hydrogen_bondifier 0.0.2

data/lib/pymol.rb

@@ -0,0 +1,93 @@
+require 'tempfile'
+
+
+class Pymol
+
+  TMP_PYTHON_SCRIPT_FILENAME = "python_script_for_pymol"
+  TMP_PYMOL_SCRIPT_FILENAME = "pymol_script"
+  TMP_PYMOL_REPLY_FILENAME = "pymol_reply"
+
+  PYMOL_SCRIPT_POSTFIX = '.pml'  # <- very freaking important
+  PYTHON_SCRIPT_POSTFIX = '.py'  # <- very freaking important
+
+  attr_accessor :cmds
+
+  def self.run(opt={}, &block)
+    self.new.run(opt, &block)
+  end
+
+  def initialize
+    @cmds = [] 
+  end
+
+  def cmd(string)
+    @cmds << string
+  end
+
+  def tmp_file(base, postfix)
+    [base, $$, Time.now.to_f].join('-') << postfix
+  end
+
+  # you can add your path to pymol to this array if you need to 
+  # or with a commandline flag
+  PYMOL_EXE_TO_TRY = ['PyMOL.exe', 'pymol']
+
+  ####################################################
+  # determine if we have pymol and how to execute it
+  ####################################################
+  PYMOL_EXE_TO_TRY.unshift(ENV['PYMOL_EXE']) if ENV['PYMOL_EXE']
+  to_use = false
+  PYMOL_EXE_TO_TRY.each do |name|
+    begin
+      _cmd = "#{name} -cq"
+      if system(_cmd)
+        to_use = _cmd
+        break
+      end
+    rescue
+    end
+  end
+
+  if to_use
+    PYMOL_QUIET = to_use
+    puts "pymol executable looks good: '#{PYMOL_QUIET}'" if $VERBOSE
+  else
+    abort "pymol not installed or can't find path, specify with --path-to-pymol" if !to_use
+  end
+
+  def run(opt={}, &block)
+    puts( "[working in pymol]: " + opt[:msg] + " ...") if (opt[:msg] && $VERBOSE)
+
+    block.call(self)
+
+    cmds_to_run = self.cmds.map
+
+    tmpfiles = []
+    if script = opt[:script]
+      script_tmpfn = tmp_file TMP_PYTHON_SCRIPT_FILENAME, PYTHON_SCRIPT_POSTFIX
+      File.open(script_tmpfn, 'w') {|out| out.print script }
+      tmpfiles << script_tmpfn
+      cmds_to_run.unshift( "run #{script_tmpfn}" )
+    end
+
+    to_run = cmds_to_run.map {|v| v + "\n" }.join
+
+    pymol_tmpfn = tmp_file TMP_PYMOL_SCRIPT_FILENAME, PYMOL_SCRIPT_POSTFIX
+    File.open(pymol_tmpfn, 'w') {|out| out.print to_run }
+    tmpfiles << pymol_tmpfn
+
+    # much more suave to open a pipe, but python does not play well with pipes
+    # and this is *MUCH* more compatible with windows
+    pymol_reply_tmpfn = tmp_file TMP_PYMOL_REPLY_FILENAME, '.pmolreply'
+    tmpfiles << pymol_reply_tmpfn
+    system_cmd = "#{PYMOL_QUIET} #{pymol_tmpfn} > #{pymol_reply_tmpfn}"
+    system system_cmd
+
+    reply = IO.read(pymol_reply_tmpfn)
+    
+    tmpfiles.each {|tmpf| File.unlink tmpf } 
+    self.cmds.clear
+    reply
+  end
+
+end

data/reference/all_connections.py

@@ -0,0 +1,24 @@
+
+from pymol import cmd
+
+def all_connections(selection):
+  """
+  USAGE
+
+  all_connections selection
+
+  returns lines: "CONNECTION: id id"
+  """
+  stored.xs = []
+  cmd.iterate(selection, 'stored.xs.append( index )')
+  #outit = open("all_connections.output.tmp", 'w')
+  for i in stored.xs:
+    selName = "neighbor%s" % i
+    ids = cmd.select(selName, ("%s and neighbor id %s" % (selection, i)))
+    base = "CONNECTION: %s - " % i 
+    to_print = base + "%s"
+    print_string = 'print "' + to_print + '" % index'
+    cmd.iterate(selName, print_string )
+
+cmd.extend("all_connections", all_connections)
+

