diff --git a/dockerfiles/api/Dockerfile b/dockerfiles/api/Dockerfile
index 0426db37..e4542906 100644
--- a/dockerfiles/api/Dockerfile
+++ b/dockerfiles/api/Dockerfile
@@ -1,3 +1,5 @@
+# MG-RAST API
+
 FROM httpd:2.4.12
 
 # MG-RAST dependencies
@@ -6,6 +8,7 @@ RUN apt-get update && apt-get install -y \
   libpq-dev \
   make \
   curl \
+  r-base \
   perl-modules \
   liburi-perl \
   libwww-perl \
@@ -40,10 +43,8 @@ RUN apt-get install -y \
   libtemplate-perl
 
 RUN pip install gspread xlrd openpyxl lepl
-
 RUN cd / && git clone https://github.com/MG-RAST/pipeline.git
 
-
 RUN mkdir -p /sites/1/ && \
   cd /sites/1/ && \
   ln -s /MG-RAST/
@@ -58,4 +59,4 @@ RUN mkdir -p /m5nr && \
   ln -s /api-server-data/20131215 /m5nr/20131215
 
 # Execute:
-# /usr/local/apache2/bin/httpd -DFOREGROUND -f /MG-RAST/conf/httpd.conf
+# /usr/local/apache2/bin/httpd -DFOREGROUND -f /api-server-conf/httpd.conf
diff --git a/dockerfiles/web/Dockerfile b/dockerfiles/web/Dockerfile
index ff850f56..f3fb6519 100644
--- a/dockerfiles/web/Dockerfile
+++ b/dockerfiles/web/Dockerfile
@@ -64,6 +64,7 @@ RUN cd / && \
 # certificates need to be in daemon home directory
 RUN ln -s /config/postgresql/ /usr/sbin/.postgresql
 
+# docker run -t -i --name web -v /home/core/mgrast-config/services/web_v3/:/config -p 80:80 httpd:2.4.12 bash
 # Start
-#bash -c "/bin/chmod 600 /config/postgresql/* ; /bin/chown daemon:daemon /config/postgresql/* ; /usr/local/apache2/bin/apachectl -f /config/httpd.conf -D FOREGROUND"
+# bash -c "/bin/chmod 600 /config/postgresql/*; /bin/chown daemon:daemon /config/postgresql/*; /usr/local/apache2/bin/apachectl -f /config/httpd.conf -D FOREGROUND"
 
diff --git a/src/MGRAST/Schema/m5nr_copy.cql.tt b/src/MGRAST/Schema/m5nr_copy.cql.tt
new file mode 100644
index 00000000..04aa3ed9
--- /dev/null
+++ b/src/MGRAST/Schema/m5nr_copy.cql.tt
@@ -0,0 +1,17 @@
+
+use m5nr_v[% version %];
+
+COPY ontologies (source, name, level1, level2, level3, level4) FROM '[% data_dir %]/m5nr_v[% version %].ontology.all';
+COPY ont_level1 (source, level1, name) FROM '[% data_dir %]/m5nr_v[% version %].ontology.level1';
+COPY ont_level2 (source, level2, name) FROM '[% data_dir %]/m5nr_v[% version %].ontology.level2';
+COPY ont_level3 (source, level3, name) FROM '[% data_dir %]/m5nr_v[% version %].ontology.level3';
+COPY ont_level4 (source, level4, name) FROM '[% data_dir %]/m5nr_v[% version %].ontology.level4';
+
+COPY organisms_ncbi (name, tax_domain, tax_phylum, tax_class, tax_order, tax_family, tax_genus, tax_species, ncbi_tax_id) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.all';
+COPY tax_domain (tax_domain, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.domain';
+COPY tax_phylum (tax_phylum, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.phylum';
+COPY tax_class (tax_class, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.class';
+COPY tax_order (tax_order, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.order';
+COPY tax_family (tax_family, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.family';
+COPY tax_genus (tax_genus, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.genus';
+COPY tax_species (tax_species, name) FROM '[% data_dir %]/m5nr_v[% version %].taxonomy.species';
diff --git a/src/MGRAST/Schema/m5nr_v4.cql b/src/MGRAST/Schema/m5nr_table.cql.tt
similarity index 60%
rename from src/MGRAST/Schema/m5nr_v4.cql
rename to src/MGRAST/Schema/m5nr_table.cql.tt
index 02d87d36..04f94a3c 100644
--- a/src/MGRAST/Schema/m5nr_v4.cql
+++ b/src/MGRAST/Schema/m5nr_table.cql.tt
@@ -1,10 +1,10 @@
 
