def histogram(direction, var, bins, nsector, normed=False, blowto=False):
if len(var) != len(direction):
raise ValueError("var and direction must have same length")
angle = 360.0 / nsector
dir_bins = np.arange(-angle / 2, 360.0 + angle, angle, dtype=np.float)
dir_edges = dir_bins.tolist()
dir_edges.pop(-1)
dir_edges[0] = dir_edges.pop(-1)
dir_bins[0] = 0.0
var_bins = bins.tolist()
var_bins.append(np.inf)
if blowto:
direction = direction + 180.0
direction[direction >= 360.0] = direction[direction >= 360.0] - 360
table = histogram2d(x=var,
y=direction,
bins=[var_bins, dir_bins],
normed=False)[0]
# add the last value to the first to have the table of North winds
table[:, 0] = table[:, 0] + table[:, -1]
# and remove the last col
table = table[:, :-1]
if normed:
table = table * 100 / table.sum()
return dir_edges, var_bins, table
def histogram(direction, var, bins, nsector, normed=False, blowto=False):
"""
Returns an array where, for each sector of wind
(centred on the north), we have the number of time the wind comes with a
particular var (speed, polluant concentration, ...).
Parameters
----------
direction : 1D array
directions the wind blows from, North centred
var : 1D array
values of the variable to compute. Typically the wind speeds
bins : list
list of var category against we're going to compute the table
nsector : integer
number of sectors
Other Parameters
----------------
normed : boolean, default False
The resulting table is normed in percent or not.
blowto : boolean, default False
Normaly a windrose is computed with directions as wind blows from. If
true, the table will be reversed (usefull for pollutantrose)
"""
if len(var) != len(direction):
raise (ValueError("var and direction must have same length"))
angle = 360. / nsector
dir_bins = np.arange(-angle / 2, 360. + angle, angle, dtype=np.float)
dir_edges = dir_bins.tolist()
dir_edges.pop(-1)
dir_edges[0] = dir_edges.pop(-1)
dir_bins[0] = 0.
var_bins = bins.tolist()
var_bins.append(np.inf)
if blowto:
direction = direction + 180.
direction[direction >= 360.] = direction[direction >= 360.] - 360
table = histogram2d(x=var,
y=direction,
bins=[var_bins, dir_bins],
normed=False)[0]
# add the last value to the first to have the table of North winds
table[:, 0] = table[:, 0] + table[:, -1]
# and remove the last col
table = table[:, :-1]
if normed:
table = table * 100 / table.sum()
return dir_edges, var_bins, table
def histogram(dir, var, bins, nsector, normed=False, blowto=False):
"""
Returns an array where, for each sector of wind
(centred on the north), we have the number of time the wind comes with a
particular var (speed, polluant concentration, ...).
* dir : 1D array - directions the wind blows from, North centred
* var : 1D array - values of the variable to compute. Typically the wind
speeds
* bins : list - list of var category against we're going to compute the table
* nsector : integer - number of sectors
* normed : boolean - The resulting table is normed in percent or not.
* blowto : boolean - Normaly a windrose is computed with directions
as wind blows from. If true, the table will be reversed (usefull for
pollutantrose)
"""
if len(var) != len(dir):
raise ValueError("var and dir must have same length")
angle = old_div(360., nsector)
dir_bins = np.arange(old_div(-angle, 2),
360. + angle,
angle,
dtype=np.float)
dir_edges = dir_bins.tolist()
dir_edges.pop(-1)
dir_edges[0] = dir_edges.pop(-1)
dir_bins[0] = 0.
var_bins = bins.tolist()
var_bins.append(np.inf)
if blowto:
dir = dir + 180.
dir[dir >= 360.] = dir[dir >= 360.] - 360
table = histogram2d(x=var, y=dir, bins=[var_bins, dir_bins],
normed=False)[0]
# add the last value to the first to have the table of North winds
table[:, 0] = table[:, 0] + table[:, -1]
# and remove the last col
table = table[:, :-1]
# count the number of calms
numcalms = np.average(table[0, :])
table[0, :] = numcalms
if normed:
table = table * 100 / table.sum()
return dir_edges, var_bins, table
def histogram(direction, var, bins, nsector, normed=False,
blowto=False):
"""
Returns an array where, for each sector of wind
(centred on the north), we have the number of time the wind comes with a
particular var (speed, polluant concentration, ...).
