egor 0.0.1 → 0.0.2

data/History.txt

@@ -1,3 +1,9 @@
+== 0.0.2 2008-11-13
+
+* 2 major enhancement:
+  * Constrained environment features are properly handled
+  * Sane log-odds ratio matrices are produced
+
 == 0.0.1 2008-11-07
 
 * 1 major enhancement:

data/README.rdoc

@@ -9,8 +9,10 @@ egor: Esst GeneratOR, a program for calculating environment-specific substitutio
 == FEATURES/PROBLEMS:
 
 * No more segmentation fault
+* Fast enough not to leave your place
+* Slow enough to check your emails or have some chats with your colleagues next you
 * Full smoothing supported
-* Infinite number of environment features can be handled (in theory)
+* In theory, infinite number of environment features can be handled
 
 == BASIC USAGE:
 

data/lib/egor/cli.rb

@@ -41,19 +41,19 @@ Usage:
     egor [ options ] -f TEM-file -c CLASSDEF-file
 
 Options:
-    --tem-file (-f) STRING: a tem file
-    --tem-list (-l) STRING: a list for tem files
-    --classdef (-c) STRING: a file for the defintion of environments (default: 'classdef.dat')
-    --outfile (-o) STRING: output filename ("allmat.dat" if not specified)
+    --tem-file (-f) FILE: a tem file
+    --tem-list (-l) FILE: a list for tem files
+    --classdef (-c) FILE: a file for the defintion of environments (default: 'classdef.dat')
+    --outfile (-o) FILE: output filename ("allmat.dat" if not specified)
     --weight (-w) INTEGER: clustering level (PID) for the BLOSUM-like weighting
     --noweight: calculate substitution counts with no weights (default)
     --smooth (-s) INTEGER:
         0 for parial smoothing (default)
         1 for full smoothing
     --nosmooth: perform no smoothing operation
-    --cys (-y) INTEGER: (NOT implemented yet)
-        0 for using C and J only for structure
-        1 for both structure and sequence (default)
+    --cys (-y) INTEGER:
+        0 for using C and J only for structure (default)
+        1 for both structure and sequence
     --output INTEGER:
         0 for raw counts (no-smoothing performed)
         1 for probabilities
@@ -61,7 +61,7 @@ Options:
     --scale INTEGER: log-odds matrices in 1/n bit units (default 3)
     --sigma DOUBLE: change the sigma value for smoothing (default 5)
     --add DOUBLE: add this value to raw counts when deriving log-odds without smoothing (default 1/#classes)
-    --penv: use environment-dependent frequencies for log-odds calculation (default false) (NOT implemented yet)
+    --penv: use environment-dependent frequencies for log-odds calculation (default false) (NOT implemented yet!!!)
     --pidmin DOUBLE: count substitutions only for pairs with PID equal to or greater than this value (default none)
     --pidmax DOUBLE: count substitutions only for pairs with PID smaller than this value (default none)
     --verbose (-v) INTEGER
@@ -137,12 +137,6 @@ Options:
         $outfile      = "allmat.dat"
         $outfh        = nil # file hanfle for outfile
         $output       = 2
-        $aa_tot_obs   = {}
-        $aa_mut_obs   = {}
-        $aa_mutb      = {}
-        $aa_rel_mutb  = {}
-        $aa_rel_freq  = {}
-        $env_aa_obs   = {}
         $ali_size     = 0
         $tot_aa       = 0
         $sigma        = 5.0
@@ -154,10 +148,20 @@ Options:
         $pidmin       = nil
         $pidmax       = nil
         $scale        = 3
-        $add          = 0
+        $add          = nil
+        $cys          = 0
         $penv         = false
-        $heatmap      = false
+
+        $aa_tot_obs   = {}
+        $aa_mut_obs   = {}
+        $aa_mutb      = {}
+        $aa_rel_mutb  = {}
+        $aa_rel_freq  = {}
+        $env_aa_obs   = {}
         $smooth_prob  = {}
+        $tot_freq_mat = nil
+        $tot_prob_mat = nil
+        $tot_logo_mat = nil
 