-CREATE KEYSPACE IF NOT EXISTS m5nr_v1
-    WITH REPLICATION = { 'class' : 'SimpleStrategy', 'replication_factor' : 2 };
+CREATE KEYSPACE IF NOT EXISTS m5nr_v[% version %]
+    WITH REPLICATION = { 'class' : 'SimpleStrategy', 'replication_factor' : [% replication %] };
 
-use m5nr_v1;
+use m5nr_v[% version %];
 
-CREATE TABLE IF NOT EXISTS md5_id_annotation (
+CREATE TABLE IF NOT EXISTS id_annotation (
     id int,
     source text,
     md5 text,
@@ -16,20 +16,18 @@ CREATE TABLE IF NOT EXISTS md5_id_annotation (
     organism list<text>,
     PRIMARY KEY (id, source)
 );
-COPY md5_id_annotation (id, source, md5, is_protein, single, lca, accession, function, organism) FROM '/mnt/src/m5nr_v1.annotation.id';
 
 CREATE TABLE IF NOT EXISTS md5_annotation (
     md5 text,
     source text,
     is_protein boolean,
     single text,
-    lca text,
+    lca list<text>,
     accession list<text>,
     function list<text>,
     organism list<text>,
     PRIMARY KEY (md5, source)
 );
-COPY md5_annotation (md5, source, is_protein, single, lca, accession, function, organism) FROM '/mnt/src/m5nr_v1.annotation.md5';
 
 CREATE TABLE IF NOT EXISTS ontologies (
     source text,
@@ -40,7 +38,6 @@ CREATE TABLE IF NOT EXISTS ontologies (
     level4 text,
     PRIMARY KEY (source, name)
 );
-COPY ontologies (source, name, level1, level2, level3, level4) FROM '/mnt/src/m5nr_v1.ontology.all';
 
 CREATE TABLE IF NOT EXISTS ont_level1 (
     source text,
@@ -48,7 +45,6 @@ CREATE TABLE IF NOT EXISTS ont_level1 (
     name text,
     PRIMARY KEY (source, level1, name)
 );
-COPY ont_level1 (source, level1, name) FROM '/mnt/src/m5nr_v1.ontology.level1';
 
 CREATE TABLE IF NOT EXISTS ont_level2 (
     source text,
@@ -56,7 +52,6 @@ CREATE TABLE IF NOT EXISTS ont_level2 (
     name text,
     PRIMARY KEY (source, level2, name)
 );
-COPY ont_level2 (source, level2, name) FROM '/mnt/src/m5nr_v1.ontology.level2';
 
 CREATE TABLE IF NOT EXISTS ont_level3 (
     source text,
@@ -64,7 +59,6 @@ CREATE TABLE IF NOT EXISTS ont_level3 (
     name text,
     PRIMARY KEY (source, level3, name)
 );
-COPY ont_level3 (source, level3, name) FROM '/mnt/src/m5nr_v1.ontology.level3';
 
 CREATE TABLE IF NOT EXISTS ont_level4 (
     source text,
@@ -72,7 +66,6 @@ CREATE TABLE IF NOT EXISTS ont_level4 (
     name text,
     PRIMARY KEY (source, level4, name)
 );
-COPY ont_level4 (source, level4, name) FROM '/mnt/src/m5nr_v1.ontology.level4';
 
 CREATE TABLE IF NOT EXISTS organisms_ncbi (
     name text,
@@ -86,53 +79,45 @@ CREATE TABLE IF NOT EXISTS organisms_ncbi (
     ncbi_tax_id int,
     PRIMARY KEY (name)
 );
-COPY organisms_ncbi (name, tax_domain, tax_phylum, tax_class, tax_order, tax_family, tax_genus, tax_species, ncbi_tax_id) FROM '/mnt/src/m5nr_v1.taxonomy.all';
 
 CREATE TABLE IF NOT EXISTS tax_domain (
     tax_domain text,
     name text,
     PRIMARY KEY (tax_domain, name)
 );
-COPY tax_domain (tax_domain, name) FROM '/mnt/src/m5nr_v1.taxonomy.domain';
 
 CREATE TABLE IF NOT EXISTS tax_phylum (
     tax_phylum text,
     name text,
     PRIMARY KEY (tax_phylum, name)
 );
-COPY tax_phylum (tax_phylum, name) FROM '/mnt/src/m5nr_v1.taxonomy.phylum';
 