* direction : 1D array - directions the wind blows from, North centred
* var : 1D array - values of the variable to compute. Typically the wind
speeds
* bins : list - list of var category against we're going to compute the table
* nsector : integer - number of sectors
* normed : boolean - The resulting table is normed in percent or not.
* blowto : boolean - Normaly a windrose is computed with directions
as wind blows from. If true, the table will be reversed (usefull for
pollutantrose)
"""
if len(var) != len(direction):
raise (ValueError("var and direction must have same length"))
angle = 360. / nsector
dir_bins = np.arange(-angle / 2, 360. + angle, angle,
dtype=np.float)
dir_edges = dir_bins.tolist()
dir_edges.pop(-1)
dir_edges[0] = dir_edges.pop(-1)
dir_bins[0] = 0.
var_bins = bins.tolist()
# print var_bins
# var_bins.append(np.inf)
if blowto:
direction = direction + 180.
direction[direction >= 360.] = direction[
direction >= 360.] - 360
table = histogram2d(x=var, y=direction, bins=[var_bins, dir_bins],
normed=False)[0]
# add the last value to the first to have the table of North winds
table[:, 0] = table[:, 0] + table[:, -1]
# and remove the last col
table = table[:, :-1]
if normed:
table = table * 100 / table.sum()
return dir_edges, var_bins, table
def main():
cvs_file_abs_name_gz = os.path.join(cingDirData, 'PluginCode', 'Whatif',
cvs_file_abs_name + '.gz')
gunzip(cvs_file_abs_name_gz)
reader = csv.reader(open(cvs_file_abs_name, "rb"), quoting=csv.QUOTE_NONE)
valuesBySsAndResType = {}
histJaninBySsAndResType = {}
histJaninBySsAndCombinedResType = {}
# histByCombinedSsAndResType = {}
histJaninCtupleBySsAndResType = {}
valuesByEntrySsAndResType = {}
hrange = (xRange, yRange)
# rowCount = 0
for row in reader:
# rowCount += 1
# 7a3h,A,VAL , 5,H, -62.8, -52.8
# 7a3h,A,VAL , 6,H, -71.2, -33.6
# 7a3h,A,GLU , 7,H, -63.5, -41.6
(entryId, _chainId, resType, _resNum, ssType, chi1, chi2,
_max_bfactor) = row
ssType = to3StateDssp(ssType)[0]
resType = resType.strip()
chi1 = chi1.strip()
chi2 = chi2.strip()
chi1 = floatParse(chi1)
chi2 = floatParse(chi2)
if isNaN(chi1) or isNaN(chi2):
continue
if not inRange(chi1):
nTerror("chi1 not in range for row: %s" % repr(row))
return
if not inRange(chi2):
nTerror("chi2 not in range for row: %s" % repr(row))
return
if not common20AADict.has_key(resType):
nTdebug("Residue not in common 20 for row: %s" % repr(row))
# rowCount -= 1
continue
appendDeepByKeys(valuesBySsAndResType, chi1, ssType, resType, 'chi1')
appendDeepByKeys(valuesByEntrySsAndResType, chi1, entryId, ssType,
resType, 'chi1')
appendDeepByKeys(valuesBySsAndResType, chi2, ssType, resType, 'chi2')
appendDeepByKeys(valuesByEntrySsAndResType, chi2, entryId, ssType,
resType, 'chi2')
# nTdebug('resType,ssType,chi1: %4s %1s %s' % (resType,ssType,floatFormat(chi1, "%6.1f")))
# nTdebug('resType,ssType,chi2: %4s %1s %s' % (resType,ssType,floatFormat(chi2, "%6.1f")))
del (reader) # closes the file handles
os.unlink(cvs_file_abs_name)
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
chi1 = valuesBySsAndResType[ssType][resType]['chi1']
chi2 = valuesBySsAndResType[ssType][resType]['chi2']
if chi1 and chi2:
hist2d, _xedges, _yedges = histogram2d(chi2,
chi1,
bins=binCount,
range=hrange)
setDeepByKeys(histJaninBySsAndResType, hist2d, ssType, resType)
cTuple = getEnsembleAverageAndSigmaHis(hist2d)
(c_av, c_sd, hisMin, hisMax) = cTuple
cTuple += tuple([str([ssType, resType])
]) # append the hash keys as a way of id.