         # Part 2.
         #
@@ -195,10 +199,8 @@ Options:
             $output       = arg.to_i
           when '--outfile'
             $outfile      = arg
-          when '--cyc'
-            $logger.error "!!! --cys option is not available yet"
-            exit 1
-            $cysteine     = (arg.to_i == 1 ? false : true)
+          when '--cys'
+            $cys          = (arg.to_i == 1 ? false : true)
           when '--weight'
             $weight       = arg.to_i
           when '--sigma'
@@ -210,15 +212,17 @@ Options:
           when '--noweight'
             $noweight     = true
           when '--smooth'
-            $smooth       = (arg.to_i == 1 ? :full : :parital)
+            $smooth       = (arg.to_i == 1 ? :full : :partial)
           when '--nosmooth'
             $nosmooth     = true
           when '--scale'
             $scale        = arg.to_f
           when '--add'
+            $logger.error "!!! --add option is not supported yet"
+            exit 1
             $add          = arg.to_f
           when '--penv'
-            $logger.error "!!! --penv option is not available yet"
+            $logger.error "!!! --penv option is not supported yet"
             exit 1
             $penv         = true
           when '--heatmap'
@@ -245,14 +249,19 @@ Options:
           exit 1
         end
 
+
         # Part 3.
         #
         # Reading Environment Class Definition File
         #
 
-        # a hash for storing all environment feature objects
+        # an array for storing all environment feature objects
         $env_features = []
 
+
+        # an array for storing indexes of constrained environment features
+        $cst_features = []
+
         # aa1 amino acid in a substitution itself is a environment feature
         $env_features << EnvironmentFeature.new("sequence",
                                                 $amino_acids,
@@ -262,24 +271,29 @@ Options:
 
         # read environment class definiton file and
         # store them into the hash prepared above
+        env_index = 1
+
         IO.foreach($classdef) do |line|
+          line.chomp!
           if line.start_with?("#")
             next
           elsif (env_ftr = line.chomp.split(/;/)).length == 5
-            $logger.info ">>> An environment feature, #{line.chomp} detected"
+            $logger.info ">>> An environment feature, #{line} detected"
             if env_ftr[-1] == "T"
               # skip silenced environment feature
-              $logger.warn "!!! The environment feature, #{line.chomp} silent"
+              $logger.warn "!!! The environment feature, #{line} silent"
               next
             end
             if env_ftr[-2] == "T"
-              $logger.warn "!!! The environment feature, #{line.chomp} constrained"
+              $cst_features << env_index
+              $logger.warn "!!! The environment feature, #{line} constrained"
             end
             $env_features << EnvironmentFeature.new(env_ftr[0],
                                                     env_ftr[1].split(""),
                                                     env_ftr[2].split(""),
                                                     env_ftr[3],
                                                     env_ftr[4])
+            env_index += 1
           else
             $logger.error "@@@ #{line} doesn't seem to be a proper format for class definition"
             exit 1
@@ -325,7 +339,7 @@ Options:
               end
             end
 
-            $ali_size   += ali.size
+            $ali_size   += 1
             env_labels  = {}
             disulphide  = {}
 
@@ -398,19 +412,27 @@ Options:
                       aa2 = s2[pos].upcase
 
                       if !$amino_acids.include?(aa1)
-                        $logger.warn "!!! #{id1}-#{pos}-#{aa1} is not standard amino acid" unless aa1 == "-"
+                        $logger.warn "!!! #{id1}-#{pos}-#{aa1} is not a standard amino acid" unless aa1 == "-"
                         next
                       end
 
                       if !$amino_acids.include?(aa2)
-                        $logger.warn "!!! #{id1}-#{pos}-#{aa2} is not standard amino acid" unless aa2 == "-"
+                        $logger.warn "!!! #{id1}-#{pos}-#{aa2} is not a standard amino acid" unless aa2 == "-"
                         next
                       end
 
                       aa1 = (((disulphide[id1][pos] == "F") && (aa1 == "C")) ? "J" : aa1)
                       aa2 = (((disulphide[id2][pos] == "F") && (aa2 == "C")) ? "J" : aa2)
 
