def run():
print("Running @", datetime.now())
print("Downloading data..")
collection.run()
print("Aggregating data..")
aggregation.run()
print("Analyzing data..")
analysis.run()
print("Creating plots..")
visualization.run()
print("Done!")
def run(i,j):
start = time.time()
if i == 9 and j == 9:
setcurrent = analysis.run([],[],[],None)
else:
setcurrent = analysis.run([],[],[],None,
pos_tagged = han.POS_TAGGED[i:j],
results = han.RESULTS[i:j])
end = time.time()
print("Time: %d" %(end - start))
return setcurrent
def run_commandline_program(sys_argv, current_version, latest_version):
def print_help():
print(
'There are two options for running PySoar from the commandline:\n'
'1. `python main_python` for GUI\n'
'2. `python main_pysoar [url]` - where [url] is the daily competition url'
)
def status_handle(message):
print(message)
def download_handle(new, total=None):
if total is not None:
analysis_str = 'Downloaded: %s/%s' % (new, total)
else:
analysis_str = 'Downloaded: %s' % new
print(analysis_str)
def on_success():
print('Analysis complete')
def on_failure(msg):
print('Error: %s' % msg)
def analysis_handle(new, total=None):
if total is not None:
analysis_str = 'Analyzed: %s/%s' % (new, total)
else:
analysis_str = 'Analyzed: %s' % new
print(analysis_str)
if latest_version and latest_version.lstrip('v') != current_version:
print('Latest version is %s! Current: %s' %
(latest_version, current_version))
if len(sys_argv) == 2:
if sys_argv[1] == '--help':
print_help()
else:
url = sys_argv[1]
if url_format_correct(url, status_handle):
source = get_url_source(url)
run(url,
source,
download_progress=download_handle,
analysis_progress=analysis_handle,
on_success=on_success,
on_failure=on_failure)
else:
print_help()
def hello_world(text):
#param = request.args.get('text')
param=json.loads(text)
print param
# fc=subprocess.check_output('/usr/bin/python analysis.py ' + param['text'],shell=True)
fc=analysis.run(param['text'])
# print fc
return fc #input
def hello_world(text):
#param = request.args.get('text')
param = json.loads(text)
print param
# fc=subprocess.check_output('/usr/bin/python analysis.py ' + param['text'],shell=True)
fc = analysis.run(param['text'])
# print fc
return fc #input
def compare_feature_selectors(p):
feature_selectors = [
SymmetricalUncertainty(),
Relief(),
SVM_RFE(percentage_features_to_select=p),
]
e_methods = [
ensemble_methods.Mean(data_set_feature_selectors=feature_selectors),
ensemble_methods.MeanNormalizedSum(data_set_feature_selectors=feature_selectors),
ensemble_methods.MeanWithClassifier(data_set_feature_selectors=feature_selectors, classifiers=analysis.default_classifiers),
]
fs = feature_selectors + [LassoFeatureSelector(), Random()] + e_methods
data_sets = ["colon", "arcene", "dexter", "gisette"]
# analysis.artificial(fs, jaccard_percentage=p)
analysis.run(data_sets, fs, jaccard_percentage=p)
def do_analysis():
"""Call the analysis and log meta information."""
# check what data files are available
data_dir = flask.request.form['datadir']
data_files = glob.glob(data_dir + "*")
print("found those datafiles:")
for file in data_files:
print('- ' + file)
print('perform analysis {} ...'.format(flask.request.form['analysis']))
result = {
"id": flask.request.form['id'],
"time_start_wall": f"{datetime.datetime.now():%Y-%m-%d %H:%M:%S}",
"URL": flask.request.form['url'],
"domain": flask.request.form['domain'],
"analysis": flask.request.form['analysis'],
"info": "Done",
}
# run the analysis
wall_start = time.time()
cpu_start = time.process_time()
result['result'] = analysis.run(data_dir)
wall_end = time.time()
cpu_end = time.process_time()
result['time_duration_wall'] = wall_end - wall_start
result['time_duration_cpu'] = cpu_end - cpu_start
result['time_end_wall'] = f"{datetime.datetime.now():%Y-%m-%d %H:%M:%S}"
# save results
outfilepath = flask.request.form['outfile']
print('Storing results in ' + outfilepath)
with open(outfilepath, 'w') as outfile:
json.dump(result, outfile)
# return something
return 'Finished analysis: ' + flask.request.form['id']
def analyze(args):
keys = [
'company',
]
lost_keys = get_lost_keys(args, keys)
if lost_keys:
return {
'respCode': '9999',
'respMsg': '缺少参数: %s' % ' '.join(lost_keys)
}
try:
company = args['company']
except Exception as e:
return {
'respCode': '9999',
'respMsg': '数据类型错误: %s' % str(e),
'sample_args': {
'company': 'aitai',
} # 后端传递入参都是字符, 需要检查数据类型
}
data = analysis.run(company)
res = {'respCode': '0000', 'respMsg': 'success', 'data': {'data': data}}
return res
def fig_Pcolor(outfile='fig_Pcolor.png'):
"""
Args:
outfile:
Returns:
"""
set_mplrc()
# Load f_mL
f_mL = analysis.load_f_mL()
frb_pre = analysis.get_candidates()
# Init prior
prior = associate_defs.adopted.copy()
# Plot
plt.figure(figsize=(7, 7))
# Bayesian + parse
_, model_mags, model_theta, max_PMix = analysis.run(prior, frb_pre=frb_pre)
#bins_L = np.linspace(0.01, 10., 20)
bins_L = np.linspace(-2., 1, 15)
#bins_L = np.linspace(0.1, 2., 20)
#bins_L = np.linspace(16., 25., 20) # Testing only
# High confidence
high_conf_Mstar, high_conf_ur = [], []
for frbA in frb_pre.frbA:
if np.max(frbA.candidates.P_Ox) > associate_defs.POx_secure:
# This is klunky
hg = frbA.frb.grab_host()
