#! /usr/bin/env ccp4-python
#
# Copyright (C) 2005 Ronan Keegan
#
# This code is distributed under the terms and conditions of the
# CCP4 Program Suite Licence Agreement as a CCP4 Application.
# A copy of the CCP4 licence can be obtained by writing to the
# CCP4 Secretary, Daresbury Laboratory, Warrington WA4 4AD, UK.
#
#
#
# Script to write the results HTML for bulkMR
# Ronan Keegan 24.8.05
import os, string, sys
import os.path
import time
import shutil
import MRBUMP_Gnuplot
import Model_prep_scoring
class Write_html:
""" A class to write output html as the bulkMR job progresses. """
def __init__(self):
self.page=''
self.seq_array=[]
self.elem_size=60
self.no_elems=''
try:
self.debug=eval(os.environ['MRBUMP_DEBUG'])
except:
self.debug=False
def write_header(self, init, results_html):
""" A function to write the header for the results html page. """
# Begin the html page
line = '\n'
# Set the title for the page
line += '\n'
line += 'BulkMR Results\n'
line += '\n'
# Set the background colour
line += '\n'
# Add the logos
line += '\n'
line += '\n'
line += ' | \n'
line += ' | \n'
line += '
\n'
line += '
\n'
# Add a title bar and link to the e-HTPX homepage
line += '\n'
# Open the main body of the page
line += '\n'
res_html = open(results_html, "w")
res_html.write(line)
res_html.close()
def write_footer(self, init, mstat, results_html):
""" A function to write the footer for the results html page."""
# Refresh the page every 10 seconds
if mstat.job_end == False:
if init.keywords.results_html == '':
line = '\n'
else:
line = '\n'
line += '
\n'
line += '\n'
line += '
\n'
line += '
\n'
line += '\n'
line += '\n'
else:
line = '
\n'
line += '\n'
line += 'Job completed: ' + time.ctime() + '. Time taken: %.3f seconds.\n' % mstat.job_time
line += '\n'
line += '
\n'
# Add the footer bar
line += '
\n'
line += '\n'
line += ' |
\n'
line += '
\n'
# put in the time when the results were generated
line += '\n'
line += '| Results generated: ' + time.ctime() + ' |
\n'
line += '
\n'
# Close the page
line += '
\n'
line += '\n'
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
mstat.len_html_footer = len(line)
def remove_footer(self, mstat, results_html):
""" A function to strip the footer from the results.html file so as to add extra search details.
Only to be used when one is about to update the results.html. """
# Get the size of the results.html
file_size = os.path.getsize(results_html)
# We will truncate the file to exclude the footer
truncate_size = file_size - mstat.len_html_footer
# Get the file contents minus the footer
res_html = open(results_html, "r")
old_file = res_html.read()[0:truncate_size]
res_html.close()
# Re-write the file witout the footer
res_html = open(results_html, "w")
res_html.write(old_file)
res_html.close()
def write_target_details(self, init, mstat, target_info, results_html):
""" A function to write the target details in the results html page. """
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
# Write the title
line = '\n'
line += ' BulkMR results for target structure: ' \
+ target_info.chainID + '\n'
line += '
\n'
if mstat.InitError[0] == False:
# Add the target details to the results.html
line += 'Target Information
\n'
line += 'Target chain ID: ' + target_info.chainID + '
\n'
line += 'Target sequence:
\n'
self.split_seq(target_info)
line += '\n'
for i in range(self.no_elems):
line += self.seq_array[i] + '
\n'
line += '\n'
line += '
\n'
line += 'Sequence file: ' + target_info.seqfilename + '
\n'
line += 'Reflections file: ' + target_info.hklin_filename + '
\n'
line += 'Resolution of collected data: %.3f
\n' % target_info.resolution
line += 'Number of Residues: ' + `target_info.no_of_res` + '
\n'
line += 'Molecular Weight: %.3f
\n\n' % target_info.mol_weight