-                      $envs[env_labels[id1][pos]].add_residue_count(aa2)
+                      if $cst_features.empty?
+                        $envs[env_labels[id1][pos]].increase_residue_count(aa2)
+                      elsif (env_labels[id1][pos].split("").values_at(*$cst_features) ==
+                             env_labels[id2][pos].split("").values_at(*$cst_features))
+                        $envs[env_labels[id1][pos]].increase_residue_count(aa2)
+                      else
+                        $logger.debug "*** #{id1}-#{pos}-#{aa1} and #{id2}-#{pos}-#{aa2} have different symbols for constrained environment features each other"
+                        next
+                      end
 
                       grp_label = env_labels[id1][pos][1..-1]
 
@@ -485,6 +507,7 @@ Options:
                   cluster2.each do |id2|
                     seq1 = ali[id1].split("")
                     seq2 = ali[id2].split("")
+
                     seq1.each_with_index do |aa1, pos|
                       if env_labels[id1][pos].include?("X")
                         $logger.debug "*** Substitutions from #{id1}-#{pos}-#{aa1} were masked"
@@ -511,8 +534,17 @@ Options:
                       obs1  = 1.0 / size1
                       obs2  = 1.0 / size2
 
-                      $envs[env_labels[id1][pos]].add_residue_count(aa2, 1.0 / (size1 * size2))
-                      $envs[env_labels[id2][pos]].add_residue_count(aa1, 1.0 / (size1 * size2))
+                      if $cst_features.empty?
+                        $envs[env_labels[id1][pos]].increase_residue_count(aa2, 1.0 / (size1 * size2))
+                        $envs[env_labels[id2][pos]].increase_residue_count(aa1, 1.0 / (size1 * size2))
+                      elsif (env_labels[id1][pos].split("").values_at(*$cst_features) ==
+                             env_labels[id2][pos].split("").values_at(*$cst_features))
+                        $envs[env_labels[id1][pos]].increase_residue_count(aa2, 1.0 / (size1 * size2))
+                        $envs[env_labels[id2][pos]].increase_residue_count(aa1, 1.0 / (size1 * size2))
+                      else
+                        $logger.debug "*** #{id1}-#{pos}-#{aa1} and #{id2}-#{pos}-#{aa2} have different symbols for constrained environment features each other"
+                        next
+                      end
 
                       grp_label1 = env_labels[id1][pos][1..-1]
                       grp_label2 = env_labels[id2][pos][1..-1]
@@ -605,7 +637,6 @@ HEADER
           else
             $outfh.puts "# Weighting scheme: clustering at PID #{$weight} level"
           end
-          $outfh.puts "#"
 
           # calculate amino acid frequencies and mutabilities, and
           # print them as default statistics in the header part
@@ -614,7 +645,7 @@ HEADER
 
           $outfh.puts "#"
           $outfh.puts "# Total amino acid frequencies:\n"
-          $outfh.puts "# %-3s %9s %9s %5s %8s %8s" % %w[RES MUT_OBS TOT_OBS MUTB REL_MUTB REL_FRQ]
+          $outfh.puts "# %-3s %9s %9s %5s %8s %8s" % %w[RES TOT_OBS MUT_OBS MUTB REL_MUTB REL_FRQ]
 
           $aa_tot_obs.each_pair do |res, freq|
             $aa_mutb[res]      = $aa_mut_obs[res] / freq.to_f
@@ -625,13 +656,18 @@ HEADER
           $amino_acids.each do |res|
             if $noweight
               $outfh.puts "# %-3s %9d %9d %5.2f %8d %8.4f" %
-                [res, $aa_mut_obs[res], $aa_tot_obs[res], $aa_mutb[res], $aa_rel_mutb[res], $aa_rel_freq[res]]
+                [res, $aa_tot_obs[res], $aa_mut_obs[res], $aa_mutb[res], $aa_rel_mutb[res], $aa_rel_freq[res]]
             else
               $outfh.puts "# %-3s %9.2f %9.2f %5.2f %8d %8.4f" %
-                [res, $aa_mut_obs[res], $aa_tot_obs[res], $aa_mutb[res], $aa_rel_mutb[res], $aa_rel_freq[res]]
+                [res, $aa_tot_obs[res], $aa_mut_obs[res], $aa_mutb[res], $aa_rel_mutb[res], $aa_rel_freq[res]]
             end
           end
-          $outfh.puts "#"
+
+
+          # Part 5.
+          #
+          # Calculating substitution frequency tables
+          #
 
