neurohmmer 0.1.0

data/data/raw_data/vertebrate_ghrelin_motilin.fa

@@ -0,0 +1,16 @@
+>Ensembl_ENSDART00000076405_Drer
+MPLRCRASSMFLLLCVSLSLCLESVSGGTSFLSPTQKPQGRRPPRVGRREAADPEIPVIKEDDRFMMSAPFELSMSLSEAEYEKYGPVLQNLLENLLRDSSFEF
+>gi_55792853_Asch
+MFLKRNTYLLVFLFCSLTLWCKSTSAGSSFLSPSQKPQNRGKSSRVGRQVMQEPQQPTDDKHITISAPFEIGISMTEEDYDEYGVVLQEIIQRLLGGTEAAEGPPQL
+>Uniprot_GHRL_Hsap
+MPSPGTVCSLLLLGMLWLDLAMAGSSFLSPEHQRVQQRKESKKPPAKLQPRALAGWLRPEDGGQAEGAEDELEVRFNAPFDVGIKLSGVQYQQHSQALGKFLQDILWEEAKEAPADK
+>gi_21238918_Ajap
+MRQMKRTAYIILLVCVLALWMDSVQAGSSFLSPSQRPQGKDKKPPRVGRRDSDGILDLFMRPPLQDEDIRHITFNTPFEIGITMTEELFQQYGEVMQKIMQD
+>gi_68445391_Ipun
+MLGHGRVGHMMLLLCAFSLWAETVMCGSSFLSPTQKPQNRGDRKPPRVGRRTAAELEAPLPSEEKIMVSAPFQLAVSLSDAEYEDYGPVLQRMLLDVLGDPPTLDGAN
+>Uniprot_MOTI_Hsap
+MVSRKAVAALLVVHVAAMLASQTEAFVPIFTYGELQRMQEKERNKGQKKSLSVWQRSGEEGPVDPAEPIREEENEMIKLTAPLEIGMRMNSRQLEKYPATLEGLLSEMLPQHAAK
+>Ensembl_ENSOANT00000031359_Oana
+MVSRKAVAFLLVVSVAAMMAEGFIPIFTHSDVQRMQERERNKGQKKSLTVQQRSEQGGLRTLAEPNGEEEGEIIQLAAPVEIGLRMNSRQLAKYRGILEELIMEALLSTQNGESNPDRGRGRCS
+>Ensembl_ENSOCUT00000005807_Ocun
+MVSRKAVAALLLVHATAMLASQTEAFVPIFTYSELQRMQERERNRGHKKSLSVQQRSEPAAAPPAEPTLEEENGRTQLTAPVEIGMRMNSRQLEKYRAALEAERAVHPDAPSRPCPAGGESGWSGEPSPT