if init.keywords.NMASU == None:
line += 'Estimated number of molecules in the a.s.u.: ' + `target_info.no_of_mols` + '
\n'
else:
line += 'User defined number of molecules in the a.s.u.: ' + `target_info.no_of_mols` + '
\n'
line += 'Estimated number of molecules in the a.s.u.: ' + `mstat.est_no_of_mols` + '
\n'
line += '
\n'
line += 'Number of generated models to be used in MR: ' + `mstat.MRNUM` + '
\n'
line += 'Number of generated models to be used in each Ensemble: ' + `mstat.ENSMODNUM` + '
\n'
line += 'Molecular Replacement program to be used: ' + mstat.MRPROGRAM + '
\n'
if init.keywords.PACK == None:
line += 'Number of clashes to be tolerated in Phaser: %d
\n' % int(init.keywords.PACK)
else:
line += 'Number of clashes to be tolerated in Phaser: %d
\n' % (int(init.keywords.PACK)*int(init.keywords.NMASU))
else:
# Report an error otherwise
line += 'Error in input data
\n'
line += 'Error report: ' + mstat.InitError[1] + '
\n'
mstat.setJob_End(True)
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_SEQ_search_details(self, init, mstat, results_html):
""" A function to write the search results from the SEQ search in the results html page. """
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Template Search
\n'
line += 'Sequence-based Search Results (SEQ):
\n'
if mstat.ignore_list != []:
codes_line=string.join(mstat.ignore_list,', ')
line += 'PDB id codes ignored in search: ' + codes_line + '
\n'
if mstat.include_list != []:
codes_line=string.join(mstat.include_list,', ')
line += 'Chain id codes included in search: ' + codes_line + '
\n'
if mstat.no_of_SEQhits != 0:
line += 'Number of hits from sequence-based search: ' + `mstat.no_of_SEQhits` + '
\n'
line += 'E value used in sequence-based (SEQ) search: %.2e
\n' % mstat.E_value
line += 'To view SEQ results file click here
\n'
else:
line += 'No suitable homologues found in the sequence-based search
\n'
line += 'Try a higher e value
\n'
line += 'To view SEQ results file click here
\n'
mstat.setJob_End(True)
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_SSM_search_details(self, init, mstat, results_html):
""" A function to write the search results from the SSM search in the results html page. """
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Structure-based Search Results (SSM):
\n'
line += 'Number of new hits from SSM search: ' + `mstat.no_of_SSMhits` + '
\n'
line += 'To view SSM results file (XML) click here
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_ALL_search_details(self, init, mstat, target_info, results_html):
""" A function to write a summary of all the search results in the results html page. """
# Write the scores file for gnuplot
self.print_scores_file(mstat)
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Overall Search Results:
\n'
line += 'Top match from search: ' + mstat.top_match_chainid + '
\n'
line += 'Top match percentage sequence identity: %.3f
\n' % (mstat.chain_list[mstat.top_match_chainid].seqID*100.0)
line += 'To view the multiple alignment results click here
\n'
# Plot the sequence alignment scores
self.plot_scores(target_info, init, mstat, 'AlignScore')
line += '
\n'
line += 'Alignment Score:
\n'
line += '
\n'
line += ' \n \n\n'
line += ' \n'
line += ' | \n'
self.make_logfiles(mstat, 'AlignScore')
score_table = self.compose_table(init, mstat, 'AlignScore')
line += '\n'
line += score_table
line += ' | \n'
line += '
\n'
# Plot the sequence identity scores
self.plot_scores(target_info, init, mstat, 'seqID')
line += '
\n'
line += 'Sequence Identity scores:
\n'
line += '
\n'
line += ' \n \n\n'
line += ' \n'
line += ' | \n'
self.make_logfiles(mstat, 'seqID')
score_table = self.compose_table(init, mstat, 'seqID')
line += '\n'
line += score_table
line += ' | \n'
line += '