data/reference/campbell_find_hb.py

@@ -0,0 +1,18 @@
+
+# http://www.mail-archive.com/pymol-users@lists.sourceforge.net/msg06211.html
+
+from pymol import cmd
+
+def print_hb(selection):
+  hb = cmd.find_pairs("((byres "+selection+") and n;n)","((byres "+selection+") and n;o)",mode=1,cutoff=3.7,angle=55)
+
+  pair1_list = []
+  pair2_list = []
+  dist_list = []
+  for pairs in hb:
+    cmd.iterate("%s and ID %s" % (pairs[0][0],pairs[0][1]), 'print "%s/%3s`%s/%s " % (chain,resn,resi,name),')
+    cmd.iterate("%s and ID %s" % (pairs[1][0],pairs[1][1]), 'print "%s/%3s`%s/%s " % (chain,resn,resi,name),')
+    print "%.2f" % cmd.dist("%s and ID %s" % (pairs[0][0],pairs[0][1]),"%s and ID %s" % (pairs[1][0],pairs[1][1]))
+
+cmd.extend("print_hb",print_hb)
+

data/reference/list_hbonds.py

@@ -0,0 +1,47 @@
+
+# list_hb <selection>, cutoff=3.2 (default)
+
+# exclude waters first:
+# list_hb 1xnb &! r. hoh, cutoff=3.2
+
+# from: ./pymol/trunk/pymol/layer3/Selector.c
+#
+#  if(mode == 1) {
+#    angle_cutoff = (float) cos(PI * h_angle / 180.8);
+#  }
+#             ##   cos(PI * 55 / 180.8)  => 0.577  # just converts to radians and takes the cosine
+#             ##   so, they take the cosine of the angle as the "angle_cutoff"
+
+# if(mode == 1) {   /* coarse hydrogen bonding assessment */
+#             ##   so, mode 1 is hydrogen bonding finding.
+#                flag = false;
+#                if(ObjectMoleculeGetAvgHBondVector(obj1, at1, state1, v1, NULL) > 0.3)
+#                  if(dot_product3f(v1, dir) < -angle_cutoff)
+#                    flag = true;
+#                if(ObjectMoleculeGetAvgHBondVector(obj2, at2, state2, v2, NULL) > 0.3)
+#                  if(dot_product3f(v2, dir) > angle_cutoff)
+#                    flag = true;
+
+from pymol import cmd
+
+def list_hb(selection,cutoff=3.2,angle=55,hb_list_name='hbonds'):
+  """
+  USAGE
+
+  list_hb selection, [cutoff (default=3.2)], [angle (default=55)], [hb_list_name]
+  
+  e.g.
+    list_hb 1abc & c. a &! r. hoh, cutoff=3.2, hb_list_name=abc-hbonds
+  """
+  cutoff=float(cutoff)
+  angle=float(angle)
+  hb = cmd.find_pairs("((byres "+selection+") and n;n)","((byres "+selection+") and n;o)",mode=1,cutoff=cutoff,angle=angle)
+# sort the list for easier reading
+  hb.sort(lambda x,y:(cmp(x[0][1],y[0][1])))
+
+  for pairs in hb:
+    for ind in [0,1]:
+      cmd.iterate("%s and index %s" % (pairs[ind][0],pairs[ind][1]), 'print "%s/%3s`%s/%s/%s " % (chain,resn,resi,name,index),')
+    print "%.2f" % cmd.distance(hb_list_name,"%s and index %s" % (pairs[0][0],pairs[0][1]),"%s and index %s" % (pairs[1][0],pairs[1][1]))
+
+cmd.extend("list_hb",list_hb)