data/data/raw_data/vertebrate_ghrelin_motilin.fa.aligned.fa

@@ -0,0 +1,32 @@
+>Ensembl_ENSDART00000076405_Drer
+MPLRCRASSMFLLLCVSLSLCLESVSGGTSFLSPT-QKPQG-------RRPPRVGRREAA
+------DPEIPVIKEDDR-FMMSAPFELSMSLSEAEYEKYGPVLQNLLENLLRD-SSFEF
+------------------------
+>gi_55792853_Asch
+MFLKRNTYLLVFLFC-SLTLWCKSTSAGSSFLSPS-QKPQNR------GKSSRVGRQ---
+------VMQEPQQPTDDKHITISAPFEIGISMTEEDYDEYGVVLQEIIQRLLGGTEAAEG
+PPQL--------------------
+>Uniprot_GHRL_Hsap
+MPSPGTVCSLLLLGM----LWLDLAMAGSSFLSPEHQRVQQRK--ESKKPPAKLQPRALA
+GWLRPEDGGQAEGAEDELEVRFNAPFDVGIKLSGVQYQQHSQALGKFLQDILWE-EAKEA
+PADK--------------------
+>gi_21238918_Ajap
+MRQMKRTAYIILLVC-VLALWMDSVQAGSSFLSPS-QRPQGK-----DKKPPRVGRRDSD
+GIL-DLFMRPPLQDEDIRHITFNTPFEIGITMTEELFQQYGEVMQKIMQD----------
+------------------------
+>gi_68445391_Ipun
+MLGHGRVGHMMLLLC-AFSLWAETVMCGSSFLSPT-QKPQNR----GDRKPPRVGRRTAA
+------ELEAPLPSEEK--IMVSAPFQLAVSLSDAEYEDYGPVLQRMLLDVLGD------
+PPTLDGA----------------N
+>Uniprot_MOTI_Hsap
+MVSRKAVAALLVVHV-AAMLASQTEAFVPIFTYGELQRMQEKERNKGQKKSLSVWQRSGE
+EGPV--DPAEPIREEENEMIKLTAPLEIGMRMNSRQLEKYPATLEGLLSEML--------
+-PQHAAK-----------------
+>Ensembl_ENSOANT00000031359_Oana
+MVSRKAVAFLLVVSV-AAMMA---EGFIPIFTHSDVQRMQERERNKGQKKSLTVQQRSEQ
+GGLR--TLAEPNGEEEGEIIQLAAPVEIGLRMNSRQLAKYRGILEELIMEALLSTQNGES
+NPDRGRGRC--------------S
+>Ensembl_ENSOCUT00000005807_Ocun
+MVSRKAVAALLLVHA-TAMLASQTEAFVPIFTYSELQRMQERERNRGHKKSLSVQQRSEP
+AAA---PPAEPTLEEENGRTQLTAPVEIGMRMNSRQLEKYRAALE----------AERAV
+HPDAPSRPCPAGGESGWSGEPSPT

data/lib/neurohmmer.rb

@@ -0,0 +1,89 @@
+require 'bio'
+require 'fileutils'
+
+require 'neurohmmer/arg_validators'
+require 'neurohmmer/hmmer'
+require 'neurohmmer/output'
+
+# Top level module / namespace.
+module Neurohmmer
+  class <<self
+    attr_accessor :opt
+    attr_accessor :conf
+
+    def init(opt)
+      @opt = ArgumentsValidators.run(opt)
+      @conf = {
+        hmm_dir: File.expand_path('../../data/hmm', __FILE__),
+        raw_data: File.expand_path('../../data/raw_data', __FILE__),
+        raw_alignments: File.expand_path('../../data/raw_data/alignments',
+                                         __FILE__),
+        hmm_output: File.join(@opt[:temp_dir], 'input.hmm_search.out'),
+        html_output: "#{@opt[:input_file]}.neurohmmer.html"
+      }
+      init_input
+    end
+
+    def run
+      Hmmer.search
+      hmm_analysis = Hmmer.analyse_output
+      Output.to_html(hmm_analysis)
+      remove_temp_dir
+    end
+
+    def extract_sequence(id)
+      id = id.gsub(/\s+/, '')
+      idx = @input_index[id]
+      seq = IO.binread(@opt[:input_file], idx[1] - idx[0], idx[0])
+      seq.scan(/>([^\n]*)\n([A-Za-z\n\*]*)/)[0]
+    end
+
+    private
+
+    def init_input
+      FileUtils.mkdir_p(@opt[:temp_dir])
+      @opt[:input_file] = translate_input if @opt[:type] == :genetic
+      @input_index = index_input_file
+    end
+
+    # Translates the input data in all 6 frames
+    def translate_input(input = @opt[:input_file])
+      translated_file = File.join(@opt[:temp_dir], 'input.translated.fa')
+      File.open(translated_file, 'w') do |file|
+        Bio::FlatFile.open(Bio::FastaFormat, input).each_entry do |entry|
+          (1..6).each do |f|
+            file.puts ">#{entry.definition}-frame:#{f}"
+            file.puts entry.naseq.translate(f)
+          end
+        end
+      end
+      translated_file
+    end
+
+    # Indexes the input file - returns a hash in the following format:
+    # {seq id: [start byte in file, end byte in file] }
+    def index_input_file
+      c = IO.binread(@opt[:input_file])
+      keys   = c.scan(/>(.*)\n/).flatten
+      values = c.enum_for(:scan, /(>[^>]+)/).map { Regexp.last_match.begin(0) }
+      index(c, keys, values)
+    end
+
+    # A method run from index_input_file that creates a simple hash with the
+    # {seq id: [start byte in file, end byte in file] }
+    def index(content, keys, values)
+      fasta_index = {}
+      keys.each_with_index do |k, i|
+        id = k[0..115].gsub(/\s+/, '')
+        endf  = (i == values.length - 1) ? content.length - 1 : values[i + 1]
+        fasta_index[id] = [values[i], endf]
+      end
+      fasta_index
+    end
+
+    def remove_temp_dir
+      return unless File.directory?(@opt[:temp_dir])
+      FileUtils.rm_rf(@opt[:temp_dir])
+    end
+  end
+end