           # calculating probabilities for each environment
           $envs.values.each do |e|
@@ -641,7 +677,7 @@ HEADER
           end
 
           # count raw frequencies
-          $tot_freq_matrix = ($noweight ? NMatrix.int(21,21) : NMatrix.float(21,21))
+          $tot_freq_mat = ($noweight ? NMatrix.int(21,21) : NMatrix.float(21,21))
 
           # for each combination of environment features
           env_groups = $envs.values.group_by { |env| env.label[1..-1] }
@@ -652,28 +688,33 @@ HEADER
               $env_features[i + 1].labels.index(l)
             }
           }.each_with_index do |group, group_no|
-            grp_freq_matrix = ($noweight ? NMatrix.int(21,21) : NMatrix.float(21,21))
+            grp_freq_mat = ($noweight ? NMatrix.int(21,21) : NMatrix.float(21,21))
 
             $amino_acids.each_with_index do |aa, ai|
               freq_array = group[1].find { |e| e.label.start_with?(aa) }.freq_array
-              0.upto(20) { |j| grp_freq_matrix[ai, j] = freq_array[j] }
+              0.upto(20) { |j| grp_freq_mat[ai, j] = freq_array[j] }
             end
 
-            $tot_freq_matrix += grp_freq_matrix
+            $tot_freq_mat += grp_freq_mat
 
             if $output == 0
               $outfh.puts ">#{group[0]} #{group_no}"
-              $outfh.puts grp_freq_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+              $outfh.puts grp_freq_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
             end
           end
 
           if $output == 0
             $outfh.puts ">Total"
-            $outfh.puts $tot_freq_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+            $outfh.puts $tot_freq_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
             exit 0
           end
 
-          # for probability
+
+          # Part 6.
+          #
+          # Calculating substitution probability tables
+          #
+
           if $output == 1
             $outfh.puts <<HEADER
 #
@@ -687,7 +728,7 @@ HEADER
 
           if ($output > 0) && $nosmooth
             # Probability matrices
-            tot_prob_matrix = NMatrix.float(21, 21)
+            $tot_prob_mat = NMatrix.float(21, 21)
 
             # for each combination of environment features
             env_groups = $envs.values.group_by { |env| env.label[1..-1] }
@@ -697,24 +738,24 @@ HEADER
                 $env_features[i + 1].labels.index(l)
               }
             }.each_with_index do |group, group_no|
-              grp_prob_matrix = NMatrix.float(21,21)
+              grp_prob_mat = NMatrix.float(21,21)
 
               $amino_acids.each_with_index do |aa, ai|
                 prob_array = group[1].find { |e| e.label.start_with?(aa) }.prob_array
-                0.upto(20) { |j| grp_prob_matrix[ai, j] = prob_array[j] }
+                0.upto(20) { |j| grp_prob_mat[ai, j] = prob_array[j] }
               end
 
-              tot_prob_matrix += grp_prob_matrix
+              $tot_prob_mat += grp_prob_mat
 
               if ($output == 1)
                 $outfh.puts ">#{group[0]} #{group_no}"
-                $outfh.puts grp_prob_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+                $outfh.puts grp_prob_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
               end
             end
 
             if ($output == 1)
               $outfh.puts ">Total"
-              $outfh.puts tot_prob_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+              $outfh.puts $tot_prob_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
               $outfh.close
               exit 0
             end
@@ -749,6 +790,7 @@ HEADER
 
