def test_model(self):
"""PDBDope test final model"""
from Biskit import PDBModel
if self.local:
print '\nData added to info record of model (key -- value):'
for k in self.d.m.info.keys():
print '%s -- %s' % (k, self.d.m.info[k])
print '\nAdded atom profiles:'
print self.M.atoms
print '\nAdded residue profiles:'
print self.M.residues
## check that nothing has changed
print '\nChecking that models are unchanged by doping ...'
m_ref = PDBModel(self.f)
m_ref = m_ref.compress(m_ref.maskProtein())
for k in m_ref.atoms.keys():
#ref = [ m_ref.atoms[i][k] for i in m_ref.atomRange() ]
#mod = [ self.M.atoms[i][k] for i in self.M.atomRange() ]
self.assert_(N.all(m_ref[k] == self.M[k]))
## display in Pymol
if self.local:
print "Starting PyMol..."
from Biskit.Pymoler import Pymoler
pm = Pymoler()
pm.addPdb(self.M, 'm')
pm.colorAtoms('m', N0.clip(self.M.profile('relAS'), 0.0, 100.0))
pm.show()
def logConfidence( x, R, clip=0 ):
"""
Estimate the probability of x NOT beeing a random observation from a
lognormal distribution that is described by a set of random values.
@param x: observed value
@type x: float
@param R: sample of random values
@type R: [float]
@param clip: clip zeros at this value 0->don't clip (default: 0)
@type clip: float
@return: confidence that x is not random, median of random distr.
@rtype: (float, float)
"""
if clip and 0 in R:
R = N0.clip( R, clip, max( R ) )
if clip and x == 0:
x = clip
## remove 0 instead of clipping
R = N0.compress( R, R )
if x == 0:
return 0, 0
## get mean and stdv of log-transformed random sample
alpha = N0.average( N0.log( R ) )
n = len( R )
beta = N0.sqrt(N0.sum(N0.power(N0.log( R ) - alpha, 2)) / (n - 1.))
return logArea( x, alpha, beta ), logMedian( alpha )
def test_conservation(self):
"""PDBDope.addConservation (Hmmer) test"""
if self.local: print "Adding conservation data...",
self.d.addConservation()
if self.local: print 'Done.'
## display in Pymol
if self.local:
print "Starting PyMol..."
from Biskit.Pymoler import Pymoler
pm = Pymoler()
pm.addPdb(self.M, 'm')
pm.colorAtoms('m', N0.clip(self.M.profile('cons_ent'), 0.0, 100.0))
pm.show()
def fractionNativeSurface(self, cont, contRef):
"""
fraction of atoms/residues that are involved in B{any} contacts
in both complexes.
@param cont: contact matrix
@type cont: matrix
@param contRef: reference contact matrix
@type contRef: matrix
@return: (fractRec, fractLig), fraction of atoms/residues that
are involved in any contacts in both complexes
@rtype: (float, float)
"""
lig, ligRef = N0.clip(N0.sum(cont), 0,
1), N0.clip(N0.sum(contRef), 0, 1)
rec = N0.clip(N0.sum(cont, 1), 0, 1)
recRef = N0.clip(N0.sum(contRef, 1), 0, 1)
fLig = N0.sum(N0.logical_and(lig, ligRef)) * 1. / N0.sum(ligRef)
fRec = N0.sum(N0.logical_and(rec, recRef)) * 1. / N0.sum(recRef)
return (fRec, fLig)
def relExposure(model, absSurf, key='AS', clip=1):
"""
Calculate how exposed an atom is relative to the same
atom in a GLY-XXX-GLY tripeptide, an approximation of
the unfolded state.
@param absSurf: Absolute MS OR AS values
@type absSurf: [float]
@param key: MS or AS
@type key: MS|AS
@param clip: clip values above 100% (default: 1)
@type clip: 1|0
@return: rel - list of relative accessible surfaces
@rtype: [float]
"""
if not key == 'MS' and not key == 'AS':
raise Exception,\
'Incorrect key for relative exposiure: %s '%key
rel = []
i = 0
## loop over chains
for j in range(model.lenChains()):
c = model.takeChains([j])
k = 0
cIdx = c.resIndex()
## and loop over atoms in chain
for a in c.atoms.iterDicts():
## N-terminal residue
if k < cIdx[1]:
rel = __Nter(a, rel, absSurf, key, i)
## C-terminal residue
if k >= cIdx[-1]:
rel = __Cter(a, rel, absSurf, key, i)
## everything but N- and C termini
if not k < cIdx[1] and not k >= cIdx[-1]:
rel = __bulk(a, rel, absSurf, key, i)
i += 1
k += 1
if clip:
return N0.clip(N0.array(rel), 0.0, 100.0)
else:
return N0.array(rel)
def logConfidence(x, R, clip=1e-32):
"""
Estimate the probability of x NOT beeing a random observation from a
lognormal distribution that is described by a set of random values.
The exact solution to this problem is in L{Biskit.Statistics.lognormal}.
@param x: observed value
@type x: float
@param R: sample of random values; 0 -> don't clip (default: 1e-32)
@type R: [float]
@param clip: clip zeros at this value
@type clip: float
@return: confidence that x is not random, mean of random distrib.
@rtype: (float, float)
"""
if clip and 0 in R:
R = N0.clip(R, clip, max(R))
## get mean and stdv of log-transformed random sample
mean = N0.average(N0.log(R))
n = len(R)
stdv = N0.sqrt(N0.sum(N0.power(N0.log(R) - mean, 2)) / (n - 1.))
## create dense lognormal distribution representing the random sample
stop = max(R) * 50.0
step = stop / 100000
start = step / 10.0
X = [(v, p_lognormal(v, mean, stdv)) for v in N0.arange(start, stop, step)]
## analyse distribution
d = Density(X)
return d.findConfidenceInterval(x * 1.0)[0], d.average()
def calc_membership_matrix(self, d2):
## remove 0s (if a cluster center is exactly on one item)
d2 = N0.clip( d2, N0.power(1e200, 1-self.w), 1e300 )
q = N0.power(d2, 1. / (1. - self.w))
return q / N0.sum(q)