nTdebug(
"For ssType %s residue type %s found (av/sd/min/max) %8.0f %8.0f %8.0f %8.0f"
% (ssType, resType, c_av, c_sd, hisMin, hisMax))
if c_sd == None:
nTdebug(
'Failed to get c_sd when testing not all residues are present in smaller sets.'
)
continue
if c_sd == 0.:
nTdebug(
'Got zero c_sd, ignoring histogram. This should only occur in smaller sets. Not setting values.'
)
continue
setDeepByKeys(histJaninCtupleBySsAndResType, cTuple, ssType,
resType)
for ssType in valuesBySsAndResType.keys():
chi1 = []
chi2 = []
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
chi1 += valuesBySsAndResType[ssType][resType]['chi1']
chi2 += valuesBySsAndResType[ssType][resType]['chi2']
if chi1 and chi2:
hist2d, _xedges, _yedges = histogram2d(
chi2, # Note that the x is the chi2 for some stupid reason,
chi1, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, Cav, Csd )
setDeepByKeys(histJaninBySsAndCombinedResType, hist2d, ssType)
# Throws a verbose error message on python 2.6.3 as per issue http://code.google.com/p/cing/issues/detail?id=211
# Using Pickle instead
# dbase = shelve.open( dbase_file_abs_name )
# dbase.close()
if os.path.exists(dbase_file_abs_name):
os.unlink(dbase_file_abs_name)
output = open(dbase_file_abs_name, 'wb')
dbase = {}
dbase['histJaninBySsAndCombinedResType'] = histJaninBySsAndCombinedResType
dbase['histJaninBySsAndResType'] = histJaninBySsAndResType
dbase['histJaninCtupleBySsAndResType'] = histJaninCtupleBySsAndResType
# histJaninCtupleBySsAndResType
cPickle.dump(dbase, output, 2)
output.close()
def main():
cvs_file_abs_name_gz = os.path.join(cingDirData, 'PluginCode', 'Whatif', cvs_file_abs_name + '.gz')
gunzip(cvs_file_abs_name_gz)
reader = csv.reader(open(cvs_file_abs_name, "rb"), quoting=csv.QUOTE_NONE)
valuesBySsAndResType = {}
histJaninBySsAndResType = {}
histJaninBySsAndCombinedResType = {}
# histByCombinedSsAndResType = {}
histJaninCtupleBySsAndResType = {}
valuesByEntrySsAndResType = {}
hrange = (xRange, yRange)
# rowCount = 0
for row in reader:
# rowCount += 1
# 7a3h,A,VAL , 5,H, -62.8, -52.8
# 7a3h,A,VAL , 6,H, -71.2, -33.6
# 7a3h,A,GLU , 7,H, -63.5, -41.6
(entryId, _chainId, resType, _resNum, ssType, chi1, chi2, _max_bfactor) = row
ssType = to3StateDssp(ssType)[0]
resType = resType.strip()
chi1 = chi1.strip()
chi2 = chi2.strip()
chi1 = floatParse(chi1)
chi2 = floatParse(chi2)
if isNaN(chi1) or isNaN(chi2):
continue
if not inRange(chi1):
nTerror("chi1 not in range for row: %s" % repr(row))
return
if not inRange(chi2):
nTerror("chi2 not in range for row: %s" % repr(row))
return
if not common20AADict.has_key(resType):
nTdebug("Residue not in common 20 for row: %s" % repr(row))
# rowCount -= 1
continue
appendDeepByKeys(valuesBySsAndResType, chi1, ssType, resType, 'chi1')
appendDeepByKeys(valuesByEntrySsAndResType, chi1, entryId, ssType, resType, 'chi1')
appendDeepByKeys(valuesBySsAndResType, chi2, ssType, resType, 'chi2')
appendDeepByKeys(valuesByEntrySsAndResType, chi2, entryId, ssType, resType, 'chi2')
# nTdebug('resType,ssType,chi1: %4s %1s %s' % (resType,ssType,floatFormat(chi1, "%6.1f")))
# nTdebug('resType,ssType,chi2: %4s %1s %s' % (resType,ssType,floatFormat(chi2, "%6.1f")))
del(reader) # closes the file handles