             if $smooth == :partial
               $outfh.puts <<HEADER
+#
 # Partial Smoothing:
 #
 # p1(ri) (i.e., amino acid composition) is estimated by summing over
@@ -771,7 +813,6 @@ HEADER
 # Weights (omegas) are calculated as in Topham et al. 1993)
 # 
 # sigma value used is:  5.00
-#
 HEADER
               1.upto($env_features.size) do |ci|
                 # for partial smoothing, only P1 ~ P3, and Pn are considered
@@ -872,6 +913,7 @@ HEADER
               end
             else
               $outfh.puts <<HEADER
+#
 # Full Smoothing:
 #
 # p1(ri) is estimated as:
@@ -897,7 +939,6 @@ HEADER
 # Weights (omegas) are calculated as in Topham et al. 1993)
 # 
 # sigma value used is:  5.00
-#
 HEADER
               # full smooting
               1.upto($env_features.size) do |ci|
@@ -959,7 +1000,7 @@ HEADER
             $envs.values.each { |e| e.smooth_prob_array = $smooth_prob[$env_features.size + 1][e.label_set] }
 
             # for a total substitution probability matrix
-            tot_smooth_prob_matrix = NMatrix.float(21,21)
+            $tot_prob_mat = NMatrix.float(21,21)
 
             # grouping environments by its environment labels but amino acid label
             env_groups = $envs.values.group_by { |env| env.label[1..-1] }
@@ -972,46 +1013,54 @@ HEADER
               }
             }.each_with_index do |group, group_no|
               # calculating 21X21 substitution probability matrix for each envrionment
-              grp_prob_matrix = NMatrix.float(21,21)
+              grp_prob_mat = NMatrix.float(21,21)
 
               $amino_acids.each_with_index do |aa, ai|
                 smooth_prob_array = group[1].find { |e| e.label.start_with?(aa) }.smooth_prob_array
-                0.upto(20) { |j| grp_prob_matrix[ai, j] = smooth_prob_array[j] }
+                0.upto(20) { |j| grp_prob_mat[ai, j] = smooth_prob_array[j] }
               end
 
-              tot_smooth_prob_matrix += grp_prob_matrix
+              $tot_prob_mat += grp_prob_mat
 
               if $output == 1
                 $outfh.puts ">#{group[0]} #{group_no}"
-                $outfh.puts grp_prob_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+                $outfh.puts grp_prob_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
               end
             end
 
-            tot_smooth_prob_matrix /= env_groups.size
+            $tot_prob_mat /= env_groups.size
 
             if $output == 1
               $outfh.puts ">Total"
-              $outfh.puts tot_smooth_prob_matrix.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+              $outfh.puts $tot_prob_mat.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
               $outfh.close
               exit 0
             end
 
+
+            # Part 7.
+            #
+            # Calculating log-add ratio scoring matrices
+            #
             if $output == 2
               $outfh.puts <<HEADER
 # 
 # The probabilities were then divided by the background probabilities
+HEADER
+              if $penv
+                $outfh.puts <<HEADER
 # which were derived from the environment-independent amino acid frequencies.
 #                             ^^^^^^^^^^^^^^^^^^^^^^^
-# 
-# Shown here are logarithms of these values multiplied by 3/log(2) 
-# rounded to the nearest integer (log-odds scores in 1/3 bit units).
-# 
-# For total (composite) matrix, Entropy = XXX bits, Expected score = XXX
-#
 HEADER
+              else
+                $outfh.puts <<HEADER
+# which were derived from the environment-dependent amino acid frequencies.
+#                             ^^^^^^^^^^^^^^^^^^^^^
+HEADER
+              end
 
-              # log-add ratio matrices from now on
-              tot_logo_mat  = NMatrix.float(21,21)
+              $tot_logo_mat = $cys ? NMatrix.float(21,22) : NMatrix.float(21,21)
+              grp_logo_mats = []
               factor        = $scale / Math::log(2)
 
               # grouping environments by its environment labels but amino acid label
@@ -1027,30 +1076,82 @@ HEADER
                 # calculating 21X21 substitution probability matrix for each envrionment
                 grp_label     = group[0]
                 grp_envs      = group[1]
-                grp_logo_mat  = NMatrix.float(21,21)
+                grp_logo_mat  = $cys ? NMatrix.float(21, 22) : NMatrix.float(21,21)
 