if 'z_spec' not in hg.redshift.keys():
continue
imax = np.argmax(frbA.candidates.P_Ox)
# u-r
if 'u-r' not in hg.derived.keys():
continue
ur = hg.derived['u-r']
Mstar = hg.derived['Mstar']
# Save
high_conf_Mstar.append(np.log10(Mstar))
high_conf_ur.append(ur)
high_conf_Mstar = np.array(high_conf_Mstar)
high_conf_ur = np.array(high_conf_ur)
# Plot m_r vs. z
df = pandas.DataFrame(dict(Mstar=high_conf_Mstar, ur=high_conf_ur))
jg = sns.jointplot(data=df, x='Mstar', y='ur')
jg.ax_marg_x.set_xlim(8, 10.5)
jg.ax_marg_y.set_ylim(0.5, 2.0)
jg.ax_joint.set_xlabel(r'$\log_{10} \, (M*/M_\odot)$')
jg.ax_joint.set_ylabel(r'$u-r$')
jg.ax_joint.minorticks_on()
jg.ax_joint.yaxis.set_major_locator(plt.MultipleLocator(0.5))
jg.ax_joint.xaxis.set_major_locator(plt.MultipleLocator(1.0))
#jg.ax_joint.yaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
# Font size
set_fontsize(jg.ax_joint, 15.)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def fig_PL(outfile='fig_PL.png'):
"""
Args:
outfile:
Returns:
"""
set_mplrc()
sns.set_theme()
sns.set_style('whitegrid')
sns.set_context('paper')
# Load f_mL
f_mL = analysis.load_f_mL()
frb_pre = analysis.get_candidates()
# Init prior
prior = associate_defs.adopted.copy()
# Plot
plt.figure(figsize=(4, 8))
gs = gridspec.GridSpec(2, 1)
# Bayesian + parse
_, model_mags, model_theta, max_PMix = analysis.run(prior, frb_pre=frb_pre)
#bins_L = np.linspace(0.01, 10., 20)
bins_L = np.linspace(-2., 1, 15)
#bins_L = np.linspace(0.1, 2., 20)
#bins_L = np.linspace(16., 25., 20) # Testing only
# High confidence
high_conf_L, high_conf_z, high_conf_mr = [], [], []
for frbA in frb_pre.frbA:
if np.max(frbA.candidates.P_Ox) > associate_defs.POx_secure:
# This is klunky
hg = frbA.frb.grab_host()
if 'z_spec' not in hg.redshift.keys():
continue
imax = np.argmax(frbA.candidates.P_Ox)
# Magnitude
m = frbA.candidates.iloc[imax][frbA.filter]
# m_r(L*)
m_r_Lstar = float(f_mL(frbA.frb.z))
# Now magnitude time
log10_Lstar = (m_r_Lstar - m) / 2.5
print('FRB, z, m, m_L*, log10_L: ', frbA.frb.frb_name, frbA.frb.z,
m, m_r_Lstar, log10_Lstar)
# Save
high_conf_L.append(log10_Lstar)
high_conf_z.append(frbA.frb.z)
high_conf_mr.append(m)
high_conf_L = np.array(high_conf_L)
high_conf_z = np.array(high_conf_z)
high_conf_mr = np.array(high_conf_mr)
# Plot m_r vs. z
ax = plt.subplot(gs[0])
sns.scatterplot(x=high_conf_z, y=high_conf_mr, ax=ax)
ax.set_xlabel(r'$z$')
ax.set_ylabel(r'$m_r$')
# Add a line
zs = np.linspace(0.02, 0.5, 100)
mrss = f_mL(zs)
ax.plot(zs, mrss, 'k--')
fsz = 15.
set_fontsize(ax, fsz)
# Plot
ax = plt.subplot(gs[1])
weights = np.ones_like(high_conf_L) / high_conf_L.size
lbl = r'$P(O_i) > ' + '{}'.format(associate_defs.POx_secure) + '$ FRBs'
ax.hist(high_conf_L,
bins=bins_L,
weights=weights,
color='b',
label=lbl,
histtype='stepfilled')
# Stats
print("Median L/L* = {}".format(np.median(10**high_conf_L)))
print("RMS log(L/L*) = {}".format(np.std(high_conf_L)))
# Label me
ax.set_xlabel(r'$\log_{10} \, (L/L*)$')
ax.set_ylabel('PDF')
ax.xaxis.set_major_locator(plt.MultipleLocator(1.))
#ax.xaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
ax.set_ylim(0., 0.3)
# Legend
legend = ax.legend(loc='upper right',
scatterpoints=1,
borderpad=0.2,
handletextpad=handletextpad,
fontsize=13.)
# Font size
set_fontsize(ax, fsz)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def fig_mag_vs_DM(outfile='fig_mag_vs_DM.png'):
"""
mini Maquart relation
Args:
outfile:
Returns:
"""
set_mplrc()
# Load f_mL
f_mL = analysis.load_f_mL()
# Init prior
prior = associate_defs.adopted.copy()
# Plot
plt.figure(figsize=(8, 5))
gs = gridspec.GridSpec(1, 1)
cm = plt.get_cmap('jet')
# Bayesian + parse
frbA_tbl, model_mags, model_theta, max_PMix = analysis.run(prior)
# Colors
N = len(frbA_tbl) + 1
plt.rcParams["axes.prop_cycle"] = plt.cycler("color",
cm(np.linspace(0, 1, N)))
# Plot -- must come after colors
ax = plt.subplot(gs[0])
# Plot em
Pmin = 0.01
unit_size = 150.
order = [0, 1, 6, 3, 2, 4, 5, 7, 8, 9, 10, 11, 12]
DMcosmic = [
frb_row.frbA.frb.DM.value - frb_row.frbA.frb.DMISM.value - 100
for _, frb_row in frbA_tbl.iterrows()
]
order = np.argsort(DMcosmic)
#for _, frb_row in frbA_tbl.iterrows():
for ss in order:
# Restrict to candiates with min
frb_row = frbA_tbl.iloc[ss]
ok_c = frb_row.frbA.candidates.P_Ox > Pmin
N_c = np.sum(ok_c)
# Scatter plot em
ax.scatter(
[frb_row.frbA.frb.DM.value - frb_row.frbA.frb.DMISM.value - 100] *
N_c,
frb_row.frbA.candidates[ok_c][frb_row.frbA.filter],
label=frb_row.frb,
edgecolor='darkslategrey',
marker='o',
s=unit_size * frb_row.frbA.candidates[ok_c].P_Ox,
)
# Fiducial relation
L_Lstar = 0.270 # Grabbed from PL
numL = int(np.round(1. / L_Lstar))
log10_L_Lstar = np.log10(L_Lstar)
std_log10_L_Lstar = 0.47 # Grabbed from fig_PL
# Load up DM_cosmic
dm_cosmic, z = igm.average_DM(1., cumul=True)
f_DMz = interp1d(z, dm_cosmic.value)
zval = np.linspace(0.02, 1., 100)
DMs = f_DMz(zval)
mLstar = f_mL(zval)
m_FRB = mLstar - 2.5 * log10_L_Lstar
ax.plot(DMs, m_FRB, 'k-', label=r'$L=L*/${}'.format(numL))
# Label me
ax.set_xlabel(
r'DM$_{\rm cosmic} \equiv$ DM$_{\rm FRB}$ - DM$_{\rm ISM}$ - 100')
ax.set_ylabel(r'$m_r$')
#ax.xaxis.set_major_locator(plt.MultipleLocator(1.))