 CREATE TABLE IF NOT EXISTS tax_class (
     tax_class text,
     name text,
     PRIMARY KEY (tax_class, name)
 );
-COPY tax_class (tax_class, name) FROM '/mnt/src/m5nr_v1.taxonomy.class';
 
 CREATE TABLE IF NOT EXISTS tax_order (
     tax_order text,
     name text,
     PRIMARY KEY (tax_order, name)
 );
-COPY tax_order (tax_order, name) FROM '/mnt/src/m5nr_v1.taxonomy.order';
 
 CREATE TABLE IF NOT EXISTS tax_family (
     tax_family text,
     name text,
     PRIMARY KEY (tax_family, name)
 );
-COPY tax_family (tax_family, name) FROM '/mnt/src/m5nr_v1.taxonomy.family';
 
 CREATE TABLE IF NOT EXISTS tax_genus (
     tax_genus text,
     name text,
     PRIMARY KEY (tax_genus, name)
 );
-COPY tax_genus (tax_genus, name) FROM '/mnt/src/m5nr_v1.taxonomy.genus';
 
 CREATE TABLE IF NOT EXISTS tax_species (
     tax_species text,
     name text,
     PRIMARY KEY (tax_species, name)
 );
-COPY tax_species (tax_species, name) FROM '/mnt/src/m5nr_v1.taxonomy.species';
diff --git a/src/MGRAST/Schema/mgrast_analysis_v4.cql b/src/MGRAST/Schema/mgrast_analysis_v4.cql
index 7d4bde93..9376df40 100644
--- a/src/MGRAST/Schema/mgrast_analysis_v4.cql
+++ b/src/MGRAST/Schema/mgrast_analysis_v4.cql
@@ -12,9 +12,15 @@ CREATE TYPE source_info (
 
 CREATE TYPE md5_info (
     abundance int,
+    exp_avg float,
+    ident_avg float,
+    len_avg float,
+    exp_stdv float,
+    ident_stdv float,
+    len_stdv float,
     seek bigint,
     length int,
-    annotation map<text, frozen <source_info>>
+    is_protein boolean
 );
 
 CREATE TYPE lca_info (
@@ -42,6 +48,7 @@ INSERT INTO job_test (version, job, exp_avg, md5, data) VALUES (1, 20, 60, 'foo'
 INSERT INTO job_test (version, job, exp_avg, md5, data) VALUES (1, 20, 70, 'foo', 'hello world');
 INSERT INTO job_test (version, job, exp_avg, md5, data) VALUES (1, 20, 70, 'bar', 'hello world');
 
+### version 1 ###
 CREATE TABLE IF NOT EXISTS job_md5s (
     version int,
     job int,
@@ -70,6 +77,7 @@ CREATE TABLE IF NOT EXISTS job_lcas (
 
 # see: http://docs.datastax.com/en/cql/3.1/cql/cql_using/use-slice-partition.html
 
+### version 2 ###
 CREATE TABLE IF NOT EXISTS job_md5s (
     version int,
     job int,
@@ -107,3 +115,17 @@ CREATE TABLE IF NOT EXISTS job_lcas (
     PRIMARY KEY ((version, job), exp_avg, ident_avg, len_avg, lca)
 );
 