data/lib/neurohmmer/arg_validators.rb

@@ -0,0 +1,71 @@
+require 'bio'
+
+# Top level module / namespace.
+module Neurohmmer
+  # A class that validates the command line opts
+  class ArgumentsValidators
+    class << self
+      def run(opt)
+        assert_file_present('input fasta file', opt[:input_file])
+        assert_input_file_not_empty(opt[:input_file])
+        assert_input_file_probably_fasta(opt[:input_file])
+        opt[:type] = assert_input_sequence(opt[:input_file])
+        # TODO: Assert hmm & mafft binaries
+        opt
+      end
+
+      private
+
+      def assert_file_present(desc, file, exit_code = 1)
+        return if file && File.exist?(File.expand_path(file))
+        $stderr.puts "*** Error: Couldn't find the #{desc}: #{file}."
+        exit exit_code
+      end
+
+      def assert_input_file_not_empty(file)
+        return unless File.zero?(File.expand_path(file))
+        $stderr.puts "*** Error: The input_file (#{file})" \
+                     ' seems to be empty.'
+        exit 1
+      end
+
+      def assert_input_file_probably_fasta(file)
+        File.open(file, 'r') do |f|
+          fasta = (f.readline[0] == '>') ? true : false
+          return fasta if fasta
+        end
+        $stderr.puts "*** Error: The input_file (#{file})" \
+                     ' does not seems to be a fasta file.'
+        exit 1
+      end
+
+      def assert_input_sequence(file)
+        type = type_of_sequences(file)
+        return type unless type.nil?
+        $stderr.puts '*** Error: The input files seems to contain a mixture of'
+        $stderr.puts '    both protein and nucleotide data.'
+        $stderr.puts '    Please correct this and try again.'
+        exit 1
+      end
+
+      def type_of_sequences(file)
+        fasta_content = IO.binread(file)
+        # the first sequence does not need to have a fasta definition line
+        sequences = fasta_content.split(/^>.*$/).delete_if(&:empty?)
+        # get all sequence types
+        sequence_types = sequences.collect { |seq| guess_sequence_type(seq) }
+                         .uniq.compact
+        return nil if sequence_types.empty?
+        sequence_types.first if sequence_types.length == 1
+      end
+
+      def guess_sequence_type(seq)
+        # removing non-letter and ambiguous characters
+        cleaned_sequence = seq.gsub(/[^A-Z]|[NX]/i, '')
+        return nil if cleaned_sequence.length < 10 # conservative
+        type = Bio::Sequence.new(cleaned_sequence).guess(0.9)
+        (type == Bio::Sequence::NA) ? :genetic : :protein
+      end
+    end
+  end
+end