#ax.xaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
ax.set_xlim(0., 800.)
# Legend
legend = ax.legend(loc='lower right',
scatterpoints=1,
borderpad=0.0,
handletextpad=handletextpad,
fontsize=10.)
# Font size
set_fontsize(ax, 15.)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def execfile():
channel_id = e1.get()
from analysis import run
an = run(channel_id)
an.rest()
def fig_Phalflight(outfile='fig_Phalflight.png'):
"""
Args:
outfile:
Returns:
"""
set_mplrc()
# Load f_mL
f_mL = analysis.load_f_mL()
frb_pre = analysis.get_candidates()
# Init prior
prior = associate_defs.adopted.copy()
# Plot
plt.figure(figsize=(7, 7))
# Bayesian + parse
_, model_mags, model_theta, max_PMix = analysis.run(prior, frb_pre=frb_pre)
#bins_L = np.linspace(0.01, 10., 20)
bins_L = np.linspace(-2., 1, 15)
#bins_L = np.linspace(0.1, 2., 20)
#bins_L = np.linspace(16., 25., 20) # Testing only
# High confidence
high_conf_z, high_conf_hl = [], []
for frbA in frb_pre.frbA:
if np.max(frbA.candidates.P_Ox) > associate_defs.POx_secure:
# Half light
imax = np.argmax(frbA.candidates.P_Ox)
# Save
high_conf_z.append(frbA.frb.z)
high_conf_hl.append(frbA.candidates.iloc[imax].half_light)
high_conf_z = np.array(high_conf_z)
high_conf_hl = np.array(high_conf_hl)
# Plot m_r vs. z
df = pandas.DataFrame(dict(z=high_conf_z, hl=high_conf_hl))
jg = sns.jointplot(data=df, x='z', y='hl')
jg.ax_marg_x.set_xlim(0, 0.6)
#jg.ax_marg_y.set_ylim(0.5, 2.5)
jg.ax_joint.set_xlabel(r'$z$')
jg.ax_joint.set_ylabel(chalf + ' (arcsec)')
#jg.ax_joint.yaxis.set_major_locator(plt.MultipleLocator(0.5))
#jg.ax_joint.yaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
# Font size
#set_fontsize(ax, 13.)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def combinations():
fs = [
SymmetricalUncertainty(),
Relief(),
SVM_RFE(),
LassoFeatureSelector(),
]
e_methods = [
ensemble_methods.Mean(data_set_feature_selectors=fs),
ensemble_methods.Influence(data_set_feature_selectors=fs),
ensemble_methods.MeanNormalizedSum(data_set_feature_selectors=fs),
ensemble_methods.MeanWithClassifier(
data_set_feature_selectors=fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.InfluenceWithClassifier(
data_set_feature_selectors=fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.MeanNormWithClassifier(
data_set_feature_selectors=fs,
classifiers=analysis.default_classifiers
),
]
for comb in itertools.combinations(list(range(4)), 3):
comb_fs = [fs[i] for i in comb]
e_methods.extend([
ensemble_methods.Mean(data_set_feature_selectors=comb_fs),
ensemble_methods.Influence(data_set_feature_selectors=comb_fs),
ensemble_methods.MeanNormalizedSum(data_set_feature_selectors=comb_fs),
ensemble_methods.MeanWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.InfluenceWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.MeanNormWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
])
for comb in itertools.combinations(list(range(4)), 2):
comb_fs = [fs[i] for i in comb]
e_methods.extend([
ensemble_methods.Mean(data_set_feature_selectors=comb_fs),
ensemble_methods.Influence(data_set_feature_selectors=comb_fs),
ensemble_methods.MeanNormalizedSum(data_set_feature_selectors=comb_fs),
ensemble_methods.MeanWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.InfluenceWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
ensemble_methods.MeanNormWithClassifier(
data_set_feature_selectors=comb_fs,
classifiers=analysis.default_classifiers
),
])
data_sets = ["artificial", "colon", "arcene", "dexter", "gisette"]
analysis.run(data_sets, fs + e_methods, prefix="combinations")
#
# START
#
answer = "n" # input("Start Generate Data? (y/n)")
if (answer.lower() == "y" or 'generate' in sys.argv):
import generate
if ('test' in sys.argv):
generate.test()
else:
generate.run()
answer = "n" # input("Start Prepare Data? (y/n)")
if (answer.lower() == "y" or 'prepare' in sys.argv):
import prepare
if ('analyze' in sys.argv):
prepare.analyze()
else:
prepare.run()
answer = "n" # input("Start Training? (y/n)")
if (answer.lower() == "y" or 'train' in sys.argv):
import train
train.run()
answer = "n" #input("Start Visualize? (y/n)")
if (answer.lower() == "y" or 'visualize' in sys.argv):
import analysis
analysis.run()
def main(analysis_type, command_variables, threshold):
pdb_dir = fopen(
"D:\\xampp\\htdocs\\python\\SSP\\absolute_path_of_pdb_files")[0].strip(
) ## EDIT
dssp_dir = None
#dssp_dir = "..\\pdb_files\\dsspout\\"
#pdb_dir = "..\\pdb_files\\"
x = Workbook()
x.add_sheet("X-RAY")
x.add_sheet("X-RAY 2")
x.add_sheet("X-RAY 3")
x.add_sheet("NMR averaged model")
x.add_sheet("NMR multiple models 1")
x.add_sheet("NMR multiple models 2")
x.add_sheet("Other")
date_ = time.strftime("%d%b%Y_%H%M%S")
write_path = "D:\\xampp\\htdocs\\python\\Analyses\\" + str(
int(random.random() * 100000000)) + "_" + date_ + ".xls"
inA = 0
if len(command_variables) == 2:
if command_variables[1] == '-A':
date_ = time.strftime("%d_%b_%Y")
write_path = "D:\\xampp\\htdocs\\python\\Analyses\\disulfide_analysis_" + date_ + ".xls"
command_variables.pop(1)
inA = 1
## EDIT
#x.save(write_path)
#x.save("..\\Analyses\\test.xls")
# file pdb1a5n is missing, possibly because 1a5n does not contain ssbonds (correct)
#"pdb1tdy.ent","pdb1a5n.ent","pdb1cb6.ent","pdb2ac5.ent"
#for file in [pdb_dir+name for name in ["pdb1kdg.ent", "pdb1kdk.ent", "pdb1kdm.ent", "pdb1kdq.ent", "pdb1kdu.ent", "pdb1kdv.ent", "pdb1kdy.ent", "pdb1kdz.ent", "pdb1ke1.ent", "pdb1ke2.ent", "pdb1keb.ent", "pdb1keg.ent", "pdb1kek.ent", "pdb1kel.ent", "pdb1kem.ent", "pdb1ken.ent", "pdb1keo.ent", "pdb1kex.ent", "pdb1kf2.ent", "pdb1kf3.ent", "pdb1kf4.ent", "pdb1kf5.ent", "pdb1kf7.ent"]]:
# run(file,x)
#try:
# open("..\\Analyses\\basic_analysis_"+time.strftime("%a_%d_%b_%Y")+".xls","r")
#except IOError:
#for file in os.listdir(pdb_dir):
#for file in ["pdb1axi.ent"]:
# run(pdb_dir+file,x)
#n = len(os.listdir(pdb_dir))
erfile = open("D:\\xampp\\htdocs\\python\\Errors\\error_" + analysis_type +
"_" + date_ + ".log", "w") ## EDIT
runlist = None
#check command parameters
if len(command_variables) < 2:
#command_variables.pop(0)
#run_list = []
run_list = os.listdir(pdb_dir)
run_list.remove("dsspout")
elif len(command_variables) >= 2:
command_variables.pop(0)
run_list = []
for i in command_variables:
if i.find("\\") == -1:
i = ".\\" + i
if i.lower().find('.ent') != -1 or i.lower().find('.pdb') != -1:
if not os.path.isfile(i):
erfile.write("The file " + i + " was not found.")