data/reference/test_pymol.rb

@@ -0,0 +1,125 @@
+
+require 'open3'
+
+cmd = 'pymol -cq -p'
+
+#output = IO.popen("pymol -cq -p", 'w+') do |pipe|
+  #reader = Thread.new { pipe.to_a }
+
+  #pipe.puts "load 2pERK2_Hadded.pdb, mymodel"
+  #pipe.puts "run all_connections.py"
+  ## this command will generate a whole bunch of output to stdout
+  #pipe.puts "all_connections mymodel"
+
+  #pipe.close_write
+  #reader.value
+#end
+
+#p output
+
+
+my_string = ""
+Open3.popen3("pymol -cq -p") do |si, so, se|
+  si.puts "load 2pERK2_Hadded.pdb, mymodel\n"
+  si.puts "run all_connections.py\n"
+  si.puts "all_connections mymodel\n"
+
+  # await input for 0.5 seconds, will return nil and
+  # break the loop if there is nothing to read from so after 0.5s
+  while ready = IO.select([so], nil, nil, 0.5)
+    # ready.first == so # in this case
+
+    # read until the current pipe buffer is empty
+    begin
+      my_string << so.read_nonblock(4096)
+    rescue Errno::EAGAIN
+      break
+    end while true
+  end
+end 
+p my_string
+
+
+
+#IO.popen("pymol -cq -p", 'w+') do |pipe|
+#pipe.puts "load 2pERK2_Hadded.pdb, mymodel\n"
+#pipe.puts "run all_connections.py"
+#pipe.puts "all_connections mymodel"
+#pipe.puts "quit"
+
+#IO.select([pipe]) # you shouldn't need to uncomment this...
+
+#output = pipe.read
+#end 
+#p output
+
+
+#my_string = ""
+#Open3.popen3(cmd) do |si, so, se|
+#si.puts "run all_connections.py\n"
+#si.puts "load 2pERK2_Hadded.pdb, mdl\n"
+##si.puts "show surface, mdl\n"
+#si.puts "all_connections mdl\n"
+#si.puts "print \"DONE\""
+#stringsz = -1
+
+#forstdout = Thread.new do
+#Thread.current['lines'] = ""
+#loop do
+#Thread.current['lines'] << so.read(2024)
+#end
+#end
+#past_size = -1
+#loop do
+#sleep(0.7)
+#csize = forstdout['lines'].size
+#break if csize == past_size
+#past_size = csize
+#end
+#my_string = forstdout['lines']
+#forstdout.kill
+#si.close
+#end
+#p my_string.size
+
+
+
+
+=begin
+IO.popen(cmd, 'w') do |pipe|
+  pipe.puts "run all_connections.py\n"
+  pipe.puts "load 2pERK2_Hadded.pdb, mdl\n"
+  #pipe.puts "show surface, mdl\n"
+  pipe.puts "all_connections mdl\n"
+  pipe.puts "quit\n"
+  can_do = pipe.fsync
+  puts can_do
+  pipe.close_write
+  puts pipe.read
+end
+=end
+
+
+
+=begin
+
+cmd = 'pymol -cq -p'
+
+my_string = ""
+Open3.popen3(cmd) do |si, so, se|
+  si.puts "run all_connections.py\n"
+  si.puts "load 2pERK2_Hadded.pdb, mdl\n"
+  si.puts "all_connections mdl\n"
+  loop do
+    if File.exist?("all_connections.output.tmp")
+      break
+    end
+  end
+  si.close
+end
+p my_string.size
+
+
+
+
+=end

data/spec/pymol/connections_spec.rb

@@ -0,0 +1,25 @@
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+
+require 'fileutils'
+require 'pymol/connections'
+
+describe 'pymol connections' do
+
+  before do
+    @file = 'little.pdb'
+    origfile = TESTFILES + '/' + @file
+    FileUtils.copy origfile, @file
+  end
+
+  after do
+    File.unlink @file if File.exist?(@file)
+  end
+
+  it 'finds all connections in the molecule' do
+    id_pairs = Pymol::Connections.from_pdb(@file)
+    id_pairs[0,2].enums [[1, 2], [1, 6]]
+    id_pairs[-2,2].enums [[204, 209], [205, 206]]
+    id_pairs.size.is 210
+  end
+
+end