+### version 3 ###
+CREATE TABLE IF NOT EXISTS job_md5s (
+    version int,
+    job int,
+    source text,
+    md5 text,
+    data frozen<md5_info>,
+    accession list<text>,
+    function list<text>,
+    organism list<text>,
+    PRIMARY KEY ((version, job), source, md5)
+);
+    
+
diff --git a/src/MGRAST/html/MGRAST-frontpage.tmpl b/src/MGRAST/html/MGRAST-frontpage.tmpl
index 2c08eee2..76077232 100644
--- a/src/MGRAST/html/MGRAST-frontpage.tmpl
+++ b/src/MGRAST/html/MGRAST-frontpage.tmpl
@@ -8,7 +8,7 @@
 <div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=MetagenomeSearch"><img src='./Html/lupe.png' style='width: 20px; height: 20px;' title='Search'></a></div>
 	<br>
 	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=DownloadMetagenome"><img src='./Html/mg-download.png' style='width: 20px; height: 20px;' title=Download></a></div>
-    	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=Upload"><img src='./Html/mg-upload.png' style='width: 20px; height: 20px;' title='Upload'></a></div>
+    	<div id="top_nav_links"><a class= "nav_top" href="./Html/mgmain.html?mgpage=upload"><img src='./Html/mg-upload.png' style='width: 20px; height: 20px;' title='Upload'></a></div>
 	<div id="top_nav_links"><a class= "nav_top" href="http://blog.metagenomics.anl.gov/howto/" target="blank"><img src='./Html/mg-help.png' style='width: 20px; height: 20px;' title='Support'></a></div>
 	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=Contact"><img src='./Html/mg-contact.png' style='width: 20px; height: 20px;' title='Contact'></a></div>
       </div>
diff --git a/src/MGRAST/html/MGRAST.tmpl b/src/MGRAST/html/MGRAST.tmpl
index c6914940..d449bc10 100644
--- a/src/MGRAST/html/MGRAST.tmpl
+++ b/src/MGRAST/html/MGRAST.tmpl
@@ -14,7 +14,7 @@
 	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=MetagenomeSearch"><img src='./Html/lupe.png' style='width: 20px; height: 20px;' title='Search'></a></div>
 	<br>
 	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=DownloadMetagenome"><img src='./Html/mg-download.png' style='width: 20px; height: 20px;' title=Download></a></div>
-    	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=Upload"><img src='./Html/mg-upload.png' style='width: 20px; height: 20px;' title='Upload'></a></div>
+    	<div id="top_nav_links"><a class= "nav_top" href="./Html/mgmainv3.html?mgpage=upload"><img src='./Html/mg-upload.png' style='width: 20px; height: 20px;' title='Upload'></a></div>
 	<div id="top_nav_links"><a class= "nav_top" href="http://blog.metagenomics.anl.gov/howto/" target=_blank><img src='./Html/mg-help.png' style='width: 20px; height: 20px;' title='Support'></a></div>
 	<div id="top_nav_links"><a class= "nav_top" href="metagenomics.cgi?page=Contact"><img src='./Html/mg-contact.png' style='width: 20px; height: 20px;' title='Contact'></a></div>
       </div>
diff --git a/src/MGRAST/html/js/MetagenomeSearch.js b/src/MGRAST/html/js/MetagenomeSearch.js
index 77e3d759..747ec77c 100644
--- a/src/MGRAST/html/js/MetagenomeSearch.js
+++ b/src/MGRAST/html/js/MetagenomeSearch.js
@@ -1,4 +1,4 @@
-var api_url = 'http://api.metagenomics.anl.gov/1/metagenome?verbosity=mixs&';
+var api_url = RetinaConfig ? RetinaConfig.mgrast_api + "/metagenome?verbosity=mixs&" : 'http://api.metagenomics.anl.gov/1/metagenome?verbosity=mixs&';
 var datastore = {};
 var result = 'result'; // div where results are to be displayed
 var saved_params = {};
diff --git a/src/MGRAST/lib/JobDB/Metadata.pm b/src/MGRAST/lib/JobDB/Metadata.pm
index dd6b2bb3..01741ca3 100644
--- a/src/MGRAST/lib/JobDB/Metadata.pm
+++ b/src/MGRAST/lib/JobDB/Metadata.pm
@@ -2,6 +2,7 @@ package JobDB::Metadata;
 
 use strict;
 use warnings;
+use Conf;
 
 1;
 
@@ -130,7 +131,7 @@ sub get_data {
     $obj->{id} = "mgm".$job->{metagenome_id};
     $obj->{about} = "mobedac metagenome";
     $obj->{name} = $job->{name};
-    $obj->{url} = 'http://metagenomics.anl.gov/linkin.cgi?id='.$obj->{id};
+    $obj->{url} = $Conf::cgi_url.'linkin.cgi?id='.$obj->{id};
     $obj->{version} = $job->{server_version};
     $obj->{creation} = $job->{created_on};
 #    $obj->{sample} = $job->sample->{ID};
diff --git a/src/MGRAST/lib/WebPage/Analysis.pm b/src/MGRAST/lib/WebPage/Analysis.pm
index 6e57ac68..1a558cae 100644
--- a/src/MGRAST/lib/WebPage/Analysis.pm
+++ b/src/MGRAST/lib/WebPage/Analysis.pm
@@ -6603,8 +6603,8 @@ sub selectable_metagenomes {
       push @{$collections->{$name}}, [ $pj->{metagenome_id}, $pj->{name} ];
     }
     foreach my $coll ( sort keys %$collections ) {
-      if ( @{$collections->{$coll}} == 0 ) { next; }
-      push(@$colls, { label => $coll." [".scalar(@{$collections->{$coll}})."]", value => join('||', map { $_->[0]."##".$_->[1] } @{$collections->{$coll}}) });
+      if ( (! $coll) || (! $collections->{$coll}) || (@{$collections->{$coll}} == 0) ) { next; }
+      push(@$colls, { label => $coll." [".scalar(@{$collections->{$coll}})."]", value => join('||', map { ($_->[0] || "")."##".($_->[1] || "") } @{$collections->{$coll}}) });
     }
   }    
   