sys.exit(1)
run_list.append(i)
pdb_dir = ""
dssp_dir = pdb_dir + os.path.dirname(i) + "\\dsspout"
else:
run_list = []
run_list = command_variables[:]
print "Checking DSSP files... "
if dssp_dir == None:
dssp_dir = pdb_dir + os.path.dirname(run_list[0]) + "\\dsspout"
if not os.path.isdir(dssp_dir):
dssplist = []
else:
dssplist = os.listdir(dssp_dir)
newlist = {}
for j in dssplist:
newlist[j.split(".")[0].split("_")[0]] = 1
print len(newlist)
print len(run_list)
for i in run_list:
found = 0
print i[:-3].split("\\")[-1].rstrip(".")
if newlist.has_key(i[:-3].split("\\")[-1].rstrip(".")):
found = 1
if found == 0:
if not os.path.isdir(pdb_dir + os.path.dirname(i) + "\\dsspout"):
os.mkdir(pdb_dir + os.path.dirname(i) + "\\dsspout")
outpath = pdb_dir + os.path.dirname(i) + "\\dsspout" + "\\"
erfile.write("python create_dssp.py " + pdb_dir +
os.path.dirname(i) + "\\" + os.path.basename(i) +
" " + outpath)
os.system("python D:\\xampp\htdocs\python\SSP\create_dssp.py " +
pdb_dir + os.path.dirname(i) + "\\" +
os.path.basename(i) + " " + outpath)
print "CREATE "
print "Done"
# print "Checking DSSP files... "
# for i in run_list:
# dssp_dir = pdb_dir+os.path.dirname(i)+"\\dsspout"
# if not os.path.isdir(dssp_dir):
# dssplist = []
# else:
# dssplist = os.listdir(dssp_dir)
# found = 0
# for j in dssplist:
# if j.find(i[:-3].split("\\")[-1]) != -1:
# found = 1
# break;
# if found == 0:
# if not os.path.isdir(pdb_dir+os.path.dirname(i)+"\\dsspout"):
# os.mkdir(pdb_dir+os.path.dirname(i)+"\\dsspout")
# os.system("D:\\xampp\\htdocs\\python\\SSP\\dsspcmbi.exe "+pdb_dir+os.path.dirname(i)+"\\"+os.path.basename(i)+" > "+pdb_dir+os.path.dirname(i)+"\\dsspout\\"+i[:-3].split("\\")[-1]+"dssp") ## EDIT
# print "DSSP -> "+pdb_dir+os.path.basename(i)
# print "Done"
#else:
# try:
# run_list = [z.strip() for z in fopen("new_files_downloaded_on_"+command_variables[1])]
# except IOError:
# tag = "new_files_downloaded_on_"
# print "The file \""+tag+command_variables[1]+"\" does not exist. Possible options are:"
# for line in [z for z in os.listdir(".") if z[:24] == tag]:
# print line
# sys.exit(1)
###
###temp
# import string
# run_list = [z.replace(".dssp",".ent") for z in open("dssplist","r").read().split("\n") if z != ""]
# aer = run_list.index("pdb2gi7.ent")
# run_list = run_list[aer:]
# print run_list
# run_list = ["pdb2gi7.ent"]
###
#histogram - for analysing relationships between secondary structures and bond types
hist_ = {}
#run with specified command parameters in run list
# run_list = ["pdb1gpq.ent","pdb1gps.ent","pdb1gpt.ent","pdb1gpz.ent","pdb1gqb.ent","pdb1gqr.ent","pdb1gqs.ent","pdb1gqv.ent","pdb1gqz.ent","pdb1gr2.ent","pdb1gra.ent","pdb1grn.ent","pdb1grt.ent","pdb1gsk.ent","pdb1gsm.ent","pdb1gsn.ent","pdb1gsp.ent","pdb1gt6.ent","pdb1gtp.ent","pdb1gts.ent","pdb1gtt.ent","pdb1gu2.ent","pdb1gu3.ent","pdb1guj.ent","pdb1gur.ent","pdb1guv.ent","pdb1gv7.ent","pdb1gv8.ent","pdb1gv9.ent","pdb1gvc.ent","pdb1gvk.ent","pdb1gvl.ent","pdb1gvt.ent","pdb1gvu.ent","pdb1gvv.ent","pdb1gvw.ent","pdb1gvx.ent","pdb1gvz.ent","pdb1gw0.ent","pdb1gw2.ent","pdb1gwa.ent","pdb1gwb.ent","pdb1gwd.ent","pdb1gwn.ent","pdb1gwo.ent","pdb1gwt.ent","pdb1gwu.ent","pdb1gx2.ent","pdb1gx8.ent","pdb1gx9.ent","pdb1gxa.ent","pdb1gxd.ent","pdb1gxs.ent","pdb1gxv.ent","pdb1gxx.ent","pdb1gxy.ent","pdb1gxz.ent","pdb1gy0.ent","pdb1gyc.ent","pdb1gyd.ent","pdb1gye.ent","pdb1gyh.ent","pdb1gyo.ent","pdb1gz1.ent","pdb1gz2.ent","pdb1gz7.ent","pdb1gza.ent","pdb1gzb.ent","pdb1gzj.ent","pdb1gzm.ent","pdb1gzp.ent","pdb1gzq.ent","pdb1gzr.ent","pdb1gzy.ent","pdb1gzz.ent","pdb1h02.ent","pdb1h03.ent","pdb1h04.ent","pdb1h0b.ent","pdb1h0d.ent","pdb1h0g.ent","pdb1h0h.ent","pdb1h0i.ent","pdb1h0j.ent","pdb1h0l.ent","pdb1h0z.ent","pdb1h12.ent","pdb1h13.ent","pdb1h14.ent","pdb1h15.ent","pdb1h1b.ent","pdb1h1h.ent","pdb1h1n.ent","pdb1h20.ent","pdb1h22.ent","pdb1h23.ent","pdb1h2b.ent","pdb1h2p.ent","pdb1h2q.ent","pdb1h30.ent","pdb1h34.ent","pdb1h3j.ent","pdb1h3p.ent","pdb1h3t.ent","pdb1h3u.ent","pdb1h3v.ent","pdb1h3w.ent","pdb1h3x.ent","pdb1h3y.ent","pdb1h43.ent","pdb1h44.ent","pdb1h45.ent","pdb1h46.ent","pdb1h49.ent","pdb1h4i.ent","pdb1h4j.ent","pdb1h4p.ent","pdb1h4u.ent","pdb1h4w.ent","pdb1h52.ent","pdb1h53.ent","pdb1h55.ent","pdb1h57.ent","pdb1h58.ent","pdb1h59.ent","pdb1h5a.ent","pdb1h5b.ent","pdb1h5c.ent","pdb1h5d.ent","pdb1h5e.ent","pdb1h5f.ent","pdb1h5g.ent","pdb1h5h.ent","pdb1h5i.ent","pdb1h5j.ent","pdb1h5k.ent","pdb1h5l.ent","pdb1h5m.ent","pdb1h5o.ent","pdb1h5x.ent","pdb1h6m.ent","pdb1h6r.ent","pdb1h6v.ent","pdb1h75.ent","pdb1h76.ent","pdb1h7l.ent","pdb1h7q.ent","pdb1h80.ent","pdb1h81.ent","pdb1h82.ent","pdb1h83.ent","pdb1h84.ent","pdb1h86.ent","pdb1h87.ent","pdb1h8d.ent","pdb1h8i.ent","pdb1h8l.ent","pdb1h8n.ent","pdb1h8o.ent","pdb1h8p.ent","pdb1h8s.ent","pdb1h8u.ent","pdb1h8v.ent","pdb1h8x.ent","pdb1h8y.ent","pdb1h8z.ent","pdb1h91.ent","pdb1h9h.ent","pdb1h9i.ent","pdb1h9l.ent","pdb1h9v.ent","pdb1h9z.ent","pdb1ha0.ent","pdb1ha2.ent","pdb1ha6.ent","pdb1ha8.ent","pdb1ha9.ent","pdb1haa.ent","pdb1hae.ent","pdb1haf.ent","pdb1hag.ent","pdb1hah.ent","pdb1hai.ent","pdb1haj.ent"]
n = len(run_list)
typedict = {}
for type_ in ["lys_arg", "his", "asp_glu", "trp", "phe", "tyr"]:
typedict[type_] = (analysis_type == type_)
timestart = time.ctime()
for i, file in enumerate(run_list):
#for i,file in enumerate(["pdb1w4y.ent"]):
# print type(hist_)
print pdb_dir + file
hist_ = analysis.run(pdb_dir + file,
x,
erfile,
dist=threshold,
histogram=hist_,
lys_arg=typedict["lys_arg"],
his=typedict["his"],
asp_glu=typedict["asp_glu"],
trp=typedict["trp"],
phe=typedict["phe"],
tyr=typedict["tyr"])
m = int((i + 1) / float(n) * 10000) / 100.0
print str(m) + "% complete"
timefinish1 = time.ctime()
print "\nResults file is being saved. Please do not close the program window.\n"
##################################################################
##The histogram bit
def histcmp(a, b):
return (hist_[a] < hist_[b]) * 2 - 1
#fffff = open("histogram_","w")
#hkeys = hist_.keys()[:]
#hkeys.sort(histcmp)
#x.add_sheet("Histogram")
#histosheet = x.get_sheet(4)
#histosheet.write(0,0,"Secondary structure 1",style0)
#histosheet.write(0,1,"Secondary structure 2",style0)
#histosheet.write(0,2,"Disulfide Bond type",style0)
#histosheet.write(0,3,"Count",style0)
#for row,h in enumerate(hkeys):
# fffff.write(str(h)+" "*(21-len(h))+str(hist_[h])+"\n")
# histosheet.write(row+1,0,h[0])
# histosheet.write(row+1,1,h[1])
# histosheet.write(row+1,2,h[2:])
# histosheet.write(row+1,3,hist_[h])
#fffff.close()
##################################################################
realactive = x.get_active_sheet()
firstactive = 6
sheet = x.get_sheet(0)
if sheet.row(0).get_str_count() > 0:
firstactive = 0
sheet.set_first_visible_row(0)
sheet = x.get_sheet(1)
if sheet.row(0).get_str_count() > 0 and firstactive > 1:
firstactive = 1
sheet.set_first_visible_row(0)
sheet = x.get_sheet(2)
if sheet.row(0).get_str_count() > 0 and firstactive > 2:
firstactive = 2
sheet.set_first_visible_row(0)
sheet = x.get_sheet(3)
if sheet.row(0).get_str_count() > 0 and firstactive > 3:
firstactive = 3
sheet.set_first_visible_row(0)
sheet = x.get_sheet(4)
if sheet.row(0).get_str_count() > 0 and firstactive > 4:
firstactive = 3
sheet.set_first_visible_row(0)
sheet = x.get_sheet(5)
sheet.set_first_visible_row(0)
if sheet.row(0).get_str_count() > 0 and firstactive > 5:
firstactive = 5
sheet.set_first_visible_row(0)
sheet = x.get_sheet(6)
sheet.set_first_visible_row(0)
x.set_active_sheet(firstactive)
x.save(write_path)
if inA:
os.system(
"7z a -tzip -mx=9 D:\\xampp\\htdocs\\python\\Analyses\\disulfide_analysis_"
+ date_ + ".zip " + write_path)
print "\nAnalysis complete and results saved.\n"
timefinish2 = time.ctime()
erfile.close()
print "start =", timestart
print "prcessed =", timefinish1
print "saved =", timefinish2
#sorting and printing of histogram
#gives descending sort
#write histogram to spreadsheet
#cys["SSBOND 7 CYS A 1348 CYS A 1380"]
return write_path
if args.barcode_level:
counts['unique_id'] = list(counts.index)
label = "barcode_level"
elif args.feature_level:
counts['unique_id'] = [
str(i) + "_" + str(j) + "_" + str(k) + "_" + str(l)
for i, j, k, l in zip(counts['sseqid'], counts['guide_status'],
counts['donor_status'], counts['library'])
]
label = "feature_level"
"""elif args.v2:
counts['unique_id'] = [str(i)+"_"+str(j)+"_"+str(k) for i,j,k in zip(counts['ref'], counts['guide_status'], counts['MD'])]
label = "feature_level"
else: # default is barcode level analysis
counts['unique_id'] = list(counts.index)
label = "barcode_level"
"""
#counts = counts[samples + ['unique_id', 'ref', 'guide_status', 'MD']].dropna(subset=samples, how='all')
counts = counts[samples + [
"unique_id", "sseqid", "library", "subpool", "guide_status",
"donor_status", "bsp_status"
]].dropna(subset=samples, how='all')
counts['percentNA'] = counts[samples].isnull().sum(axis=1) / float(
len(samples))
counts = counts[counts['percentNA'] < 0.25]
counts.to_csv(outfile + "_" + label + "_counts.csv")
results, log = analysis.run(counts, model, log=True)
results.to_csv(outfile + "_" + label + "_analysis.csv")
log.to_csv(outfile + "_" + label + "_log.csv")
def fig_Psizesep(outfile='fig_Psizesep.png'):
"""
Args:
outfile:
Returns:
"""
set_mplrc()
sns.set_theme()
sns.set_style('whitegrid')
sns.set_context('paper')
# Load f_mL
f_mL = analysis.load_f_mL()
frb_pre = analysis.get_candidates()
# Init prior
prior = associate_defs.adopted.copy()
# Plot
plt.figure(figsize=(7, 7))
# Bayesian + parse
_, model_mags, model_theta, max_PMix = analysis.run(prior, frb_pre=frb_pre)
#bins_L = np.linspace(0.01, 10., 20)
bins_L = np.linspace(-2., 1, 15)
#bins_L = np.linspace(0.1, 2., 20)
#bins_L = np.linspace(16., 25., 20) # Testing only
# High confidence
high_conf_size, high_conf_sep = [], []
max_sep = 0.
for frbA in frb_pre.frbA:
if np.max(frbA.candidates.P_Ox) > associate_defs.POx_secure:
# Max
imax = np.argmax(frbA.candidates.P_Ox)
# Measures
ang_phys = associate_defs.cosmo.kpc_proper_per_arcmin(frbA.frb.z)
high_conf_sep.append((frbA.candidates.iloc[imax].separation *
units.arcsec * ang_phys).to('kpc').value)
high_conf_size.append(
((frbA.candidates.iloc[imax].half_light * units.arcsec) *
ang_phys).to('kpc').value)
#
if high_conf_sep[-1] > max_sep:
max_sep = high_conf_sep[-1]
max_FRB = frbA.frb
high_conf_size = np.array(high_conf_size)
high_conf_sep = np.array(high_conf_sep)
print("Max sep: {}".format(max_FRB))
# Plot m_r vs. z
df = pandas.DataFrame(dict(size=high_conf_size, sep=high_conf_sep))
jg = sns.jointplot(data=df, x='sep', y='size')
#jg.ax_marg_x.set_xlim(0, 0.6)
jg.ax_marg_y.set_ylim(0., 7)
jg.ax_joint.set_xlabel('Physical Separation (kpc)')
jg.ax_joint.set_ylabel('Galaxy size (kpc)')
# One-to-one line
jg.ax_joint.plot([0., 11.], [0., 11.], 'k--')
#jg.ax_joint.yaxis.set_major_locator(plt.MultipleLocator(0.5))
#jg.ax_joint.yaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
# Font size
set_fontsize(jg.ax_joint, 15.)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def fig_prior_vs_posterior(outfile='fig_prior_vs_posterior.png',
POx_secure=associate_defs.POx_secure,
prior_mode='conservative'):
"""
Args:
outfile:
Returns:
"""
set_mplrc()
frb_pre = analysis.get_candidates()
# Init prior
prior = getattr(associate_defs, prior_mode).copy()
thetas = np.linspace(0., associate_defs.theta_max, 1000)
dtheta = thetas[1] - thetas[0]
bins_theta = np.linspace(0., 6., 20)
# Plot
plt.figure(figsize=(6, 5))
gs = gridspec.GridSpec(2, 2)
# Loop on theta priors
for ss, clr, tprior in zip(np.arange(3), ['k', 'b', 'g', 'gray'], [
associate_defs.theta_u,
associate_defs.theta_c,
associate_defs.theta_e,
associate_defs.theta_e,
]):
# Plot
ax = plt.subplot(gs[ss])
# Prior
prior['theta'] = tprior
if ss < 3:
# Bayesian + parse
_, model_mags, model_theta, max_PMix = analysis.run(
prior, frb_pre=frb_pre)
# Distribution
ptheta = bayesian.pw_Oi(thetas, 1., prior['theta'])
scl = np.sum(ptheta * dtheta)
print('scl:', scl)
convolved_ptheta = None
# High confidence
high_conf_theta = []
nsecure = 0
for frbA in frb_pre.frbA:
if np.max(frbA.candidates.P_Ox) > POx_secure:
imax = np.argmax(frbA.candidates.P_Ox)
high_conf_theta.append(
frbA.candidates.iloc[imax].separation /
frbA.candidates.iloc[imax].half_light)
# Convolve with prior
sigR_phi = np.sqrt(
frbA.frb.sig_a *
frbA.frb.sig_b) / frbA.candidates.iloc[imax].half_light
conv_theta = convolve(ptheta,
np.exp(-thetas**2 / 2 / sigR_phi**2),
mode='full')
scl_c = np.sum(conv_theta * dtheta)
if convolved_ptheta is None:
convolved_ptheta = conv_theta / scl_c
