Possibly fixed zipfulldata job, currently testing. Added rake task to remove annotation for vg-SNPs and to remove annotation without SNPs.

app/workers/zipfulldata.rb

@@ -28,6 +28,7 @@ class Zipfulldata
 
   def run(target_address)
     genotypes = Genotype.includes(user: :user_phenotypes).all
+    log "Got #{genotypes.length} genotypes"
 
     # only try to create csv & zip-file if there is data at all.
     if genotypes.empty?
@@ -42,7 +43,9 @@ class Zipfulldata
     end
 
     begin
+      log "Making tmpdir #{tmp_dir}"
       Dir.mkdir(tmp_dir)
+      log "Starting zipfile #{zip_fs_path}"
       Zip::ZipFile.open(zip_fs_path, Zip::ZipFile::CREATE) do |zipfile|
         create_user_csv(genotypes, zipfile)
         create_fitbit_csv(zipfile)

lib/tasks/clean_annotation.rake

@@ -0,0 +1,46 @@
+def check_to_destroy(type)
+  set_to_destroy = Set.new []
+  type.find_each do |m|
+    parental = m.snp
+    if parental == nil
+      set_to_destroy.add(m)
+      next
+    end
+    # check if name of parental snp is a pesky vg-SNP
+    if parental.name.start_with? "vg"
+      set_to_destroy.add(m)
+    end
+  end 
+  set_to_destroy
+end
+
+def destroy(set, type)
+  set.each do |s|
+    if s.parental != nil and s.parental.name.start_with? "vg"
+      # has to be a vg-SNP, we don't want annotation for these right now as they are mostly bogus
+      s.parental.update_attributes(:ranking => 0)
+    end
+    # delete
+    puts "Want to destroy: #{s.inspect}"
+    # type.destroy(s)
+  end
+end
+
+namespace :snps do
+  desc "Iterates over all annotations, deletes a) ones without SNP and b) annotations for SNPs starting in 'vg' (also sets vg-SNPs scores to 0)"
+  task :clean_annotation => :environment do
+    require 'set'
+    # need to store annotations to destroy first, iterating over stuff
+    # while destroying it only makes problems
+    ms_to_destroy = check_to_destroy(MendeleyPaper)
+    sn_to_destroy = check_to_destroy(SnpediaPaper)
+    sp_to_destroy = check_to_destroy(PlosPaper)
+    spg_to_destroy = check_to_destroy(PgpAnnotation)
+    gg_to_destroy = check_to_destroy(GenomeGovPaper)
+    destroy(ms_to_destroy, MendeleyPaper)
+    destroy(sn_to_destroy, SnpediaPaper)
+    destroy(sp_to_destroy, PlosPaper)
+    destroy(spg_to_destroy, PgpAnnotation)
+    destroy(gg_to_destroy, GenomeGovPaper)
+  end
+end

lib/tasks/dump_annotation.rake

@@ -10,6 +10,10 @@ namespace :snps do
     f.write("SNP\tPosition\tChromosome\tYear\tFirst Author\tTitle\tDOI\tOpen Access\tLink\n")
     MendeleyPaper.find_each do |m|
       parental = m.snp
+      if parental == nil
+        puts m.snp_id
+        next
+      end
       position = parental.position
       name = parental.name
       chrom = parental.chromosome
@@ -26,6 +30,10 @@ namespace :snps do
     f.write("SNP\tPosition\tChromosome\tSummary\tLink\n")
     SnpediaPaper.find_each do |sn|
       parental = sn.snp
+      if parental == nil
+        puts m.snp_id
+        next
+      end
       position = parental.position
       name = parental.name
       chrom = parental.chromosome
@@ -38,6 +46,10 @@ namespace :snps do
     f.write("SNP\tPosition\tChromosome\tFirst author\tTitle\tDOI\tYear\n")
     PlosPaper.find_each do |sp|
       parental = sp.snp
+      if parental == nil
+        puts m.snp_id
+        next
+      end
       position = parental.position
       name = parental.name
       chrom = parental.chromosome
@@ -53,6 +65,10 @@ namespace :snps do
     f.write("SNP\tPosition\tChromosome\tGene\tQualified Impact\tInheritance\tSummary\tTrait\n")
     PgpAnnotation.find_each do |spg|
       parental = sp.snp
+      if parental == nil
+        puts m.snp_id
+        next
+      end
       position = parental.position
       name = parental.name
       chrom = parental.chromosome
@@ -68,6 +84,10 @@ namespace :snps do
     f.write("SNP\tPosition\tChromosome\tFirst author\tTitle\tPubmed-link\tYear\tJournal\tTrait\tp-value\tp-value description\tConfidence Interval\n")
     GenomeGovPaper.find_each do |gg|
       parental = gg.snp
+      if parental == nil
+        puts m.snp_id
+        next
+      end
       position = parental.position
       name = parental.name
       chrom = parental.chromosome