diff --git a/src/MGRAST/lib/WebPage/DownloadMetagenome.pm b/src/MGRAST/lib/WebPage/DownloadMetagenome.pm
index da4eb29f..698277bb 100644
--- a/src/MGRAST/lib/WebPage/DownloadMetagenome.pm
+++ b/src/MGRAST/lib/WebPage/DownloadMetagenome.pm
@@ -73,7 +73,7 @@ sub init {
   $self->data('default', $default);
 
   # api info for download
-  $self->data('api', "http://api.metagenomics.anl.gov");
+  $self->data('api', $Conf::api_url || "http://api.metagenomics.anl.gov");
 
   # get to metagenome using the metagenome ID
   if ( $cgi->param('metagenome') ) {
diff --git a/src/MGRAST/lib/WebPage/Home.pm b/src/MGRAST/lib/WebPage/Home.pm
index c133c4af..c6610911 100644
--- a/src/MGRAST/lib/WebPage/Home.pm
+++ b/src/MGRAST/lib/WebPage/Home.pm
@@ -58,7 +58,7 @@ sub output {
   if ($is_old) {
     $content .= qq~<script>
  function bookmark(url, title) {
-   if (confirm('The URL you are using to reach this page is deprecated.\\nThe new URL is:\\n\\n\\thttp://metagenomics.anl.gov/\\n\\nWould you like to bookmark the new location?')) {
+   if (confirm('The URL you are using to reach this page is deprecated.\\nThe new URL is:\\n\\n\\t~.$Conf::cgi_url.qq~\\n\\nWould you like to bookmark the new location?')) {
      var url = url || location.protocol + '//' + location.host;
      var title = title || location.host;
      if(document.all) {
@@ -69,7 +69,7 @@ sub output {
    }
    return false;
  }
- </script><img src='./Html/clear.gif' onload='bookmark("http://metagenomics.anl.gov/", "MG-RAST");'>~;
+ </script><img src='./Html/clear.gif' onload='bookmark("~.$Conf::cgi_url.qq~", "MG-RAST");'>~;
   }
 
   $content .= "<div class='clear' style='height:30px;'></div>";
@@ -117,7 +117,7 @@ function forward_to_search (e) {
   $content .= "<a href='?page=Register' title='Register a new account'><div style='float:left; cursor: pointer;padding-top:5px;'>".$register."</div></a>";  
   $content .= "<a href='?page=Contact' title='Click here for contact information'><div style='float:left; cursor: pointer;padding-top:5px;'>".$contact."</div></a>";
   $content .= "<a href='http://blog.metagenomics.anl.gov' title='Click here for support and FAQs'><div style='float:left; cursor: pointer;padding-top:5px;'>".$help."</div></a>";
-  $content .= "<a href='?page=Upload' title='Upload a new metagenome'><div style='float:left; cursor: pointer;padding-top:5px; margin-right:20px;'>".$upload."</div></a>";
+  $content .= "<a href='./Html/mgmainv3.html?mgpage=upload' title='Upload a new metagenome'><div style='float:left; cursor: pointer;padding-top:5px; margin-right:20px;'>".$upload."</div></a>";
   $content .= "<a href='http://blog.metagenomics.anl.gov/' target=_blank><div style='float:left; cursor: pointer;padding-top:5px;'>".$news."</div></a>";
   $content .= "</div>";
   $content .= "<div class='clear'></div>";
@@ -176,13 +176,7 @@ END
 
   $content .= "<a href=\"http://www.biomedcentral.com/1471-2105/9/386\" target=\"_blank\"><p style='border-radius: 5px; -webkit-border-radius: 5px; -moz-border-radius: 5px; color: white; padding: 5px 0px 3px 10px; color: white; font-size:14; background-color: #5281B0; width: 100px; float: left; margin-right: 10px; top: -12px; position: relative;'>cite MG-RAST</p></a><a href=\"http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004008\" target=\"_blank\" style=\"float: right;\"><p style='border-radius: 5px; -webkit-border-radius: 5px; -moz-border-radius: 5px; color: white; padding: 5px 0px 3px 10px; color: white; font-size:14; background-color: #5281B0; width: 120px; float: left; margin-right: 10px; top: -12px; position: relative;'>cite MG-RAST API</p></a>";
 