else:
convolved_ptheta += conv_theta / scl_c
# Increment
nsecure += 1
# Nomalize
high_conf_theta = np.array(high_conf_theta)
convolved_ptheta /= nsecure
# Save
if tprior == associate_defs.theta_e:
save_high = high_conf_theta.copy()
# KS test
#cumsum = np.cumsum(ptheta*dtheta) / scl
cumsum = np.cumsum(convolved_ptheta * dtheta)
more_thetas = np.arange(convolved_ptheta.size) * dtheta
#f_CDF = interp1d(thetas, cumsum)
f_CDF = interp1d(more_thetas, cumsum)
res = kstest(high_conf_theta, f_CDF)
pvalue = res.pvalue
else:
# Fit secure with a new exponential
scale_lengths = np.linspace(0.25, 2., 100)
P_KS = []
for scale_length in scale_lengths:
ptheta = bayesian.pw_Oi(thetas,
1.,
associate_defs.theta_e,
scale_half=scale_length)
scl = np.sum(ptheta * dtheta)
# KS test
cumsum = np.cumsum(ptheta * dtheta) / scl
f_CDF = interp1d(thetas, cumsum)
res = kstest(high_conf_theta, f_CDF)
# Save
P_KS.append(res.pvalue)
imax = np.argmax(P_KS)
best_sl = scale_lengths[imax]
print("Best scale length = {}".format(best_sl))
pvalue = P_KS[imax]
# Once more
ptheta = bayesian.pw_Oi(thetas,
1.,
associate_defs.theta_e,
scale_half=best_sl)
scl = np.sum(ptheta * dtheta)
prior['theta']['method'] = 'exp ' + '{:0.1f}'.format(
best_sl) + r'$ \, ' + cmhalf + '$'
#
high_conf_theta = save_high
# Plot profiles
ax.plot(thetas,
ptheta / scl,
label=prior['theta']['method'] + ' unconvolved',
color=clr,
alpha=0.3)
more_thetas = np.arange(convolved_ptheta.size) * dtheta
ax.plot(more_thetas,
convolved_ptheta,
label=prior['theta']['method'] + r' $P_{\rm KS} =' +
'{:0.2f}'.format(pvalue) + '$',
color=clr)
# Plot secure
weights3 = np.ones_like(high_conf_theta) / high_conf_theta.size
lbl = r'$P(O_i) > ' + '{}'.format(
POx_secure) + '$ FRBs' if ss == 0 else None
ax.hist(high_conf_theta,
bins=bins_theta,
weights=weights3,
color='darkred',
label=lbl,
histtype='stepfilled')
# Plot all
if ss < 3:
weights2 = np.ones_like(model_theta) / model_theta.size
lbl = 'All FRB candidates' if ss == 0 else None
ax.hist(model_theta,
weights=weights2,
bins=bins_theta,
color='darkgrey',
label=lbl,
histtype='step')
# Label me
ax.set_xlabel(r'$\theta/' + cmhalf + '$')
ax.set_ylabel('PDF')
ax.xaxis.set_major_locator(plt.MultipleLocator(1.))
#ax.xaxis.set_major_formatter(FormatStrFormatter(r'$%.3f$'))
ax.set_xlim(0., 8)
ax.set_ylim(0., 0.5)
# Legend
legend = ax.legend(loc='upper right',
scatterpoints=1,
borderpad=0.2,
handletextpad=handletextpad,
fontsize=11.)
# Font size
set_fontsize(ax, 13.)
# End
plt.tight_layout(pad=0.2, h_pad=0., w_pad=0.1)
print('Writing {:s}'.format(outfile))
kwargs = {}
if 'png' in outfile:
kwargs['dpi'] = 700
plt.savefig(outfile, **kwargs)
plt.close()
def main(analysis_type, command_variables, threshold):
#pdb_dir = fopen("absolute_path_of_pdb_files")[0].strip()
pdb_dir = "../pdb_files"
x = Workbook()
x.add_sheet("NMR averaged model")
x.add_sheet("NMR multiple models")
x.add_sheet("X-RAY")
x.add_sheet("Other")
date_ = time.strftime("%d_%b_%Y")
x.save("../Analyses/" + analysis_type + "_" + date_ + ".xls")
# file pdb1a5n is missing, possibly because 1a5n does not contain ssbonds (correct)
#"pdb1tdy.ent","pdb1a5n.ent","pdb1cb6.ent","pdb2ac5.ent"
#for file in [pdb_dir+name for name in ["pdb1kdg.ent", "pdb1kdk.ent", "pdb1kdm.ent", "pdb1kdq.ent", "pdb1kdu.ent", "pdb1kdv.ent", "pdb1kdy.ent", "pdb1kdz.ent", "pdb1ke1.ent", "pdb1ke2.ent", "pdb1keb.ent", "pdb1keg.ent", "pdb1kek.ent", "pdb1kel.ent", "pdb1kem.ent", "pdb1ken.ent", "pdb1keo.ent", "pdb1kex.ent", "pdb1kf2.ent", "pdb1kf3.ent", "pdb1kf4.ent", "pdb1kf5.ent", "pdb1kf7.ent"]]:
# run(file,x)
#try:
# open("..\\Analyses\\basic_analysis_"+time.strftime("%a_%d_%b_%Y")+".xls","r")
#except IOError:
#for file in os.listdir(pdb_dir):
#for file in ["pdb1axi.ent"]:
# run(pdb_dir+file,x)
#n = len(os.listdir(pdb_dir))
erfile = open("error_" + analysis_type + "_" + date_ + ".log", "w")
#check command parameters
if len(command_variables) <= 1:
run_list = os.listdir(pdb_dir)
run_list.remove("dsspout")
print "1"
elif command_variables[1].find(".ent") != -1:
try:
run_list = [command_variables[1]]
print "2"
#pdb_dir = "./"
except IOError:
print "The file " + command_variables[1] + " was not found."
