Author: bugman
Date: Thu Apr 11 17:46:46 2013
New Revision: 19469
URL: http://svn.gna.org/viewcvs/relax?rev=19469&view=rev
Log:
Implemented the relax_disp.cluster user function.
This is for clustering spins together for a dispersion analysis.
Modified:
branches/relax_disp/specific_analyses/relax_disp.py
branches/relax_disp/user_functions/relax_disp.py
Modified: branches/relax_disp/specific_analyses/relax_disp.py
URL:
http://svn.gna.org/viewcvs/relax/branches/relax_disp/specific_analyses/relax_disp.py?rev=19469&r1=19468&r2=19469&view=diff
==============================================================================
--- branches/relax_disp/specific_analyses/relax_disp.py (original)
+++ branches/relax_disp/specific_analyses/relax_disp.py Thu Apr 11 17:46:46
2013
@@ -282,6 +282,65 @@
# Return the correct peak intensity series.
return results[0, index]
+
+
+ def _cluster(self, cluster_id=None, spin_id=None):
+ """Define spin clustering.
+
+ @keyword cluster_id: The cluster ID string.
+ @type cluster_id: str
+ @keyword spin_id: The spin ID string for the spin or group of
spins to add to the cluster.
+ @type spin_id: str
+ """
+
+ # Initialise.
+ if not hasattr(cdp, 'clustering'):
+ # Create the dictionary.
+ cdp.clustering = {}
+ cdp.clustering['free spins'] = []
+
+ # Add all spin IDs to the cluster.
+ for spin, id in spin_loop(return_id=True):
+ cdp.clustering['free spins'].append(id)
+
+ # Add the key.
+ if cluster_id not in cdp.clustering:
+ cdp.clustering[cluster_id] = []
+
+ # Loop over the spins to add to the cluster.
+ for spin, id in spin_loop(selection=spin_id, return_id=True):
+ # First remove the ID from all clusters.
+ for key in cdp.clustering.keys():
+ if id in cdp.clustering[key]:
+ cdp.clustering[key].pop(cdp.clustering[key].index(id))
+
+ # Then add the ID to the cluster.
+ cdp.clustering[cluster_id].append(id)
+
+ # Clean up - delete any empty clusters.
+ for key in cdp.clustering.keys():
+ if cdp.clustering[key] == []:
+ cdp.clustering.pop(key)
+
+
+ def _cluster_ids(self):
+ """Return the current list of cluster ID strings.
+
+ @return: The list of cluster IDs.
+ @rtype: list of str
+ """
+
+ # Initialise.
+ ids = ['free spins']
+
+ # Add the defined IDs.
+ if hasattr(cdp, 'cluster'):
+ for key in list(cdp.cluster.keys()):
+ if key not in ids:
+ ids.append(key)
+
+ # Return the IDs.
+ return ids
def _cpmg_delayT(self, id=None, delayT=None):
Modified: branches/relax_disp/user_functions/relax_disp.py
URL:
http://svn.gna.org/viewcvs/relax/branches/relax_disp/user_functions/relax_disp.py?rev=19469&r1=19468&r2=19469&view=diff
==============================================================================
--- branches/relax_disp/user_functions/relax_disp.py (original)
+++ branches/relax_disp/user_functions/relax_disp.py Thu Apr 11 17:46:46 2013
@@ -25,6 +25,7 @@
# relax module imports.
from pipe_control import spectrum
+from pipe_control.mol_res_spin import get_spin_ids
from graphics import WIZARD_IMAGE_PATH
from specific_analyses.setup import relax_disp_obj
from user_functions.data import Uf_info; uf_info = Uf_info()
@@ -107,6 +108,43 @@
uf.backend = relax_disp_obj._calc_r2eff
uf.menu_text = "&calc_r2eff"
uf.wizard_size = (900, 600)
+
+
+# The relax_disp.cluster user function.
+uf = uf_info.add_uf('relax_disp.cluster')
+uf.title = "Define clusters of spins for joint optimisation."
+uf.title_short = "Spin clustering."
+uf.add_keyarg(
+ name = "cluster_id",
+ py_type = "str",
+ desc_short = "cluster ID",
+ desc = "The cluster identification string.",
+ wiz_element_type = 'combo',
+ wiz_combo_iter = relax_disp_obj._cluster_ids
+)
+uf.add_keyarg(
+ name = "spin_id",
+ py_type = "str",
+ desc_short = "spin ID string",
+ desc = "The spin identifier string for the spin or group of spins to add
to the cluster.",
+ wiz_element_type = 'combo',
+ wiz_combo_iter = get_spin_ids,
+ can_be_none = True
+)
+# Description.
+uf.desc.append(Desc_container())
+uf.desc[-1].add_paragraph("In a relaxation dispersion analysis, the
parameters of the model of dispersion can either be optimised for each spin
system separately or a number of spins can be grouped or clustered and the
dispersion model parameters optimised for all spins in the cluster together.
Clusters are identified by unique ID strings. Any spins not within a cluster
will be optimised separately and individually.")
+uf.desc[-1].add_paragraph("If the cluster ID string already exists, the
spins will be added to that pre-existing cluster. If no spin ID is given,
then all spins will be added to the cluster.")
+uf.desc[-1].add_paragraph("The special cluster ID string 'free spins' is
reserved for the pool of non-clustered spins. This can be used to remove a
spin system from an already existing cluster by specifying this cluster ID
and the desired spin ID.")
+# Prompt examples.
+uf.desc.append(Desc_container("Prompt examples"))
+uf.desc[-1].add_paragraph("To add the spins ':1@N' and ':3@N' to a new
cluster called 'cluster', type one of:")
+uf.desc[-1].add_prompt("relax> relax_disp.cluster('cluster', ':1,3@N)")
+uf.desc[-1].add_prompt("relax> relax_disp.cluster(cluster_id='cluster',
spin_id=':1,3@N)")
+uf.backend = relax_disp_obj._cluster
+uf.menu_text = "c&luster"
+uf.wizard_height_desc = 500
+uf.wizard_size = (800, 600)
# The relax_disp.cpmg_delayT user function.
r19469 - in /branches/relax_disp: specific_analyses/relax_disp.py user_functions/relax_disp.py