-  #$content .= "<p style='color:#8FBC3F;text-align:left;font-size:9px;margin-top:3px;'><em>The Metagenomics RAST server - A public resource for the automatic phylogenetic and functional analysis of metagenomes</em> F. Meyer, D. Paarmann, M. D'Souza, R. Olson , E. M. Glass, M. Kubal, T. Paczian , A. Rodriguez , R. Stevens, A. Wilke, J. Wilkening, R. A. Edwards<br/><em>BMC Bioinformatics 2008, 9:386
-#, [<a href=\"http://www.biomedcentral.com/1471-2105/9/386\" target=\"_blank\">article</a>]</em></p>";
-
-  #$content .= "<p style='text-align: center; float: right;'><span style='font-size: 11px; color: white;'>powered by</span><br><img style='width: 50px;' src='./Html/kbase_logo.png'></p>";
-
   my $logos = "<img src='./Html/argonne_header_logo.jpg' style='padding-top: 15px; width: 150px;'>";
-  # news - logos
 
   return $content;
 }
@@ -198,36 +192,6 @@ sub speedometer {
     my ($speed, $mileage, $trip, $togo) = split(/\t/, $line);
 
     $content .= "<table style='background-color: #2f2f2f; position: absolute; right: 22px; top: 78px; color: rgb(143, 188, 63); font-family: Verdana,Arial,sans-serif;'><tr><td colspan=3 align=center style='font-size: 9pt;font-weight: bold;'>pipeline status</td></tr><tr title='analysis speed in basepairs/second'><td align=right style='font-size: 8pt;'>$speed</td><td align=left style='font-size: 8pt;'>bp/s</td><td align=left style='font-size: 8pt;'>speed</td></tr><tr title='data analyzed in the last 30 days'><td align=right style='font-size: 8pt;'>$trip</td><td align=left style='font-size: 8pt;'>Mbp</td><td align=left style='font-size: 8pt;'>last 30 days</td></tr><tr title='data left to be processed'><td align=right style='font-size: 8pt;'>$togo</td><td align=left style='font-size: 8pt;'>Mbp</td><td align=left style='font-size: 8pt;'>in queue</td></tr><tr title='total amount of data analyzed'><td align=right style='font-size: 8pt;'>$mileage</td><td align=left style='font-size: 8pt;'>Mbp</td><td align=left style='font-size: 8pt;'>total</td></tr></table>";
-
-    # my @trip1 = split(//, $trip);
-#     while (scalar(@trip1) < 7) {
-#       unshift(@trip1, "0");
-#     }
-#     $trip = '<table><tr>';# style="margin-left: 36px;"
-#     foreach my $t (@trip1) {
-#       $trip .= "<td style='font-size: 10px; border: 1px solid black; background-color: white; font-weight: bold; width: 12px; text-align: center;'>$t</td>";
-#     }
-#     $trip .= '</tr></table>';  
-#     my @togo1 = split(//, $togo);
-#     while (scalar(@togo1) < 7) {
-#       unshift(@togo1, "0");
-#     }
-#     $togo = '<table><tr>';# style="margin-left: 36px;"
-#     foreach my $t (@togo1) {
-#       $togo .= "<td style='font-size: 10px; border: 1px solid black; background-color: white; font-weight: bold; width: 12px; text-align: center;'>$t</td>";
-#     }
-#     $togo .= '</tr></table>';
-#     my @mileage1 = split(//, $mileage);
-#     while (scalar(@mileage1) < 7) {
-#       unshift(@mileage1, "0");
-#     }
-#     $mileage = '<table><tr>';
-#     foreach my $t (@mileage1) {
-#       $mileage .= "<td style='font-size: 10px; border: 1px solid black; background-color: white; font-weight: bold; width: 12px; text-align: center;'>$t</td>";
-#     }
-#     $mileage .= '</tr></table>';
-    
-#     $content .= "<table style='background-color: #2f2f2f; position: absolute; right: 22px; top: 79px; color: rgb(143, 188, 63); font-weight: bold; font-family: Verdana,Arial,sans-serif;'><tr title='analysis speed in basepairs/second'><td style='font-size: 10px; text-align: right;'>$speed</td><td style='font-size: 10px; vertical-align: middle;'>bp/s</td><td style='font-size: 10px; vertical-align: middle;'>speed</td></tr><tr title='data analyzed in the last 30 days'><td>$trip</td><td style='font-size: 10px; vertical-align: middle;'>Mbp</td><td style='font-size: 10px; vertical-align: middle;'>last 30 days</td></tr><tr title='data left to be processed'><td>$togo</td><td style='font-size: 10px; vertical-align: middle;'>Mbp</td><td style='font-size: 10px; vertical-align: middle;'>in queue</td></tr><tr title='total amount of data analyzed'><td>$mileage</td><td style='font-size: 10px; vertical-align: middle;'>Mbp</td><td style='font-size: 10px; vertical-align: middle;'>total</td></tr></table>";
   }
 