os.unlink(cvs_file_abs_name)
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
chi1 = valuesBySsAndResType[ssType][resType]['chi1']
chi2 = valuesBySsAndResType[ssType][resType]['chi2']
if chi1 and chi2:
hist2d, _xedges, _yedges = histogram2d(
chi2, chi1,
bins=binCount,
range=hrange)
setDeepByKeys(histJaninBySsAndResType, hist2d, ssType, resType)
cTuple = getEnsembleAverageAndSigmaHis(hist2d)
(c_av, c_sd, hisMin, hisMax) = cTuple
cTuple += tuple([str([ssType, resType])]) # append the hash keys as a way of id.
nTdebug("For ssType %s residue type %s found (av/sd/min/max) %8.0f %8.0f %8.0f %8.0f" % (
ssType, resType, c_av, c_sd, hisMin, hisMax))
if c_sd == None:
nTdebug('Failed to get c_sd when testing not all residues are present in smaller sets.')
continue
if c_sd == 0.:
nTdebug('Got zero c_sd, ignoring histogram. This should only occur in smaller sets. Not setting values.')
continue
setDeepByKeys(histJaninCtupleBySsAndResType, cTuple, ssType, resType)
for ssType in valuesBySsAndResType.keys():
chi1 = []
chi2 = []
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
chi1 += valuesBySsAndResType[ssType][resType]['chi1']
chi2 += valuesBySsAndResType[ssType][resType]['chi2']
if chi1 and chi2:
hist2d, _xedges, _yedges = histogram2d(
chi2, # Note that the x is the chi2 for some stupid reason,
chi1, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, Cav, Csd )
setDeepByKeys(histJaninBySsAndCombinedResType, hist2d, ssType)
# Throws a verbose error message on python 2.6.3 as per issue https://github.com/VuisterLab/cing/issues/211
# Using Pickle instead
# dbase = shelve.open( dbase_file_abs_name )
# dbase.close()
if os.path.exists(dbase_file_abs_name):
os.unlink(dbase_file_abs_name)
output = open(dbase_file_abs_name, 'wb')
dbase = {}
dbase[ 'histJaninBySsAndCombinedResType' ] = histJaninBySsAndCombinedResType
dbase[ 'histJaninBySsAndResType' ] = histJaninBySsAndResType
dbase[ 'histJaninCtupleBySsAndResType' ] = histJaninCtupleBySsAndResType
# histJaninCtupleBySsAndResType
cPickle.dump(dbase, output, 2)
output.close()
def main():
cvs_file_abs_name_gz = cvs_file_abs_name + '.gz'
gunzip(cvs_file_abs_name_gz)
reader = csv.reader(open(cvs_file_abs_name, "rb"), quoting=csv.QUOTE_NONE)
valuesBySsAndResType = {}
histRamaBySsAndResType = {}
histRamaBySsAndCombinedResType = {}
# histByCombinedSsAndResType = {}
histRamaCtupleBySsAndResType = {}
valuesByEntrySsAndResType = {}
hrange = (xRange, yRange)
rowCount = 0
for row in reader:
rowCount += 1
# 7a3h,A,VAL , 5,H, -62.8, -52.8
# 7a3h,A,VAL , 6,H, -71.2, -33.6
# 7a3h,A,GLU , 7,H, -63.5, -41.6
(entryId, _chainId, resType, _resNum, ssType, phi, psi,
_max_bfactor) = row
ssType = to3StateDssp(ssType)[0]
resType = resType.strip()
phi = float(phi)
psi = float(psi)
if not (inRange(phi, isRange360=isRange360)
and inRange(psi, isRange360=isRange360)):
nTerror("phi and/or psi not in range for row: %s" % repr(row))
return
if not common20AADict.has_key(resType):
nTdebug("Residue not in common 20 for row: %s" % repr(row))
rowCount -= 1
continue