sys.exit(1)
else:
print "3"
try:
run_list = [
z.strip() for z in fopen("new_files_downloaded_on_" +
command_variables[1])
]
except IOError:
tag = "new_files_downloaded_on_"
print "The file \"" + tag + command_variables[
1] + "\" does not exist. Possible options are:"
for line in [z for z in os.listdir(".") if z[:24] == tag]:
print line
sys.exit(1)
###
###temp
# import string
# run_list = [z.replace(".dssp",".ent") for z in open("dssplist","r").read().split("\n") if z != ""]
# aer = run_list.index("pdb2gi7.ent")
# run_list = run_list[aer:]
# print run_list
# run_list = ["pdb2gi7.ent"]
###
#run_list = ["pdb2fz3.ent"]
#histogram - for analysing relationships between secondary structures and bond types
hist_ = {}
#run with specified command parameters in run list
# run_list = ["pdb1gpq.ent","pdb1gps.ent","pdb1gpt.ent","pdb1gpz.ent","pdb1gqb.ent","pdb1gqr.ent","pdb1gqs.ent","pdb1gqv.ent","pdb1gqz.ent","pdb1gr2.ent","pdb1gra.ent","pdb1grn.ent","pdb1grt.ent","pdb1gsk.ent","pdb1gsm.ent","pdb1gsn.ent","pdb1gsp.ent","pdb1gt6.ent","pdb1gtp.ent","pdb1gts.ent","pdb1gtt.ent","pdb1gu2.ent","pdb1gu3.ent","pdb1guj.ent","pdb1gur.ent","pdb1guv.ent","pdb1gv7.ent","pdb1gv8.ent","pdb1gv9.ent","pdb1gvc.ent","pdb1gvk.ent","pdb1gvl.ent","pdb1gvt.ent","pdb1gvu.ent","pdb1gvv.ent","pdb1gvw.ent","pdb1gvx.ent","pdb1gvz.ent","pdb1gw0.ent","pdb1gw2.ent","pdb1gwa.ent","pdb1gwb.ent","pdb1gwd.ent","pdb1gwn.ent","pdb1gwo.ent","pdb1gwt.ent","pdb1gwu.ent","pdb1gx2.ent","pdb1gx8.ent","pdb1gx9.ent","pdb1gxa.ent","pdb1gxd.ent","pdb1gxs.ent","pdb1gxv.ent","pdb1gxx.ent","pdb1gxy.ent","pdb1gxz.ent","pdb1gy0.ent","pdb1gyc.ent","pdb1gyd.ent","pdb1gye.ent","pdb1gyh.ent","pdb1gyo.ent","pdb1gz1.ent","pdb1gz2.ent","pdb1gz7.ent","pdb1gza.ent","pdb1gzb.ent","pdb1gzj.ent","pdb1gzm.ent","pdb1gzp.ent","pdb1gzq.ent","pdb1gzr.ent","pdb1gzy.ent","pdb1gzz.ent","pdb1h02.ent","pdb1h03.ent","pdb1h04.ent","pdb1h0b.ent","pdb1h0d.ent","pdb1h0g.ent","pdb1h0h.ent","pdb1h0i.ent","pdb1h0j.ent","pdb1h0l.ent","pdb1h0z.ent","pdb1h12.ent","pdb1h13.ent","pdb1h14.ent","pdb1h15.ent","pdb1h1b.ent","pdb1h1h.ent","pdb1h1n.ent","pdb1h20.ent","pdb1h22.ent","pdb1h23.ent","pdb1h2b.ent","pdb1h2p.ent","pdb1h2q.ent","pdb1h30.ent","pdb1h34.ent","pdb1h3j.ent","pdb1h3p.ent","pdb1h3t.ent","pdb1h3u.ent","pdb1h3v.ent","pdb1h3w.ent","pdb1h3x.ent","pdb1h3y.ent","pdb1h43.ent","pdb1h44.ent","pdb1h45.ent","pdb1h46.ent","pdb1h49.ent","pdb1h4i.ent","pdb1h4j.ent","pdb1h4p.ent","pdb1h4u.ent","pdb1h4w.ent","pdb1h52.ent","pdb1h53.ent","pdb1h55.ent","pdb1h57.ent","pdb1h58.ent","pdb1h59.ent","pdb1h5a.ent","pdb1h5b.ent","pdb1h5c.ent","pdb1h5d.ent","pdb1h5e.ent","pdb1h5f.ent","pdb1h5g.ent","pdb1h5h.ent","pdb1h5i.ent","pdb1h5j.ent","pdb1h5k.ent","pdb1h5l.ent","pdb1h5m.ent","pdb1h5o.ent","pdb1h5x.ent","pdb1h6m.ent","pdb1h6r.ent","pdb1h6v.ent","pdb1h75.ent","pdb1h76.ent","pdb1h7l.ent","pdb1h7q.ent","pdb1h80.ent","pdb1h81.ent","pdb1h82.ent","pdb1h83.ent","pdb1h84.ent","pdb1h86.ent","pdb1h87.ent","pdb1h8d.ent","pdb1h8i.ent","pdb1h8l.ent","pdb1h8n.ent","pdb1h8o.ent","pdb1h8p.ent","pdb1h8s.ent","pdb1h8u.ent","pdb1h8v.ent","pdb1h8x.ent","pdb1h8y.ent","pdb1h8z.ent","pdb1h91.ent","pdb1h9h.ent","pdb1h9i.ent","pdb1h9l.ent","pdb1h9v.ent","pdb1h9z.ent","pdb1ha0.ent","pdb1ha2.ent","pdb1ha6.ent","pdb1ha8.ent","pdb1ha9.ent","pdb1haa.ent","pdb1hae.ent","pdb1haf.ent","pdb1hag.ent","pdb1hah.ent","pdb1hai.ent","pdb1haj.ent"]
n = len(run_list)
typedict = {}
for type_ in ["lys_arg", "his", "asp_glu", "trp", "phe", "tyr"]:
typedict[type_] = (analysis_type == type_)
timestart = time.ctime()
for i, file in enumerate(run_list):
#for i,file in enumerate(["pdb1w4y.ent"]):
# print type(hist_)
hist_ = analysis.run(pdb_dir + file,
x,
erfile,
dist=threshold,
histogram=hist_,
lys_arg=typedict["lys_arg"],
his=typedict["his"],
asp_glu=typedict["asp_glu"],
trp=typedict["trp"],
phe=typedict["phe"],
tyr=typedict["tyr"])
m = int((i + 1) / float(n) * 10000) / 100.0
print str(m) + "% complete"
timefinish1 = time.ctime()
print "\nResults file is being saved. Please do not close the program window.\n"
##################################################################
##The histogram bit
def histcmp(a, b):
return (hist_[a] < hist_[b]) * 2 - 1
fffff = open("histogram_", "w")
hkeys = hist_.keys()[:]
hkeys.sort(histcmp)
x.add_sheet("Histogram")
histosheet = x.get_sheet(4)
histosheet.write(0, 0, "Secondary structure 1", style0)
histosheet.write(0, 1, "Secondary structure 2", style0)
histosheet.write(0, 2, "Disulfide Bond type", style0)
histosheet.write(0, 3, "Count", style0)
for row, h in enumerate(hkeys):
fffff.write(str(h) + " " * (21 - len(h)) + str(hist_[h]) + "\n")
histosheet.write(row + 1, 0, h[0])
histosheet.write(row + 1, 1, h[1])
histosheet.write(row + 1, 2, h[2:])
histosheet.write(row + 1, 3, hist_[h])
fffff.close()
##################################################################
x.save("..\\Analyses\\" + analysis_type + "_" + date_ + ".xls")
print "\nAnalysis complete and results saved.\n"
timefinish2 = time.ctime()
erfile.close()
print "start =", timestart
print "prcessed =", timefinish1
print "saved =", timefinish2