data/lib/neurohmmer/hmmer.rb

@@ -0,0 +1,71 @@
+require 'bio-hmmer3_report'
+require 'forwardable'
+
+# Top level module / namespace.
+module Neurohmmer
+  # A class that holds methods related to Hmmer
+  class Hmmer
+    class <<self
+      extend Forwardable
+      def_delegators Neurohmmer, :opt, :conf
+
+      def search
+        Dir.foreach(conf[:hmm_dir]) do |h|
+          hmm_file = File.join(conf[:hmm_dir], h)
+          next if hmm_file !~ /hmm$/
+          hmm_search(opt[:input_file], hmm_file, conf[:hmm_output])
+        end
+      end
+
+      def analyse_output
+        hmm_results       = {}
+        hmm_search_output = File.open(conf[:hmm_output])
+        hmm_reports       = Bio::HMMER::HMMER3.reports(hmm_search_output)
+        hmm_reports.each_with_index do |report, idx|
+          next if idx + 1 == hmm_reports.length
+          hmm_results[report.query] = analyse_hmm_search_report(report)
+        end
+        hmm_results
+      end
+
+      def generate_hmm_models
+        Dir.foreach(conf[:raw_data]) do |file|
+          next if file !~ /fa(sta)?$/
+          np_fasta_file  = File.join(conf[:raw_data], file)
+          aligned_file   = File.join(conf[:raw_alignments],
+                                     "#{file.gsub(/fa(sta)?$/, '')}.aligned")
+          hmm_model_file = File.join(conf[:hmm_dir],
+                                     "#{file.gsub(/fa(sta)?$/, '')}.hmm")
+          mafft(np_fasta_file, aligned_file, conf[:num_threads])
+          hmm_build(aligned_file, hmm_model_file)
+        end
+      end
+
+      private
+
+      def analyse_hmm_search_report(report)
+        report_result = []
+        report.hits.each do |hit|
+          seq  = Neurohmmer.extract_sequence(hit.sequence_name.strip)
+          hsps = []
+          hit.hsps.each { |hsp|  hsps << hsp.flatseq }
+          report_result << { id: seq[0], seq: seq[1], flatseq: hsps }
+        end
+        report_result
+      end
+
+      def mafft(input, aligned_file, num_threads)
+        `mafft --maxiterate 1000 --thread #{num_threads} '#{input}' > \
+         '#{aligned_file}'`
+      end
+
+      def hmm_build(aligned_file, hmm_model_file)
+        `hmmbuild '#{hmm_model_file}' '#{aligned_file}'`
+      end
+
+      def hmm_search(input_file, hmm_file, hmm_output)
+        `hmmsearch '#{hmm_file}' '#{input_file}' >> '#{hmm_output}'`
+      end
+    end
+  end
+end

data/lib/neurohmmer/output.rb

@@ -0,0 +1,50 @@
+require 'slim'
+require 'forwardable'
+
+# Top level module / namespace.
+module Neurohmmer
+  # A class that holds methods related to Output
+  class Output
+    class <<self
+      extend Forwardable
+      def_delegators Neurohmmer, :conf
+
+      def to_html(hmm_results)
+        @html_results = format_seqs_for_html(hmm_results)
+        template_path = File.expand_path(File.join(__FILE__, '../../../',
+                                                   'template/contents.slim'))
+        contents_temp = File.read(template_path)
+        html_content = Slim::Template.new { contents_temp }.render(self)
+        File.open(conf[:html_output], 'w') { |f| f.puts html_content }
+      end
+
+      def format_seqs_for_html(hmm_results)
+        results = {}
+        hmm_results.each do |query, hits|
+          next if hits.length == 0
+          results[query] = []
+          hits.each do |hit|
+            html_seq = format_html_seq(hit[:seq], hit[:flatseq])
+            results[query] << { id: hit[:id], seq: html_seq }
+          end
+        end
+        results
+      end
+
+      private
+
+      def format_html_seq(seq, flatseq)
+        seq.gsub!("\n", '')
+        flatseq.each do |hsp|
+          seq.gsub!(/#{hsp.gsub('-', '')}/i, '<span class=hsp>\0</span>')
+        end
+        seq.gsub(/KR|KK|RR/i, '<span class=clv>\0</span>')
+          .gsub(/(K|R)<span class=hsp>(K|R)/i, '<span class=clv>\1</span>' \
+                '<span class=clv_i>\2</span><span class=hsp>')
+          .gsub('<span class=clv>R</span><span class=clv_i>K</span><span' \
+                ' class=hsp>', 'R<span class=hsp>K')
+          .gsub(/G<span class=clv>/, '<span class=gly>G</span><span class=clv>')
+      end
+    end
+  end
+end

data/lib/neurohmmer/version.rb

@@ -0,0 +1,4 @@
+# Top level module / namespace.
+module Neurohmmer
+  VERSION = '0.1.0'
+end

