def PlotFracNativeContact(DataDir, Prefix = None):
print 'PLOTTING FRACTION OF NATIVE CONTACTS'
print '---------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.cm as cmap
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 5*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_title(p, fontsize = 8)
# extract data
print 'Extracting native contact fraction for ', p
hasFracNativeContact, FracNativeContact = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'fracnativecontacts')
if not hasFracNativeContact:
print 'Contact maps for %s not found' % p
ax.set_xticks([])
ax.set_yticks([])
ax.set_xticklabels([])
ax.set_yticklabels([])
continue
frac, hist, err = FracNativeContact
# plot
ax.errorbar(frac, hist, yerr = err, lw = 2, markersize = 4)
ax.set_xlim([0, 1])
if Prefix is None: Prefix = 'fracnativecontact'
figname = os.path.join(OutDir, Prefix+'.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
def PlotContactOrder(DataDir, Prefix = None):
print 'PLOTTING CONTACT ORDERS'
print '---------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.cm as cmap
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 4*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_title(p, fontsize = 8)
# extract data
print 'Extracting native contact fraction for ', p
hasCO, CO = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'co')
if not hasCO:
print 'Contact orders for %s not found' % p
ax.set_xticks([])
ax.set_yticks([])
ax.set_xticklabels([])
ax.set_yticklabels([])
continue
nativeCO, TrajCOHist = CO
TrajCO, hist, err = TrajCOHist
AvgCO = np.sum(TrajCO * hist) / np.sum(hist)
# plot
ax.errorbar(TrajCO, hist, yerr = err, lw = 2, markersize = 4)
ax.axvline(nativeCO, ls = '-', lw = 2, color = 'red')
ax.axvline(AvgCO, ls = '-', lw = 2, color = 'blue')
figname = os.path.join(OutDir, 'co.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
def PlotPanel(DataDir, NativeDir = None, Prefix = None):
print 'PLOTTING PANELS'
print '---------------'
import matplotlib ; matplotlib.use('Agg')
import vis
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
NativePdbs = []
ClustPdbs = []
Labels = []
for i, p in enumerate(pset):
print 'Aligning ', p
# extract native pdb
nativepdb = utils.parseNative(p, MasterDir = NativeDir)
NativePdbs.append(nativepdb)
# see if cluster pdb is present
hasclustpdb, clustpdb = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'clustpdb')
ClustPdbs.append(clustpdb)
# extract rmsds for labels
if not hasclustpdb: Labels.append(None)
else:
hasrmsdhist, rmsdhist = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'rmsd')
if hasrmsdhist:
rmsdavg = np.sum(rmsdhist[0] * rmsdhist[1]) / np.sum(rmsdhist[1])
label = p + ' ' + r'$RMSD = %1.1f \AA$' % rmsdavg
Labels.append(label)
else: Labels.append(None)
# plot panel
print 'Rendering...'
vis.Panel(NativePdbs, ClustPdbs, layout['NRows'], layout['NCols'], Labels, OutDir = OutDir, PanelPrefix = Prefix)
return
def PlotFoldCurve(DataDir, Prefix = None):
print 'PLOTTING FOLDING CURVES'
print '------------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.cm as cm
pset_type = sys.argv[1]
OutDir = os.path.abspath(sys.argv[2])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 4*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_title(p)
# extract data
print 'Extracting folding curves for ', p
hasFoldCurve, FoldCurvePickle = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'foldcurve')
if (not hasFoldCurve) or (not os.path.isfile(FoldCurvePickle)):
print 'Folding curve for %s not found' % p
continue
with open(FoldCurvePickle, 'r') as of:
Temps, Frac, Err = pickle.load(of)
TFold = Temps[np.argmin(abs(Frac-0.5))]
# plot
ax.errorbar(Temps, Frac, yerr = Err, color = 'red', marker = 'o', lw = 3, markersize = 4)
ax.axvline(TFold, color = 'black', ls = '--', lw = 2)
if Prefix is None: Prefix = 'foldcurve'
figname = os.path.join(OutDir, Prefix+'.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
def PlotRamaProb(DataDir, Prefix = None):
print 'PLOTTING RAMACHANDRAN MAPS'
print '--------------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.cm as cm
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 4*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_xlim([-190, 190])
ax.set_ylim([-190, 190])
ax.set_title(p)