data/spec/pymol/hydrogen_bonds_spec.rb

@@ -0,0 +1,38 @@
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+
+require 'fileutils'
+require 'pymol/hydrogen_bonds'
+
+describe 'pymol hydrogen bonds' do
+
+  before do
+    @file = 'little.pdb'
+    origfile = TESTFILES + '/' + @file
+    FileUtils.copy origfile, @file
+  end
+
+  after do
+    File.unlink @file if File.exist?(@file)
+  end
+
+  it 'finds hydrogen bonds' do
+    triplets = Pymol::HydrogenBonds.from_pdb(@file)
+
+    answers_but_atoms_as_ids = [[75, 82, 125, 2.70110007356142, 2.87, 1.9000997342245],
+      [122, 131, 78, 2.65823167164358, 2.94, 1.98408492761777],
+      [155, 166, 46, 2.77786431401053, 2.89, 1.90544535476618],
+      [187, 194, 158, 2.19624439858352, 2.86, 2.13271728084151]]
+
+    triplets.zip(answers_but_atoms_as_ids) do |row1, row2|
+      row1[0,3] = row1[0,3].map(&:serial)
+      row1.zip(row2) do |v1, v2| 
+        if v1.is_a? Integer
+          v1.is v2
+        else
+          v1.should.be.close v2, 0.00001
+        end
+      end
+    end
+  end
+
+end

data/spec/pymol/surface_spec.rb

@@ -0,0 +1,47 @@
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+
+require 'fileutils'
+require 'pymol/surface'
+
+class Object
+  def enums_close(other, delta)
+    self_ar = []
+    self.each do |v|
+      self_ar << v
+    end
+    other_ar = []
+    other.each do |v|
+      other_ar << v
+    end
+    self_ar.zip(other_ar) do |a,b|
+      if a.is_a? Enumerable
+        a.enums_close b, delta
+      else
+        a.should.be.close b, delta
+      end
+    end
+  end
+end
+
+describe 'pymol surface' do
+
+  before do
+    @file = 'little.pdb'
+    origfile = TESTFILES + '/' + @file
+    FileUtils.copy origfile, @file
+  end
+
+  after do
+    File.unlink @file if File.exist?(@file)
+  end
+
+  it 'finds surface coordinates' do
+    coords = Pymol::Surface.from_pdb(@file)
+    first_two = [[9.083281, 60.507313, 9.422514], [9.540758, 60.98122, 9.483328]]
+    last_two = [[7.323, 47.034184, 6.684179], [5.81, 47.229004, 6.411]]
+    coords[0,2].enums_close first_two, 0.0001
+    coords[-2,2].enums_close last_two, 0.0001
+    coords.size.is 9684
+  end
+
+end

data/spec/pymol_spec.rb

@@ -0,0 +1,41 @@
+require File.expand_path(File.dirname(__FILE__) + '/spec_helper')
+
+require 'fileutils'
+require 'pymol'
+
+describe 'pymol running basic tests' do
+
+  before do
+    @file = 'little.pdb'
+    origfile = TESTFILES + '/' + @file
+    FileUtils.copy origfile, @file
+  end
+
+  after do
+    File.unlink @file if File.exist?(@file)
+    File.unlink @file if File.exist?(@file)
+  end
+
+  it 'can run commands and wait for a file to be written' do
+    newfile = @file.sub(/\.pdb/i, ".HADDED.pdb")
+    reply = Pymol.run do |p|
+      p.cmd "load #{@file}, mdl" 
+      p.cmd "h_add"
+      p.cmd "save #{newfile}"
+    end
+    ok File.exist?(newfile)
+    fs = File.size(newfile)
+    ok ((fs == 17014) or (fs == 16805))  # linux / windows filesizes
+    File.unlink(newfile) if File.exist?(newfile)
+  end
+
+  it 'can run commands and wait for stdout output' do
+    reply = Pymol.run do |p|
+      p.cmd "load #{@file}, mdl" 
+      p.cmd "select mysel, mdl and elem o"
+      p.cmd 'iterate mysel, print "GRAB: %s" % index'
+    end
+    reply.split("\n").select {|line| line =~ /^GRAB: / }.size.is 20
+  end
+
+end