                 $amino_acids.each_with_index do |aa, ai|
                   env       = grp_envs.detect { |e| e.label.start_with?(aa) }
-                  logo_arr  = NArray.float(21)
+                  logo_arr  = $cys ? NArray.float(22) : NArray.float(21)
 
                   env.smooth_prob_array.to_a.each_with_index do |prob, j|
-                    paj = 100.0 * $aa_rel_freq[$amino_acids[j]]
-                    odds = prob == 0.0 ? 0.000001 / paj : prob / paj
+                    paj         = 100.0 * $aa_rel_freq[$amino_acids[j]]
+                    odds        = prob == 0.0 ? 0.000001 / paj : prob / paj
                     logo_arr[j] = factor * Math::log(odds)
                   end
+
                   0.upto(20) { |j| grp_logo_mat[ai, j] = logo_arr[j] }
+
+                  # adding log odds ratio for "U" (J or C) when --cyc is ON
+                  if $cys
+                    paj   = 100.0 * ($aa_rel_freq["C"] + $aa_rel_freq["J"])
+                    prob  = env.smooth_prob_array[$amino_acids.index("C")] + env.smooth_prob_array[$amino_acids.index("J")]
+                    odds  = prob == 0.0 ? 0.000001 / paj : prob / paj
+                    logo_arr[logo_arr.size - 1] = factor * Math::log(odds)
+                    grp_logo_mat[ai, logo_arr.size - 1] = logo_arr[logo_arr.size - 1]
+                  end
                 end
 
-                tot_logo_mat += grp_logo_mat
+                $tot_logo_mat += grp_logo_mat
+                grp_logo_mats << [grp_label, grp_logo_mat]
+              end
 
-                $outfh.puts ">#{grp_label} #{group_no}"
-                $outfh.puts grp_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+              $tot_logo_mat /= env_groups.size
+
+              # calculating relative entropy for each amino acid pair H and
+              # the expected score E in bit units
+              #
+              # I'm a bit suspicious about this part...
+              tot_E = 0.0
+              tot_H = 0.0
+
+              0.upto($tot_logo_mat.shape[0] - 1) do |i|
+                0.upto($tot_logo_mat.shape[0] - 1) do |j|
+                  if i != j
+                    tot_E += $tot_logo_mat[i, j] * $aa_rel_freq[$amino_acids[i]] * $aa_rel_freq[$amino_acids[j]] / 2.0
+                    tot_H += $tot_logo_mat[i, j] * $tot_prob_mat[i, j] / 2.0 / 10000.0
+                  else
+                    tot_E += $tot_logo_mat[i, j] * $aa_rel_freq[$amino_acids[i]] * $aa_rel_freq[$amino_acids[j]]
+                    tot_H += $tot_logo_mat[i, j] * $tot_prob_mat[i, j] / 10000.0
+                  end
+                end
               end
 
-              tot_logo_mat /= env_groups.size
+              $outfh.puts <<HEADER
+# 
+# Shown here are logarithms of these values multiplied by #{$scale}/log(2) 
+# rounded to the nearest integer (log-odds scores in 1/3 bit units).
+# 
+# For total (composite) matrix, Entropy = #{"%5.4f" % tot_H} bits, Expected score = #{"%5.4f" % tot_E}
+#
+HEADER
+
+              grp_logo_mats.each_with_index do |arr, grp_no|
+                grp_label     = arr[0]
+                grp_logo_mat  = arr[1]
 
-              $outfh.puts ">Total"
-              $outfh.puts tot_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+                $outfh.puts ">#{grp_label} #{grp_no}"
+                if $cys
+                  $outfh.puts grp_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids + %w[U])
+                else
+                  $outfh.puts grp_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+                end
+              end
+
+              $outfh.puts ">Total #{grp_logo_mats.size}"
+              if $cys
+                $outfh.puts $tot_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids + %w[U])
+              else
+                $outfh.puts $tot_logo_mat.round.pretty_string(:col_header => $amino_acids, :row_header => $amino_acids)
+              end
               $outfh.close
               exit 0
             end

data/lib/egor.rb

@@ -2,5 +2,5 @@ $:.unshift(File.dirname(__FILE__)) unless
   $:.include?(File.dirname(__FILE__)) || $:.include?(File.expand_path(File.dirname(__FILE__)))
 
 module Egor
-  VERSION = '0.0.1'
-end
+  VERSION = '0.0.2'
+end

data/lib/environment.rb

@@ -23,7 +23,7 @@ class Environment
     @smooth_prob_array  = NArray.float(21)
   end
 