# extract data
print 'Extracting Ramachandran maps for ', p
hasRamaProb, RamaPickle = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'ramaprob')
hasRamaErr, RamaErr = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'ramaerr')
if not (hasRamaProb and hasRamaErr):
print 'Rama plot not found for %s not found' % p
ax.set_xticks([])
ax.set_yticks([])
ax.set_xticklabels([])
ax.set_yticklabels([])
continue
PhiNative, PsiNative, Err = RamaErr
with open(RamaPickle, 'r') as of: data = pickle.load(of)
PhiClust, PsiClust, RamaHist = data['Generic']
(PhiCenters, PsiCenters), h, err = RamaHist
RamaPmf = - np.log(h)
# convert all angles to degrees
PhiNative *= (180. / np.pi)
PsiNative *= (180. / np.pi)
PhiCenters *= (180. / np.pi)
PsiCenters *= (180. / np.pi)
# trim pmf
RamaPmf = RamaPmf.clip(min = -5, max = 5) # restrict within -5 kT and 5 kT
# plot
im = ax.imshow(np.transpose(RamaPmf), origin = 'lower', aspect = 'auto', interpolation = 'gaussian', cmap = cm.Blues,
extent = [PhiCenters.min(), PhiCenters.max(), PsiCenters.min(), PsiCenters.max()])
ax.scatter(PhiNative, PsiNative, s = 100, c = 'red', marker = 'o', edgecolor = 'k', lw = 4)
ax.axhline(0., color = 'black', lw = 2)
ax.axvline(0., color = 'black', lw = 2)
if Prefix is None: Prefix = 'ramaprob'
figname = os.path.join(OutDir, Prefix+'.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
def PlotContactMap(DataDir, Prefix = None):
print 'PLOTTING CONTACT MAPS'
print '---------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
import matplotlib.cm as cmap
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 4*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_title(p, fontsize = 8)
# extract data
print 'Extracting contact maps for ', p
hasCM_Traj, d1 = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'contactmap_traj')
hasCM_topclust, d2 = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'contactmap_topclust')
if not (hasCM_Traj and hasCM_topclust):
print 'Contact maps for %s not found' % p
ax.set_xticks([])
ax.set_yticks([])
ax.set_xticklabels([])
ax.set_yticklabels([])
continue
NativeCM_, TrajCM = d1
NativeCM, TopClustCM = d2
NRes = NativeCM.shape[0]
inds_Native = np.nonzero(NativeCM)
inds_TopClust = np.nonzero(TopClustCM)
# plot
im = ax.imshow(TrajCM, cmap = cmap.Blues, origin = 'lower', aspect = 'auto', interpolation = 'gaussian', extent = [0, NRes-1, 0, NRes-1], alpha = 0.8)
ax.scatter(inds_Native[0], inds_Native[1], s = 100, c = 'blue', edgecolor = 'black', lw = 3)
ax.scatter(inds_TopClust[1], inds_TopClust[0], s = 100, c = 'red', edgecolor = 'black', lw = 3)
ax.plot(range(NRes+1), 'k--', lw = 2)
ax.set_xticks(range(NRes))
ax.set_yticks(range(NRes))
ax.set_xlim([0, NRes-1])
ax.set_ylim([0, NRes-1])
ax.grid('on')
if Prefix is None: Prefix = 'contactmap'
figname = os.path.join(OutDir, Prefix+'.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
def PlotPhiPsiErr(DataDir, Prefix = None):
print 'PLOTTING RAMACHANDRAN ERRORS'
print '----------------------------'
import matplotlib; matplotlib.use('Agg')
import matplotlib.pyplot as plt
pset_type = sys.argv[2]
OutDir = os.path.abspath(sys.argv[3])
layout = utils.getPanelLayout(pset_type)
pset = layout['pset']
fig = plt.figure(figsize = (5*layout['NCols'], 4*layout['NRows']), facecolor = 'w', edgecolor = 'w')
for i, p in enumerate(pset):
ax = fig.add_subplot(layout['NRows'], layout['NCols'], i+1)
ax.set_xlim([-190, 190])
ax.set_ylim([-190, 190])
ax.set_title(p, fontsize = 8)
# extract data
print 'Extracting Ramachandran errors for ', p
hasRamaErr, RamaErr = utils.hasCGData(Prefix = 'prot_'+p, OutDir = os.path.join(OutDir, p, DataDir), Key = 'ramaerr')
if not hasRamaErr:
print 'RamaErrors for %s not found' % p
ax.set_xticks([])
ax.set_yticks([])
ax.set_xticklabels([])
ax.set_yticklabels([])
continue
PhiNative, PsiNative, Err = RamaErr
# scale the errors
PhiNative *= (180. / np.pi)
PsiNative *= (180. / np.pi)
Err *= (180. / np.pi)
Err = (Err - Err.min()) / (Err.max() - Err.min())
# plot
plt.scatter(PhiNative, PsiNative, s = 500*Err, c = 'red', alpha=0.4, edgecolors="grey", linewidth=2)
ax.axhline(0., color = 'black', lw = 2)
ax.axvline(0., color = 'black', lw = 2)
if Prefix is None: Prefix = 'ramaerr'
figname = os.path.join(OutDir, Prefix+'.png')
fig.tight_layout()
plt.savefig(figname, bbox_inches='tight')
return
