explore SRA

nakanidaremoimasenyo.rb

# -*- coding: utf-8 -*-

require "nokogiri"

hako = "/path/to/directory_contains_pubmed_html"

title_list = []
Dir::entries(hako).each do |item|
	if !item.include?("url_list")
		if !item.include?("¥.")
			kitanaihtml = Nokogiri::HTML(open("#{hako}/#{item}"))
            
            　　　　　　　title = (kitanaihtml/"title").inner_text
            　　　　　　　title_list.push(title)
		end
	end
end

title_list.each do |title|
	puts title.gsub(/( - PubMed result)$/,"")
end

journal_list = []
title_list.each do |title|
	title =‾ %r|¥[(.*)¥]|
	journal = $1
	if journal
    journal_list.push(journal)
	end
end

journal_list.sort.each do |journal|
	puts journal.gsub(/(.....)$/,"")
end

year_list = []
title_list.each do |title|
	title =‾ %r|¥[.*(....)¥]|
	year = $1
	if year
	year_list.push(year)
	end
end

puts year_list.sort                                                                           

toaruNCBInoDaikiboHairetsuKakunouko.rb

# -*- coding: utf-8 -*-

require "open-uri"
require "json"
require "nokogiri"
require "pp"


def sratohananndattanoka(kaneganaitoka,nantokasiroya) # argument: organism, study_type
        
        mishima = "http://trace.ddbj.nig.ac.jp/DRASearch/"
	ground_floor = "#{mishima}query?organism=#{kaneganaitoka}&study_type=#{nantokasiroya}&show=100"
	deeper_underground = "#{mishima}query?organism=#{kaneganaitoka}&study_type=#{nantokasiroya}&show=100&page=2"
    
	list_study = URI(ground_floor).read.scan(/href="(study¥?acc=.*)" target/).flatten
	list_study2 = URI(deeper_underground).read.scan(/href="(study¥?acc=.*)" target/).flatten
	full_list = list_study + list_study2
                        
	# save full sample list as json format file
	sample_id_full_list = full_list.map { |study_url|
	URI("#{mishima}#{study_url}").read.scan(/href="(sample¥?acc=.*)" target/)
	}.flatten
	
	n = kaneganaitoka.gsub("+","_")
	s = nantokasiroya.gsub("+","_")
	open("./#{n}_#{s}_sample_id.json","w") { |f| f.puts JSON.dump(sample_id_full_list) }
	
	return sample_id_full_list
	
end


def mouNCBInantesiruka(majide) # argument: sampleid (e.g. "sample?acc=SRAXXXXXX")
	
        begin
	    mishima = "http://trace.ddbj.nig.ac.jp/DRASearch/"
    	
    	    convert = URI("#{mishima}#{majide}").read.scan(/href="(submission¥?acc=.*)" target/).flatten.join("")
            sub_id = convert.gsub("submission¥?acc=","")
            sub_id_index = sub_id.slice(0,6)
            
            xmldono = File.read("/Users/iNut/togofarm/xmldono/Submissions/#{sub_id_index}/#{sub_id}/#{sub_id}.study.xml")
            nakami = Nokogiri::XML(xmldono)
                            
            entrez_link_db = (nakami/"ENTREZ_LINK"/"DB").inner_text
            entrez_link_id = (nakami/"ENTREZ_LINK"/"ID").inner_text
                                        
	    return entrez_link_db, entrez_link_id
		                                            
	rescue
		"no data"

    end
end                                                                           


def ronbundasumadegasequencedesu(ronbunronbun)

	pmid_list = []
	ronbunronbun.each do |pair|
    	if pair[0].include?("pubmed")
			if pair[0] =‾ %r|^pubmed|
				pmid = pair[1].slice(0,8)
				pmid_list.push(pmid)
			elsif pair[0] =‾ %r|pubmed$|
				pmid = pair[1].scan(/.*(........)$/).join("")
				pmid_list.push(pmid)
			end
		end
	end
	#pp pmid_list.uniq
	
	return pmid_list.uniq.map { |id|
		"http://www.ncbi.nlm.nih.gov/pubmed/" + id
	}
end                                                            



if __FILE__ == $0
	
#	organism = "Homo+sapiens"
	organism = "Mus+musculus"
	study_type = "Transcriptome+Analysis"
	n = organism.gsub("+","_")
	s = study_type.gsub("+","_")
	
	sample_list = sratohananndattanoka(organism,study_type)

	result = []
	sample_list.each do |id|
		result.push(mouNCBInantesiruka(id))
	end
	sorted = result.uniq
	
	open("#{n}_#{s}_db_id.json","w") { |f| JSON.dump(sorted, f) }

	puts ronbundasumadegasequencedesu(sorted)	
	
end