-  def add_residue_count(a, inc = 1.0)
+  def increase_residue_count(a, inc = 1.0)
     @freq_array[@@amino_acids.index(a.upcase)] += inc
   end
 

data/lib/environment_feature.rb

@@ -1,7 +1,7 @@
 class EnvironmentFeature < Struct.new(:name, :symbols, :labels, :constrained, :silent)
 
   def to_s
-    values.join(";")
+    [name, symbols.join, labels.join, constrained, silent].join(";")
   end
 
   def constrained?

data/website/index.html

@@ -44,6 +44,8 @@
 <h2>Features</h2>
 <ul>
 	<li>No more segmentation fault</li>
+	<li>Fast enough not to leave your place</li>
+	<li>Slow enough to check your emails or have some chats with your colleagues next you</li>
 	<li>Full smoothing supported</li>
 	<li>In theory, infinite number of environment features can be handled</li>
 </ul>
@@ -53,7 +55,7 @@
 or
 <pre>$ egor -l TEM-file -c classdef.dat</pre>
 <h2>Repository</h2>
-<p>You can download a pre-built RubyGem package from</p>
+<p>You can download a pre-built RubyGems package from</p>
 <ul>
 	<li>rubyforge: <a href="http://rubyforge.org/projects/egor">http://rubyforge.org/projects/egor</a></li>
 </ul>
@@ -67,12 +69,19 @@ or
 <h2>Contact</h2>
 <p>Comments are welcome, please send an email to me (seminlee at gmail dot com).</p>
     <p class="coda">
-      <a href="FIXME email">Semin Lee</a>, 10th November 2008<br>
+      Semin Lee, 12th November 2008<br>
       Theme extended from <a href="http://rb2js.rubyforge.org/">Paul Battley</a>
     </p>
 </div>
 
 <!-- insert site tracking codes here, like Google Urchin -->
-
+<script type="text/javascript">
+  var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
+  document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
+</script>
+<script type="text/javascript">
+  var pageTracker = _gat._getTracker("UA-6291956-1");
+  pageTracker._trackPageview();
+</script>
 </body>
 </html>

data/website/index.txt

@@ -14,6 +14,8 @@ h2. Installation
 h2. Features
 
 * No more segmentation fault
+* Fast enough not to leave your place
+* Slow enough to check your emails or have some chats with your colleagues next you
 * Full smoothing supported
 * In theory, infinite number of environment features can be handled
 
@@ -29,7 +31,7 @@ It's pretty much the same as Kenji's subst, so in most cases, you just need swap
 
 h2. Repository
 
-You can download a pre-built RubyGem package from
+You can download a pre-built RubyGems package from
 
 * rubyforge: "http://rubyforge.org/projects/egor":http://rubyforge.org/projects/egor
 

data/website/template.html.erb

@@ -39,12 +39,19 @@
     </div>
     <%= body %>
     <p class="coda">
-      <a href="FIXME email">Semin Lee</a>, <%= modified.pretty %><br>
+      Semin Lee, <%= modified.pretty %><br>
       Theme extended from <a href="http://rb2js.rubyforge.org/">Paul Battley</a>
     </p>
 </div>
 
 <!-- insert site tracking codes here, like Google Urchin -->
-
+<script type="text/javascript">
+  var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
+  document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
+</script>
+<script type="text/javascript">
+  var pageTracker = _gat._getTracker("UA-6291956-1");
+  pageTracker._trackPageview();
+</script>
 </body>
 </html>

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: egor
 version: !ruby/object:Gem::Version 
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors: 
 - Semin Lee
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2008-11-10 00:00:00 +00:00
+date: 2008-11-13 00:00:00 +00:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -60,7 +60,7 @@ dependencies:
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        version: 1.0.7
+        version: 1.1.0
     version: 
 - !ruby/object:Gem::Dependency 
   name: hoe