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_domains_search_details(self, init, mstat, results_html):
""" A function to write a summary of the domain search results in the results html page. """
# Write the SCOP results file
self.make_SCOP_results_file(mstat)
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Domains Search Results (SCOP):
\n'
line += 'Total number of domains found: ' + `mstat.no_of_SCOPhits` + '
\n'
line += 'To view the SCOP results file click here
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_model_prep_details(self, init, mstat, target_info, results_html):
""" A function to write the model preparation details into the results html page. """
# Get the scores for the Model preparation step
prep_scores=Model_prep_scoring.Model_Prep_Scores()
if init.keywords.MDLMOLREP:
prep_scores.molrep_score(init, mstat)
if init.keywords.MDLCHAINSAW:
prep_scores.chainsaw_score(init, mstat,'multiple')
# Re-write the scores file for gnuplot with Molrep and Chainsaw scores included
self.print_scores_file(mstat)
# Output to the HTML results file if required
if init.keywords.HTMLOUT:
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Model Preparation
\n'
line += 'Number of Modelling jobs spwaned: ' + `mstat.model_counter` + '
\n'
# Molrep pruning scores
if init.keywords.MDLMOLREP:
self.plot_scores(target_info, init, mstat, 'Molrep_Score')
line += '
\n'
line += 'Molrep scores:
\n'
line += "These are the results for correcting each of the model structures to match the target sequence using Molrep
\n"
line += '
\n'
line += ' \n \n\n'
line += ' \n'
line += ' | \n'
score_table = self.compose_table(init, mstat, 'Molrep')
line += '\n'
line += score_table
line += ' | \n'
line += '
\n'
# Chainsaw Score
if init.keywords.MDLCHAINSAW:
self.plot_scores(target_info, init, mstat, 'Chainsaw_Score')
line += '
\n'
line += 'Chainsaw scores:
\n'
line += "These are the results for manipulating the model structures using Chainsaw. \
The graph shows the % of the model structure conserved by Chainsaw.
\n"
line += '
\n'
line += ' \n \n\n'
line += ' \n'
line += ' | \n'
score_table = self.compose_table(init, mstat, 'Chainsaw')
line += '\n'
line += score_table
line += ' | \n'
line += '
\n'
line += '
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_PQS_search_details(self, init, mstat, target_info, results_html):
""" A function to write the PQS search results the results html page. """
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Protein Quaternary Structure Search (PQS):
\n'
line += 'Estimated number of molecules in the a.s.u.: ' + `target_info.no_of_mols` + '
\n'
line += 'Number of hits in PQS search: ' + `mstat.no_of_PQShits` + '
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def write_ensemble_details(self, init, mstat, target_info, results_html):
""" A function to write the PQS search results the results html page. """
# Remove the footer from the resuls page
self.remove_footer(mstat, results_html)
line = '
\n'
line += 'Preparation of an ensemble for Phaser:
\n'
line += 'Number of template models in each Ensemble: ' + `mstat.ENSMODNUM` + '
\n'
# Append the details onto the results.html file
res_html = open(results_html, "a")
res_html.write(line)
res_html.close()
# Add the footer to the results page
self.write_footer(init, mstat, results_html)
def split_seq(self, target_info):
""" A function to parse the sequence for display on the results page. """
seq=target_info.sequence
if len(seq) > self.elem_size:
remainder=len(seq)%self.elem_size
self.no_elems=len(seq)/self.elem_size + 1
for i in range(self.no_elems - 1):
self.seq_array.append(seq[i*self.elem_size:(i+1)*self.elem_size])
self.seq_array.append(seq[len(seq)-remainder:len(seq)])
else:
self.no_elems=1
self.seq_array.append(seq)