data/spec/scripts/confirm_angle_and_h_dist.rb

@@ -0,0 +1,126 @@
+require 'gnuplot'
+
+require 'runarray/more'
+require 'narray'
+require 'rsruby'
+
+
+class NArray
+  include Runarray::More
+  alias_method :avg, :mean
+end
+
+if ARGV.size != 2
+  puts "usage: #{File.basename(__FILE__)} <file>_hbonds.csv <file>_HBOND.out"
+  exit
+end
+
+(mine, theirs) = ARGV
+
+lines = IO.readlines(mine)
+lines.shift
+
+my_angle_dist_by_res_and_name = {}
+lines.each do |line|
+  pieces = line.split(',')
+  key = [pieces[5], pieces[3], pieces[8], pieces[9]]
+  my_angle_dist_by_res_and_name[key] = [pieces[10].to_f, pieces[12].to_f]
+end
+# ["22", "HE", "42", "OD1"] == ["22", "HE", "42", "OD1"]
+#["43", "1HD2", "24", "OG1"]
+#
+#["48", "1HH1", "25", "O"]
+#["48", "2HH2", "25", "O"]
+#["48", "2HH2", "39", "OG"]
+#
+#["48", "HH21", "39", "OG"]
+
+
+
+lines = []
+get_lines = false
+IO.foreach(theirs) do |line|
+  if get_lines
+    lines << line
+  end
+  if line =~ /^Donor/
+    get_lines = true
+  end
+end
+
+puts "*******************************************"
+puts "SECOND"
+puts "*******************************************"
+
+res_to_vals = {}
+lines.each do |v| 
+  pieces = v.chomp.split(/\s+/)
+  key = pieces[0].split(':')[1,2].push(*(pieces[1].split(':')[1,2]))
+  key[1] = key[1].sub(/^HN/,'H').sub(/^(\w+\d?)(\d)$/) {|md| $2 + $1 }
+  res_to_vals[key] = [pieces[3].to_f, pieces[2].to_f]
+end
+
+a_x = []
+a_y = []
+d_x = []
+d_y = []
+my_angle_dist_by_res_and_name.each do |id, angle_dist|
+  (angle, dist) = angle_dist
+  (o_angle, o_dist) = res_to_vals[id]
+  if o_angle && o_dist
+    a_x << angle 
+    a_y << o_angle 
+    d_x << dist
+    d_y << o_dist
+  end
+end
+puts "EXAMINE"
+p a_x
+p a_y
+
+intersection = a_x.size
+my_hbond_count = my_angle_dist_by_res_and_name.size
+their_hbond_count = res_to_vals.size
+
+a_rsi = NArray.to_na(a_x).rsq_slope_intercept(NArray.to_na(a_y)).map {|v| "%0.2f" % v }
+d_rsi = NArray.to_na(d_x).rsq_slope_intercept(NArray.to_na(d_y)).map {|v| "%0.2f" % v }
+
+set_analysis = [[:jtp, my_hbond_count], [:insight, their_hbond_count], [:intersection, intersection]]
+title_string = set_analysis.map {|pair| pair.join('=')}.join(', ')
+
+Gnuplot.open do |gp|
+  Gnuplot::Plot.new(gp) do |plot|
+    plot.terminal "png"
+    plot.output "jtp_vs_insight_angles.png"
+    plot.title "2pERK2_Hadded.pdb - #{title_string}"
+    plot.xlabel "[JTP] donor, H, acceptor angle (degrees)"
+    plot.ylabel "[Insight] angle (degrees)"
+
+    plot.data << Gnuplot::DataSet.new( [a_x, a_y] ) do |ds|
+      ds.title = "RSQ=#{a_rsi[0]} SLOPE=#{a_rsi[1]} INT=#{a_rsi[2]}"
+    end
+    plot.data << Gnuplot::DataSet.new( [[118,180], [118,180]] ) do |ds|
+      ds.with = "lines"
+      ds.title = "perfect"
+    end
+  end
+end
+
+Gnuplot.open do |gp|
+  Gnuplot::Plot.new(gp) do |plot|
+    plot.terminal "png"
+    plot.output "jtp_vs_insight_distances.png"
+    plot.title "2pERK2_Hadded.pdb - #{title_string}"
+    plot.xlabel "[JTP] distance (Angstrom)"
+    plot.ylabel "[Insight] distance (Angstrom)"
+
+    plot.data << Gnuplot::DataSet.new( [d_x, d_y] ) do |ds|
+      ds.title = "RSQ=#{d_rsi[0]} SLOPE=#{d_rsi[1]} INT=#{d_rsi[2]}"
+    end
+    plot.data << Gnuplot::DataSet.new( [[1,3], [1,3]] ) do |ds|
+      ds.with = "lines"
+      ds.title = "perfect"
+    end
+  end
+end
+