   return $content;
diff --git a/src/MGRAST/lib/WebPage/MetagenomeOverview.pm b/src/MGRAST/lib/WebPage/MetagenomeOverview.pm
index e2d284ba..c4d40bfd 100644
--- a/src/MGRAST/lib/WebPage/MetagenomeOverview.pm
+++ b/src/MGRAST/lib/WebPage/MetagenomeOverview.pm
@@ -146,7 +146,7 @@ sub output {
   my $mddb = $self->data('mddb');
   my $user = $self->application->session->user;
   my $job_id  = $job->job_id;
-  my $mg_link = "http://metagenomics.anl.gov/linkin.cgi?metagenome=$mgid";
+  my $mg_link = $Conf::cgi_url."linkin.cgi?metagenome=$mgid";
 
   # get project information
   my $project_link   = "";
diff --git a/src/MGRAST/lib/WebPage/MetagenomeProject.pm b/src/MGRAST/lib/WebPage/MetagenomeProject.pm
index fdb88cae..9f4c55b5 100644
--- a/src/MGRAST/lib/WebPage/MetagenomeProject.pm
+++ b/src/MGRAST/lib/WebPage/MetagenomeProject.pm
@@ -129,7 +129,7 @@ sub output {
       $self->{is_editor} = 1;
     }
     
-    my $proj_link = "http://metagenomics.anl.gov/linkin.cgi?project=".$self->{project_id};
+    my $proj_link = $Conf::cgi_url."linkin.cgi?project=".$self->{project_id};
     $html .= "<h1 style='display: inline;'>".$project->name.(($user and $user->has_right(undef, 'edit', 'user', '*')) ? " (ID ".$project->id.")": "")."</h1>";
     $html .= "<p><table>";
     $html .= "<tr><td><b>Visibility</b></td><td style='padding-left:15px;'>".($project->public ? 'Public' : 'Private')."</td></tr>";
diff --git a/src/MGRAST/lib/WebPage/MetagenomeSearch.pm b/src/MGRAST/lib/WebPage/MetagenomeSearch.pm
index 91cff8e3..68da4ee6 100644
--- a/src/MGRAST/lib/WebPage/MetagenomeSearch.pm
+++ b/src/MGRAST/lib/WebPage/MetagenomeSearch.pm
@@ -235,6 +235,7 @@ sub output {
   
   my $scripts  = qq~
 <script type="text/javascript" src="http://maps.google.com/maps/api/js?sensor=false"></script>
+<script type="text/javascript" src="js/config.js"></script>
 <script type="text/javascript">
 \$(document).ready( function() {
   var qList = $qlist;
diff --git a/src/MGRAST/lib/WebPage/PublishGenome.pm b/src/MGRAST/lib/WebPage/PublishGenome.pm
index 170f1a3c..75233fdf 100644
--- a/src/MGRAST/lib/WebPage/PublishGenome.pm
+++ b/src/MGRAST/lib/WebPage/PublishGenome.pm
@@ -10,6 +10,7 @@ use JSON;
 use LWP::UserAgent;
 use HTTP::Request;
 use Data::Dumper;
+use Conf;
 
 use WebConfig;
 use base qw( WebPage );
@@ -62,10 +63,10 @@ sub init {
   }
   
   # api info for making public
-  $self->data('api', "http://api.metagenomics.anl.gov");
+  $self->data('api', $Conf::api_url || "http://api.metagenomics.anl.gov");
   