def plot_scores(self, target_info, init, mstat, plot_type):
""" Plot the sequence identity for matching chains """
gp=MRBUMP_Gnuplot.Gnuplot()
gp.setPlotFile(init.search_dir + "/logs/temp.plt")
gp.setSetLine("set xrange [1:*]")
# Plot sequence identity
if plot_type == 'seqID':
gp.setTitle('Sequence identity scores for target: ' + target_info.chainID) # (optional)
gp.setXLabel('models')
gp.setYLabel('% identity')
gp.gnuplot_png(mstat.scores_file, [1], [4], ["seqID"], mstat.results_dir + '/images/p_id.png')
# Plot Alignment Score
if plot_type == 'AlignScore':
gp.setTitle('Alignment scores for matches to target: ' + target_info.chainID) # (optional)
gp.setXLabel('models')
gp.setYLabel('Alignment score')
gp.gnuplot_png(mstat.scores_file, [1], [3], ["score"], mstat.results_dir + '/images/align_scores.png')
# Plot Alignment Length Quality
if plot_type == 'AlignLenScore':
gp.setTitle('Alignment score for matches to target: ' + target_info.chainID) # (optional)
gp.setXLabel('models')
gp.setYLabel('align score')
gp.gnuplot_png(mstat.scores_file, [1], [5], ["score"], mstat.results_dir + '/images/align_len_score.png')
# Plot Molrep Score
if plot_type == 'Molrep_Score':
gp.setTitle('Molrep scores for matches to target: ' + target_info.chainID) # (optional)
gp.setXLabel('models')
gp.setYLabel('Molrep Score')
gp.gnuplot_png(mstat.scores_file, [1], [6], ["score"], mstat.results_dir + '/images/molrep_scores.png')
# Plot Chainsaw Scores
if plot_type == 'Chainsaw_Score':
gp.setTitle('Chainsaw scores for matches to target: ' + target_info.chainID) # (optional)
gp.setXLabel('models')
gp.setYLabel('% of structure conserved by Chainsaw')
gp.gnuplot_png(mstat.scores_file, [1], [7], ["score (multiple alignment)"], mstat.results_dir \
+ '/images/chainsaw_scores.png')
def make_logfiles(self, mstat, plot_name):
""" A function to make up a logfiles for the alignments of each of the sequences. """
# Keep a count of the order of each template model
count=0
max_count=17
# Copy the individual alignment files and add in the alignment details to the file
for chain in mstat.chain_list.keys():
shutil.copy(mstat.chain_list[chain].alignment.out_fasta_file, mstat.results_dir + '/data/sequences/')
line = "\n"
line += "Alignment score: %.3f" % mstat.chain_list[chain].alignment.score + "\n"
line += "Sequence identity: %.3f" % mstat.chain_list[chain].alignment.seqID + "\n"
line += "Alignment length: %d" % mstat.chain_list[chain].alignment.align_len + "\n"
line += "Alignment quality: %.3f" % mstat.chain_list[chain].alignment.align_len_score + "\n"
line += "Number of gaps: %d" % mstat.chain_list[chain].alignment.no_of_gaps + "\n"
line += "Number of gap characters: %d" % mstat.chain_list[chain].alignment.no_of_gap_chars + "\n"
line += "Gap penalty: %.3f" % mstat.chain_list[chain].alignment.gap_penalty + "\n"
line += "Number of conserved residues: %d" % mstat.chain_list[chain].alignment.no_conserved + "\n"
file = open(mstat.results_dir + '/data/sequences/' + chain + '_aln.fasta','a')
file.write(line)
file.close()
def compose_table(self, init, mstat, plot_name):
""" A function to make up a table of results for each plot. """
# Keep a count of the order of each template model
count=0
max_count=17
# Set upt the headings for the table of results
table = '\n'
table += '\n'
if plot_name == 'Chainsaw':
table += '| Scores (sorted by alignment score) | \n'
table += 'Log file (ma) | \n'
else:
table += 'Top Scores (sorted by alignment score) | \n'
table += 'Source | \n'
table += 'Log file | \n'
table += '
\n'
# Populate the scores and logfile tables for each template model
for chain in mstat.sorted_MR_list:
# 1) Sequence identity
if plot_name == 'seqID':
score = '%.3f %%' % (mstat.chain_list[chain].seqID * 100.0)
