot ID.
current-data: data ID
current-figure: figure ID
"""
store = shelve.open(os.path.join(dir, filename))
#store = dbopen(os.path.join(dir, filename), 'c', format='csv')
# The actual numeric data will be stored in a separate binary file using
# the numpy savez() method allowing for efficient storage/retrieval of data
binfile = '%s.npz' % filename
bindata = {}
store['data'] = DataStore.getData().keys()
store['binfile'] = binfile
for dID in DataStore.getData():
fdata = DataStore.get(dID)
dStr = 'data-%i' % dID
dfname = fdata.filename if (fdata.filename is not '') else binfile
bindata[dStr] = fdata.data
store[dStr] = {'filename': dfname,
'displayname': fdata.displayname,
'labels': fdata.labels,
'annotations': fdata.annotations,
'analysis': fdata.analysis,
'ID': fdata.ID,
'parent': fdata.parent,
'children': fdata.children,
'selDims': fdata.selDims,
'clustering': fdata.clustering.keys(),
'nodeExpanded': fdata.nodeExpanded,
'selectedClustering': fdata.selectedClustering}
# clusterings
for cID in fdata.clustering:
cStr = 'clust-%i-%i' % (dID, cID)
csett = {'method': fdata.methodIDs[cID],
'opts': fdata.clusteringOpts[cID],
'clusteringSelDims': fdata.clusteringSelDims[cID],
'infoExpanded': fdata.infoExpanded[cID]}
store[cStr] = csett
bindata[cStr] = fdata.clustering[cID]
# figures
sfigs = []
for figID in FigureStore.getFigures():
fig = FigureStore.get(figID)
fStr = 'fig-%i' % figID
d = dict(fig.__dict__)
splots = packSubplots(store, figID, fig.subplots)
d['subplots'] = splots
store[fStr] = d
sfigs.append(fStr)
store['figures'] = list(sfigs)
# other
store['current-data'] = DataStore.getCurrentIndex()
store['current-figure'] = FigureStore.getSelectedIndex()
# write out settings data
store.close()
# write out numeric data to binary file
np.savez(os.path.join(dir, binfile), **bindata)
def clearClusteringSelection(self):
item = self.getSanitizedItemSelectionData()
if item is not None and item[0] is DATA_SET_ITEM:
DataStore.get(item[1]).selectedClustering = None
self.updateTree()
def displayDataInfo(self):
item = self.getSanitizedItemSelectionData()
if item is not None and item[0] is DATA_SET_ITEM:
data = DataStore.get(item[1])
dlg = DataInfoDialog(self.Parent, data)
dlg.Show()