data/neurohmmer.gemspec

@@ -0,0 +1,34 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'neurohmmer/version'
+
+Gem::Specification.new do |s|
+  s.name          = 'neurohmmer'
+  s.version       = Neurohmmer::VERSION
+  s.authors       = ['Ismail Moghul', 'Maurice Elphick', 'Yannick Wurm']
+  s.email         = ['ismail.moghul@gmail.com']
+
+  s.summary       = 'Identify Neuropeptides using powerful Hidden Markov' \
+                    " Models.\n\n For further information please refer to:" \
+                    ' https://github.com/wurmlab/neurohmmer'
+  s.description   = 'Identify Neuropeptides using powerful Hidden Markov' \
+                    " Models.\n\n For further information please refer to:" \
+                    ' https://github.com/wurmlab/neurohmmer'
+  s.homepage      = 'https://github.com/wurmlab/neurohmmer'
+  s.license       = 'AGPL'
+
+  s.files         = `git ls-files -z`.split("\x0")
+  s.executables   = s.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  s.test_files    = s.files.grep(%r{^(test|spec|features)/})
+  s.require_paths = ['lib']
+
+  s.required_ruby_version = '>= 2.0.0'
+  s.add_development_dependency 'bundler', '~> 1.10'
+  s.add_development_dependency 'rake', '~> 10.0'
+  s.add_development_dependency 'rspec', '~> 3.4'
+
+  s.add_dependency 'bio', '~> 1.5'
+  s.add_dependency 'bio-hmmer3_report', '~> 0.1'
+  s.add_dependency 'slim', '~> 3.0'
+end

data/spec/neurohmmer_spec.rb

@@ -0,0 +1,11 @@
+require 'spec_helper'
+
+describe Neurohmmer do
+  it 'has a version number' do
+    expect(Neurohmmer::VERSION).not_to be nil
+  end
+
+  it 'does something useful' do
+    expect(false).to eq(true)
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,2 @@
+$LOAD_PATH.unshift File.expand_path('../../lib', __FILE__)
+require 'neurohmmer'

data/template/contents.slim

@@ -0,0 +1,55 @@
+doctype html
+html lang="en"
+  head
+    meta charset="utf-8"
+    meta content="IE=edge" http-equiv="X-UA-Compatible"
+    meta content="width=device-width, initial-scale=1" name="viewport"
+    meta content="NeuroHMMer | Identify Neuropeptide Homologs" name="description"
+    meta content="Wurmlab" name="author"
+    title NeuroHMMer | Identify Neuropeptide Homologs
+    css:
+      html { position: relative; min-height: 100%; }
+      body {margin:0 0 100px; background-color:#F5F5F5; font-family:"Helvetica Neue", Helvetica, Arial, sans-serif; font-size:15px; line-height:1.42857143; color:#2c3e50;}
+      .container {margin-right:auto; margin-left:auto; padding-left:15px; padding-right:15px; max-width:1170px; width:95%;}
+      footer {bottom:0; width:100%; margin:0 auto; position:absolute; height:100px; overflow:hidden; border-top:2px solid #DBDBDB; color:#b4bcc2;text-align:center;}
+      a{color:#18bc9c; text-decoration:none}a:hover,a:focus{color:#18bc9c; text-decoration:underline}
+      p {margin:0 0 10.5px}
+      .id {font-weight:bold;}
+      .sequence {word-break:break-all; font-family:Courier New, Courier, Mono;}
+      .clv {color:#00B050; font-weight:bold;}
+      .clv_i {color:#00B050; font-weight: bold; background-color:#FFE4B5;}
+      .gly {color:#FF0000; font-weight:bold;}
+      .hsp {background-color:#FFE4B5;}
+  body
+    .container
+      h1 NeuroHMMer Results
+      .results
+      - @html_results.each do |query, hits|
+        section
+          h2 #{query.gsub(/.fa(sta)?.aligned/, '')}
+          - hits.each do |hit|
+            p.sequence
+              span.id
+                = hit[:id]
+              br
+              span.seq
+                == hit[:seq]
+          hr
+      br
+      br
+      br
+    footer
+      p
+        | Please cite "Moghul I, Elphick M & Wurm Y 
+        em
+          | (in prep) 
+        | NeuroHMMer: A tool to identify neuropeptide homologs"
+        br
+        |  Developed at 
+        a href="https://wurmlab.github.io" target="_blank" Wurm Lab
+        | , 
+        a href="http://www.sbcs.qmul.ac.uk" target="_blank" QMUL
+        br
+        | This page was created by 
+        a href="https://github.com/wurmlab/NpSearch" target="_blank"  NpSearch 
+        | v#{Neurohmmer::VERSION}