appendDeepByKeys(valuesBySsAndResType, phi, ssType, resType, 'phi')
appendDeepByKeys(valuesBySsAndResType, psi, ssType, resType, 'psi')
# nTdebug('resType,ssType,phi,psi: %4s %1s %8.3f %8.3f' % (resType,ssType,phi,psi))
appendDeepByKeys(valuesByEntrySsAndResType, phi, entryId, ssType,
resType, 'phi')
appendDeepByKeys(valuesByEntrySsAndResType, psi, entryId, ssType,
resType, 'psi')
del (reader) # closes the file handles
os.unlink(cvs_file_abs_name)
nTdebug('Total number of included residues including PRO/GLY: %d' %
rowCount)
# nTdebug('valuesByEntrySsAndResType:\n%s'%valuesByEntrySsAndResType)
# (cAv, cSd, _Cn) = getRescaling(valuesByEntrySsAndResType)
(cAv, cSd) = (1.0, 1.0)
nTdebug("Overall found av,sd: %r %r" % (cAv, cSd))
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
hist2d, _xedges, _yedges = histogram2d(
valuesBySsAndResType[ssType][resType]['psi'],
valuesBySsAndResType[ssType][resType]['phi'],
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, cAv, cSd )
setDeepByKeys(histRamaBySsAndResType, hist2d, ssType, resType)
# nTdebug('hist2d ssType, resType: %s %s\n%s' % (ssType, resType, hist2d))
cTuple = getEnsembleAverageAndSigmaHis(hist2d)
(c_av, c_sd, hisMin, hisMax) = cTuple
cTuple += tuple([str([ssType, resType])
]) # append the hash keys as a way of id.
nTdebug(
"For ssType %s residue type %s found (av/sd/min/max) %8.0f %8.0f %8.0f %8.0f"
% (ssType, resType, c_av, c_sd, hisMin, hisMax))
# nTdebug("xedges %s" % repr(xedges))
# sys.exit(1)
if c_sd == None:
nTdebug(
'Failed to get c_sd when testing not all residues are present in smaller sets.'
)
continue
if c_sd == 0.:
nTdebug(
'Got zero c_sd, ignoring histogram. This should only occur in smaller sets. Not setting values.'
)
continue
setDeepByKeys(histRamaCtupleBySsAndResType, cTuple, ssType,
resType)
for ssType in valuesBySsAndResType.keys():
phi = []
psi = []
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
phi += valuesBySsAndResType[ssType][resType]['phi']
psi += valuesBySsAndResType[ssType][resType]['psi']
hist2d, _xedges, _yedges = histogram2d(
psi, # Note that the x is the psi for some stupid reason,
phi, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, cAv, cSd )
setDeepByKeys(histRamaBySsAndCombinedResType, hist2d, ssType)
phi = []
psi = []
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
phi += valuesBySsAndResType[ssType][resType]['phi']
psi += valuesBySsAndResType[ssType][resType]['psi']
nTdebug('Total number of residues without PRO/GLY: %d' % len(psi))
hist2d, _xedges, _yedges = histogram2d(
psi, # Note that the x is the psi for some stupid reason,
phi, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# sumHistCombined = sum( hist2d )
# sumsumHistCombined = sum( sumHistCombined )
nTdebug('hist2d : \n%s' % hist2d)
# nTdebug('sumHistCombined : %s' % repr(sumHistCombined))
# nTdebug('sumsumHistCombined: %.0f' % sumsumHistCombined)
# hist2d = zscaleHist( hist2d, cAv, cSd )
# nTdebug('hist2d scaled : \n%s' % hist2d)
if os.path.exists(dbase_file_abs_name):
os.unlink(dbase_file_abs_name)
# dbase = shelve.open( dbase_file_abs_name )
output = open(dbase_file_abs_name, 'wb')