source = '%s' % mstat.chain_list[chain].source
logfile=init.relative_link + 'data/sequences/' + chain + '_aln.fasta'
# 2) Alignment length score (alignment quality)
elif plot_name == 'AlignLenScore':
score = '%.3f' % mstat.chain_list[chain].alignment.align_len_score
source = '%s' % mstat.chain_list[chain].source
logfile=init.relative_link + 'data/sequences/' + chain + '_aln.fasta'
# 3) Alignment score
elif plot_name == 'AlignScore':
score = '%.3f' % mstat.chain_list[chain].alignment.score
source = '%s' % mstat.chain_list[chain].source
logfile=init.relative_link + 'data/sequences/' + chain + '_aln.fasta'
# 4) Molrep score
elif plot_name == 'Molrep':
score = '%.3f' % mstat.chain_list[chain].molrep_score
source = '%s' % mstat.chain_list[chain].source
#os.system('cp ' + mstat.chain_list[chain].molrep_logfile + ' ' + mstat.results_dir + '/log_files/')
shutil.copy(mstat.chain_list[chain].molrep_logfile, mstat.results_dir + '/log_files/')
logfile=init.relative_link + 'log_files/MODEL_MOLREP_' + mstat.chain_list[chain].chainName + '.log'
# 5) Chainsaw scores
elif plot_name == 'Chainsaw':
score_m = '%.3f' % mstat.chain_list[chain].chainsaw_score
#os.system('cp ' + mstat.chain_list[chain].chainsaw_logfile + ' ' + mstat.results_dir + '/log_files/')
shutil.copy(mstat.chain_list[chain].chainsaw_logfile, mstat.results_dir + '/log_files/')
logfile_m=init.relative_link + 'log_files/CHAINSAW_' + mstat.chain_list[chain].chainName + '.log'
# Fill in the values into the table
table += '\n'
table += '| ' + `count+1` + ' | \n'
table += ' ' + mstat.chain_list[chain].chainName + ' | \n'
# Chainsaw is a special case as it has two sets of results
if plot_name == 'Chainsaw':
table += '' + score_m + ' | \n'
table += ' Log file | \n'
table += '
\n'
else:
table += '' + score + ' | \n'
table += '' + source + ' | \n'
table += ' Log file | \n'
table += '\n'
# Increment the template model count
count=count + 1
if count > max_count:
break
# Close the table
table += '
\n'
return table
def print_scores_file(self, mstat):
""" A function to print out all of the scores from the searches and model preparation. """
# Setup a count of the templates
count = 1
# Set the scores file
outfile = mstat.results_dir + '/data/scores.dat'
mstat.setScoresFile(outfile)
# Write out a table of several scores to the file
o=open(outfile, 'w')
for chain in mstat.sorted_MR_list:
CHAIN=mstat.chain_list[chain]
o.write("%d \t %s \t %.3f \t %.3f \t %.3f \t %.3f \t %.3f \t %s\n" % \
(count, \
CHAIN.chainName, CHAIN.alignment.score, CHAIN.alignment.seqID, \
CHAIN.alignment.align_len_score, CHAIN.molrep_score, \
CHAIN.chainsaw_score, CHAIN.source))
count = count + 1
o.close()
def make_SCOP_results_file(self, mstat):
""" A function to create a file containing the SCOP results. """
# Make the title for the SCOP results file
line = "SCOP domains search results\n\n"
line += "---------------------------------------------------------------------\n"
line += "CHAIN_ID \t No. of Domains \t List of Domains \n"
line += "---------------------------------------------------------------------\n"
# Loop over all chains (except SCOP chains) and print the domains details
for chain in mstat.sorted_list:
CHAIN=mstat.chain_list[chain[0]]
if CHAIN.source != "SCOP":
line += "%s \t\t %d \t\t " % (chain[0], CHAIN.no_of_domains)
count = 0
for domain in CHAIN.domain_list:
count = count + 1
if count < CHAIN.no_of_domains:
line += "%s, " % domain
else:
line += "%s" % domain
line += "\n"
#' Write the results the SCOP results file
mstat.setSCOPResultsFile(mstat.results_dir + '/data/scop_results.txt')
out_file=open(mstat.SCOP_results_file, 'w')
out_file.write(line)
out_file.close
if __name__ == '__main__':
w=write_html()
w.write_header(init, "res.html")
len_footer = w.write_footer(init, "res.html")