import utility.referenceUtil as ref
# Parse the input data text file
data = io.readInputDataFiles('input_data.txt')
datatypes = data.keys()
data_dict = {}
print datatypes
if 'fasta_file' in datatypes:
# Read in the amino acid sequence and the Secondary structure assignment
handle, aa_seq = io.readFasta(data['fasta_file'])
ss_seq = ru.matchSeq2SS(aa_seq, data['ss_file'])
# Generate fuzzy +/-2 SSE combinations
ss_def, ss_combi = ss.genSSCombinations(ss_seq)
io.dumpPickle("ss_profiles.pickle", ss_combi)
data_dict['ss_seq'] = ss_seq
data_dict['aa_seq'] = aa_seq
print ss_def
print ss_seq
else:
pass
if 'native_pdbs' in datatypes:
# Read the native pdbs that you can exclude from the smotif search
native_pdbs = data['native_pdbs']
native_pdbs = native_pdbs.lower()
native_pdbs = native_pdbs.split()
data_dict['natives'] = native_pdbs
print native_pdbs
else:
data = io.readInputDataFiles('input_data.txt')
datatypes = data.keys()
handle, aa_seq = io.readFasta(data['fasta_file'])
ss_seq = matchSeq2SS(aa_seq, data['ss_file'])
print ss_seq
# ss_seq = io.readPsiPred(psipred_file)
ss_def, ss_combi = ss.genSSCombinations(ss_seq)
# ss_def, ss_combi = ss.genSSCombinations2(ss_seq)
print ss_combi
io.dumpPickle("ss_profiles.pickle", ss_combi)
# Read in contacts at a given confidence level
if 'contacts_file' in datatypes:
contacts, contacts_seq = io.readContacts(contactsfile, probability=0.7)
native_pdbs = data['native_pdbs']
native_pdbs = native_pdbs.lower()
native_pdbs = native_pdbs.split()
print native_pdbs
axrh_cutoff = data['axrh_cutoff']
axrh_cutoff = axrh_cutoff.split()
def SmotifSearch(index_array):
"""
Main()
:param index_array:
:return:
"""
exp_data = io.readPickle("exp_data.pickle")
exp_data_types = exp_data.keys(
) # ['ss_seq', 'pcs_data', 'aa_seq', 'contacts']
psmotif = uts2.getPreviousSmotif(index_array[0])
current_ss, direction = uts2.getSS2(index_array[1])
csmotif_data, smotif_def = getfromDB(psmotif, current_ss, direction,
exp_data['database_cutoff'])
if not csmotif_data:
# If the smotif library doesn't exist
# Terminate further execution
return True
"""
always narrow down to previous sse and current sse and operate on them individually
"""
sse_ordered = orderSSE(psmotif, current_ss, direction)
dump_log = []
no_clashes = False
# ************************************************************************************************
# Main
# The 'for' loop below iterates over all of the Smotifs and applies various filters
# This is the place to add new filters as you desire. For starters, look at Sequence filter.
# ************************************************************************************************
for i in range(0, len(csmotif_data)):
# ************************************************
# Applying different filters for the Smotif assembly
# ************************************************
# Exclude natives if needed
if 'natives' in exp_data_types:
natives = exp_data['natives']
tpdbid = csmotif_data[i][0][0]
pdbid = tpdbid[0:4]
if pdbid in natives:
#if pdbid not in ['2z2i']:
# Stop further execution and
continue
# ************************************************
# RMSD filter using QCP method
# quickly filters non-overlapping smotifs
# ************************************************
rmsd, transformed_coos = qcp.rmsdQCP(psmotif[0], csmotif_data[i],
direction)
if rmsd <= exp_data['rmsd_cutoff'][1]:
# Loop constraint restricts the overlapping smotifs is not drifted far away.
loop_constraint = llc.loopConstraint(transformed_coos, sse_ordered,
direction, smotif_def)
if loop_constraint:
# Check whether the SSEs with in the assembled smotifs are clashing to one another
no_clashes = qcp.clahses(transformed_coos,
exp_data['clash_distance'])
else:
no_clashes = False
if rmsd <= exp_data['rmsd_cutoff'][1] and no_clashes:
# Prepare temp log array to save data at the end
tlog, noe_fmeasure, pcs_tensor_fits, rdc_tensor_fits = [], [], [], []
tlog.append(['smotif', csmotif_data[i]])
tlog.append(['smotif_def', sse_ordered])
tlog.append(['qcp_rmsd', transformed_coos, sse_ordered, rmsd])
cathcodes = sm.orderCATH(psmotif, csmotif_data[i][0], direction)
tlog.append(['cathcodes', cathcodes])
# ************************************************
# Sequence filter
# Aligns the smotif seq to target seq and calculates
# sequence identity and the alignment score
# ************************************************
csse_seq, seq_identity, blosum62_score = Sfilter.S2SequenceSimilarity(
current_ss, csmotif_data[i], direction, exp_data)
# concat current to previous seq
concat_seq = sm.orderSeq(psmotif, csse_seq, direction)
tlog.append([
'seq_filter', concat_seq, csse_seq, seq_identity,
blosum62_score
])
# ************************************************
# Pseudocontact Shift filter
# uses experimental PCS data to filter Smotifs
# scoring based on normalised chisqr
# ************************************************
if 'pcs_data' in exp_data_types:
pcs_tensor_fits = Pfilter.PCSAxRhFit2(transformed_coos,
sse_ordered,
exp_data,
stage=2)
tlog.append(['PCS_filter', pcs_tensor_fits])
# ************************************************
# Ambiguous NOE score filter
# uses experimental ambiguous noe data to filter Smotifs
# scoring based on f-measure?
# ************************************************
if 'noe_data' in exp_data_types:
noe_fmeasure = Nfilter.s2NOEfit(transformed_coos, sse_ordered,
exp_data)
tlog.append(['NOE_filter', noe_fmeasure])
# ************************************************
# Residual dipolar coupling filter
# uses experimental RDC data to filter Smotifs
# scoring based on normalised chisqr
# ************************************************
if 'rdc_data' in exp_data_types:
if noe_fmeasure and noe_fmeasure > 0.5:
rdc_tensor_fits = Rfilter.RDCAxRhFit2(transformed_coos,
sse_ordered,
exp_data,
stage=2)
tlog.append(['RDC_filter', rdc_tensor_fits])
if pcs_tensor_fits or rdc_tensor_fits:
#dump data to the disk
print tpdbid, noe_fmeasure, rdc_tensor_fits
# print csmotif_data[i][0], 'blosum62 score', blosum62_score, "seq_id", seq_identity, "rmsd=", rmsd, cathcodes
dump_log.append(tlog)
# prevent dumping empty arrays with no data
if len(dump_log) > 0:
print "num of hits", len(dump_log),
io.dumpPickle(
"tx_" + str(index_array[0]) + "_" + str(index_array[1]) +
".pickle", dump_log)
return True
print smotif_db[i][45:]
for j in range(0, len(smotif)):
# for j in range(0, 1):
# print smotif[j][0]
# print smotif[j][0][0] #['1pp9P00', '172', '203', '222', '245']
# print smotif[j][0][1] #172- [203, 'THR', 'H', 49.366, 87.846, 102.07]]
# print smotif[j][0][2] #222- [245, 'PHE', 'H', 40.016, 57.964, 78.166]]
ss1 = rms.getcoo(
smotif[j][0]
[1]) # x,y,z, atom_type, res_no, res = [], [], [], [], [], []
ss2 = rms.getcoo(
smotif[j][0][2]) # return [x,y,z, atom_type, res_no, res]
# dumpPDBCoo2(ss1, 5)
# dumpPDBCoo2(ss2, 6)
# concatenate coordinate arrys of both SSEs to identify their center of mass
ss1_plus_ss2 = conc_sss(ss1, ss2)
cen_ss1, cm_ss1 = rms.centerCoo(ss1_plus_ss2)
# dumpPDBCoo2(cen_ss1, 7)
# Translate the original SSEs to the new center of mass
tr_ss1 = translateSSE(cm_ss1, smotif[j][0][1])
tr_ss2 = translateSSE(cm_ss1, smotif[j][0][2])
smotif_def = smotif[j][0][0]
# add the modified Smotifs to temp array
temp_smotif.append([smotif_def, tr_ss1, tr_ss2])
io.dumpPickle('./smotif_cen_db/' + smotif_db[i][45:], temp_smotif)
def SmotifSearch(index_array):
"""
Main ()
:param index_array:
:return:
"""
# print index_array
s1_def, s2_def = getSSdef(index_array)
smotif_def = sm.getSmotif(s1_def, s2_def)
# print s1_def, s2_def
exp_data = io.readPickle("exp_data.pickle")
exp_data_types = exp_data.keys(
) # ['ss_seq', 'pcs_data', 'aa_seq', 'contacts']
smotif_data = sm.readSmotifDatabase(smotif_def,
exp_data['database_cutoff'])
if not smotif_data:
# If the smotif library doesn't exist, terminate further execution.
return True
dump_log = []
# ************************************************************************************************
# Main
# The 'for' loop below iterates over all of the Smotifs and applies various filters
# This is the place to add new filters as you desire. For starters, look at Sequence filter.
# ************************************************************************************************
for i in range(0, len(smotif_data)):
# loop over for all of the entries in the smotif_db file
# ************************************************
# Excluding the natives
# ************************************************
if 'natives' in exp_data_types:
natives = exp_data['natives']
tpdbid = smotif_data[i][0][0]
pdbid = tpdbid[0:4]
if pdbid in natives:
#if pdbid not in ['2z2i']:
# Stop further execution, but, iterate.
continue
# ************************************************
# Applying different filters to Smotifs
# Prepare temp log array to save data at the end
# ************************************************
tlog, pcs_tensor_fits, rdc_tensor_fits, noe_fmeasure = [], [], [], []
tlog.append(['smotif', smotif_data[i]])
tlog.append(['smotif_def', [s1_def, s2_def]])
tlog.append(['cathcodes', [smotif_data[i][0]]])
# ************************************************
# Sequence filter
# Aligns the smotif seq to target seq and calculates
# sequence identity and the alignment score
# ************************************************
smotif_seq, seq_identity, blosum62_score = \
Sfilter.SequenceSimilarity(s1_def, s2_def, smotif_data[i], exp_data)
tlog.append(['seq_filter', smotif_seq, seq_identity, blosum62_score])
# ************************************************
# Pseudocontact Shift filter
# uses experimental PCS data to filter Smotifs
# scoring based on normalised chisqr
# ************************************************
if 'pcs_data' in exp_data_types:
pcs_tensor_fits = Pfilter.PCSAxRhFit(s1_def, s2_def,
smotif_data[i], exp_data)
tlog.append(['PCS_filter', pcs_tensor_fits])
# ************************************************
# Ambiguous NOE score filter
# uses experimental ambiguous noe data to filter Smotifs
# scoring based on f-measure?
# ************************************************
if 'noe_data' in exp_data_types:
noe_fmeasure = Nfilter.s1NOEfit(s1_def, s2_def, smotif_data[i],
exp_data)
tlog.append(['NOE_filter', noe_fmeasure])
# ************************************************
# Residual dipolar coupling filter
# uses experimental RDC data to filter Smotifs
# scoring based on normalised chisqr
# ************************************************
if 'rdc_data' in exp_data_types:
if noe_fmeasure and noe_fmeasure > 0.5:
rdc_tensor_fits = Rfilter.RDCAxRhFit(s1_def, s2_def,
smotif_data[i], exp_data)
tlog.append(['RDC_filter', rdc_tensor_fits])
# Dump the data to the disk
if pcs_tensor_fits or rdc_tensor_fits:
# print smotif_data[i][0][0], "seq_id", seq_identity, "i=", i, "/", len(smotif_data)
print tpdbid, noe_fmeasure, rdc_tensor_fits
dump_log.append(tlog)
# Save all of the hits in pickled arrays
if dump_log:
print "num of hits", len(dump_log)
io.dumpPickle(
'0_' + str(index_array[0]) + "_" + str(index_array[1]) + ".pickle",
dump_log)
return True