# dbase = {'bar':'milky'}
dbase = {}
# Pickle the list using the highest protocol available.
dbase['histRamaCombined'] = hist2d
dbase['histRamaBySsAndCombinedResType'] = histRamaBySsAndCombinedResType
dbase['histRamaBySsAndResType'] = histRamaBySsAndResType
dbase['histRamaCtupleBySsAndResType'] = histRamaCtupleBySsAndResType
# pickle.dump(dbase, output, -1)
# pickle.dump(dbase, output)
cPickle.dump(
dbase, output,
2) # Was -1 for the most recent version but this caused an issue 239
# NB 2 is the highest listed protocol too but behind the scenes cPickle will probably write something higher still.
# If the protocol parameter is omitted, protocol 0 is used.
# If protocol is specified as a negative value or HIGHEST_PROTOCOL, the highest protocol version will be used.
output.close()
def main():
cvs_file_abs_name_gz = cvs_file_abs_name + '.gz'
gunzip(cvs_file_abs_name_gz)
reader = csv.reader(open(cvs_file_abs_name, "rb"), quoting=csv.QUOTE_NONE)
valuesBySsAndResType = {}
histRamaBySsAndResType = {}
histRamaBySsAndCombinedResType = {}
# histByCombinedSsAndResType = {}
histRamaCtupleBySsAndResType = {}
valuesByEntrySsAndResType = {}
hrange = (xRange, yRange)
rowCount = 0
for row in reader:
rowCount += 1
# 7a3h,A,VAL , 5,H, -62.8, -52.8
# 7a3h,A,VAL , 6,H, -71.2, -33.6
# 7a3h,A,GLU , 7,H, -63.5, -41.6
(entryId, _chainId, resType, _resNum, ssType, phi, psi, _max_bfactor) = row
ssType = to3StateDssp(ssType)[0]
resType = resType.strip()
phi = float(phi)
psi = float(psi)
if not (inRange(phi, isRange360=isRange360) and inRange(psi, isRange360=isRange360)):
nTerror("phi and/or psi not in range for row: %s" % repr(row))
return
if not common20AADict.has_key(resType):
nTdebug("Residue not in common 20 for row: %s" % repr(row))
rowCount -= 1
continue
appendDeepByKeys(valuesBySsAndResType, phi, ssType, resType, 'phi')
appendDeepByKeys(valuesBySsAndResType, psi, ssType, resType, 'psi')
# nTdebug('resType,ssType,phi,psi: %4s %1s %8.3f %8.3f' % (resType,ssType,phi,psi))
appendDeepByKeys(valuesByEntrySsAndResType, phi, entryId, ssType, resType, 'phi')
appendDeepByKeys(valuesByEntrySsAndResType, psi, entryId, ssType, resType, 'psi')
del(reader) # closes the file handles
os.unlink(cvs_file_abs_name)
nTdebug('Total number of included residues including PRO/GLY: %d' % rowCount)
# nTdebug('valuesByEntrySsAndResType:\n%s'%valuesByEntrySsAndResType)
# (cAv, cSd, _Cn) = getRescaling(valuesByEntrySsAndResType)
(cAv, cSd) = (1.0, 1.0)
nTdebug("Overall found av,sd: %r %r" % (cAv, cSd))
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
hist2d, _xedges, _yedges = histogram2d(
valuesBySsAndResType[ssType][resType]['psi'],
valuesBySsAndResType[ssType][resType]['phi'],
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, cAv, cSd )
setDeepByKeys(histRamaBySsAndResType, hist2d, ssType, resType)
# nTdebug('hist2d ssType, resType: %s %s\n%s' % (ssType, resType, hist2d))
cTuple = getEnsembleAverageAndSigmaHis(hist2d)
(c_av, c_sd, hisMin, hisMax) = cTuple
cTuple += tuple([str([ssType, resType])]) # append the hash keys as a way of id.