data/spec/scripts/confirm_distances.rb

@@ -0,0 +1,83 @@
+#!/usr/bin/ruby -s
+
+require 'rsruby'
+require 'narray'
+require 'runarray/more'
+require 'gnuplot'
+
+
+
+class NArray
+  include Runarray::More
+  alias_method :avg, :mean
+end
+
+def pairs_from_their_format(file)
+  their_pairs = IO.readlines(file).map do |line|
+    pieces = line.chomp.split(/\s+/)
+    [pieces.first[3..-1].to_i, pieces.last.to_f]
+  end
+end
+
+if ARGV.size != 3
+  puts "[-s] theirs mine <from_what>"
+  exit
+end
+(theirs, mine, keywords) = ARGV
+
+
+if $s   # mine is same format as theirs!
+  (their_pairs, my_pairs) = [theirs, mine].map {|f| pairs_from_their_format f }
+  their_index = []
+  their_pairs.each do |res, dist|
+    their_index[res] = dist
+  end
+else
+  their_pairs = pairs_from_their_format(theirs)
+
+  lines = IO.readlines(mine)
+  lines.shift
+
+  my_pairs = lines.map do |line|
+    pieces = line.split(',')
+    [pieces[4].to_i, pieces[12].to_f]
+  end
+end
+
+x = [] ; y = []
+my_pairs.each do |i, dist|
+  if their_index[i]
+    x << dist 
+    y << their_index[i]
+  end
+end
+
+intersection = x.size
+my_hbond_count = my_pairs.size
+their_hbond_count = their_index.size
+
+set_analysis = [[:jtp, my_hbond_count], [:insight, their_hbond_count], [:intersection, intersection]]
+title_string = set_analysis.map {|pair| pair.join('=')}.join(', ')
+
+
+(rsq, slope, inter) = NArray.to_na(x).rsq_slope_intercept(NArray.to_na(y)).map {|v| "%0.2f" % v }
+
+Gnuplot.open do |gp|
+  Gnuplot::Plot.new(gp) do |plot|
+    plot.terminal "png"
+    plot.output "jtp_vs_insight_surface_distances_#{keywords.gsub(/\s+/,'_')}.png"
+    plot.title "2pERK2_Hadded.pdb - #{title_string}"
+    plot.xlabel "[JTP] distance from #{keywords} to surface (Angstrom)"
+    plot.ylabel "[Insight] distance from residue to surface (Angstrom)"
+    plot.data << Gnuplot::DataSet.new( [x, y] ) do |ds|
+      ds.title = "RSQ=#{rsq} SLOPE=#{slope} INT=#{inter}"
+    end
+    plot.data << Gnuplot::DataSet.new( [[1,7], [1,7]] ) do |ds|
+      ds.with = "lines"
+      ds.title = "perfect"
+    end
+  end
+end
+
+
+

data/spec/scripts/mins.rb

@@ -0,0 +1,9 @@
+lines = nil
+IO.foreach(ARGV.shift) do |line|
+  lines.push(line) if lines
+  if line =~ /^Donor/
+    lines = []
+  end
+end
+
+p (180.0 - lines.map {|line| line.chomp.split(/\s+/)[3].to_f }.select {|v| v > 0.0 }.min)