nTdebug("For ssType %s residue type %s found (av/sd/min/max) %8.0f %8.0f %8.0f %8.0f" % (
ssType, resType, c_av, c_sd, hisMin, hisMax))
# nTdebug("xedges %s" % repr(xedges))
# sys.exit(1)
if c_sd == None:
nTdebug('Failed to get c_sd when testing not all residues are present in smaller sets.')
continue
if c_sd == 0.:
nTdebug('Got zero c_sd, ignoring histogram. This should only occur in smaller sets. Not setting values.')
continue
setDeepByKeys(histRamaCtupleBySsAndResType, cTuple, ssType, resType)
for ssType in valuesBySsAndResType.keys():
phi = []
psi = []
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
phi += valuesBySsAndResType[ssType][resType]['phi']
psi += valuesBySsAndResType[ssType][resType]['psi']
hist2d, _xedges, _yedges = histogram2d(
psi, # Note that the x is the psi for some stupid reason,
phi, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# hist2d = zscaleHist( hist2d, cAv, cSd )
setDeepByKeys(histRamaBySsAndCombinedResType, hist2d, ssType)
phi = []
psi = []
for ssType in valuesBySsAndResType.keys():
for resType in valuesBySsAndResType[ssType].keys():
if resType == 'PRO' or resType == 'GLY':
continue
phi += valuesBySsAndResType[ssType][resType]['phi']
psi += valuesBySsAndResType[ssType][resType]['psi']
nTdebug('Total number of residues without PRO/GLY: %d' % len(psi))
hist2d, _xedges, _yedges = histogram2d(
psi, # Note that the x is the psi for some stupid reason,
phi, # otherwise the imagery but also the [row][column] notation is screwed.
bins=binCount,
range=hrange)
# sumHistCombined = sum( hist2d )
# sumsumHistCombined = sum( sumHistCombined )
nTdebug('hist2d : \n%s' % hist2d)
# nTdebug('sumHistCombined : %s' % repr(sumHistCombined))
# nTdebug('sumsumHistCombined: %.0f' % sumsumHistCombined)
# hist2d = zscaleHist( hist2d, cAv, cSd )
# nTdebug('hist2d scaled : \n%s' % hist2d)
if os.path.exists(dbase_file_abs_name):
os.unlink(dbase_file_abs_name)
# dbase = shelve.open( dbase_file_abs_name )
output = open(dbase_file_abs_name, 'wb')
# dbase = {'bar':'milky'}
dbase = {}
# Pickle the list using the highest protocol available.
dbase[ 'histRamaCombined' ] = hist2d
dbase[ 'histRamaBySsAndCombinedResType' ] = histRamaBySsAndCombinedResType
dbase[ 'histRamaBySsAndResType' ] = histRamaBySsAndResType
dbase[ 'histRamaCtupleBySsAndResType' ] = histRamaCtupleBySsAndResType
# pickle.dump(dbase, output, -1)
# pickle.dump(dbase, output)
cPickle.dump(dbase, output, 2) # Was -1 for the most recent version but this caused an issue 239
# NB 2 is the highest listed protocol too but behind the scenes cPickle will probably write something higher still.
# If the protocol parameter is omitted, protocol 0 is used.
# If protocol is specified as a negative value or HIGHEST_PROTOCOL, the highest protocol version will be used.
output.close()