   $self->data('job', $job);
-  $self->data('linkin', "http://metagenomics.anl.gov/linkin.cgi?metagenome=$id");
+  $self->data('linkin', $Conf::cgi_url."linkin.cgi?metagenome=$id");
 }
 
 
diff --git a/src/MGRAST/lib/resources/profile.pm b/src/MGRAST/lib/resources/profile.pm
index 46afd651..bf4a0f3b 100644
--- a/src/MGRAST/lib/resources/profile.pm
+++ b/src/MGRAST/lib/resources/profile.pm
@@ -285,7 +285,7 @@ sub prepare_data {
     }
     elsif ($params->{type} eq 'feature') {
         $ttype = 'Gene';
-        my $id2ann = {}; # md5_id => [ [accession, function, organism] ]
+        my $id2ann = {}; # md5_id => { accession => [], function => [], organism => [], ontology => [] }
         my $id2md5 = {}; # md5_id => md5
         # mgid, md5_id, abundance, exp_avg, exp_stdv, ident_avg, ident_stdv, len_avg, len_stdv
         my $result = $mgdb->get_md5_data(undef, $params->{evalue}, $params->{identity}, $params->{length}, 1);
@@ -296,27 +296,33 @@ sub prepare_data {
         #    push @{ $id2ann->{$a->[0]} }, [ $a->[1], $a->[3], $a->[4] ];
         #}
         # from m5nr solr / by batch of 1000
-        my $fields = ['md5_id', 'md5', 'organism', 'function', 'accession'];
-        my $srcid  = $mgdb->_src_id->{$params->{source}};
-        my $iter   = natatime 1000, @md5s;
+        my $fields = ['md5_id', 'md5', 'organism', 'function', 'accession', 'source', 'type'];
+        my $squery = "source:".$params->{source};
+        if ($params->{source} eq "SEED") {
+            $squery .= " OR source:Subsystems";
+        } elsif ($params->{source} eq "KEGG") {
+            $squery .= " OR source:KO";
+        }
+        my $iter = natatime 1000, @md5s;
         while (my @curr = $iter->()) {
-            my $query_str = "(source_id:$srcid) AND (md5_id:(".join(" OR ", @curr)."))";
+            my $query_str = "(object:annotation) AND ($squery) AND (md5_id:(".join(" OR ", @curr)."))";
             my ($solr_data, $row_count) = $self->get_solr_query("POST", $Conf::m5nr_solr, $Conf::m5nr_collect.'_'.$mgdb->_version, $query_str, "", 0, 1000000000, $fields);
             foreach my $info (@$solr_data) {
                 $id2md5->{$info->{md5_id}} = $info->{md5};
-                push @{ $id2ann->{$info->{md5_id}} }, [ $info->{accession}, $info->{function}, $info->{organism} ];
+                if ($info->{source} eq $params->{source}) {
+                    push @{ $id2ann->{$info->{md5_id}}{accession} }, $info->{accession};
+                    push @{ $id2ann->{$info->{md5_id}}{function} }, $info->{function};
+                    push @{ $id2ann->{$info->{md5_id}}{organism} }, $info->{organism};
+                } elsif ($info->{type} eq "ontology") {
+                    push @{ $id2ann->{$info->{md5_id}}{ontology} }, $info->{accession};
+                }
             }
         }
         # add metadata to rows
         foreach my $row (@$result) {
             my $mid = $row->[1];
             next unless (exists $id2md5->{$mid});
-            my $metadata = {
-                accession => [ map {$_->[0]} @{$id2ann->{$mid}} ],
-                function  => [ map {$_->[1]} @{$id2ann->{$mid}} ],
-                organism  => [ map {$_->[2]} @{$id2ann->{$mid}} ]
-            };
-            push(@$rows, { "id" => $id2md5->{$mid}, "metadata" => $metadata });
+            push(@$rows, { "id" => $id2md5->{$mid}, "metadata" => $id2ann->{$mid} });
             push(@$values, [ $self->toFloat($row->[2]), $self->toFloat($row->[3]), $self->toFloat($row->[5]), $self->toFloat($row->[7]) ]);
         }
     }
diff --git a/src/MGRAST/lib/resources/validation.pm b/src/MGRAST/lib/resources/validation.pm
index 97468ce5..6652b9a0 100644
--- a/src/MGRAST/lib/resources/validation.pm
+++ b/src/MGRAST/lib/resources/validation.pm
@@ -515,7 +515,7 @@ sub reformat_template {
   use JSON;
   my $ua = LWP::UserAgent->new;
   my $json = new JSON;
-  my $data = $json->decode($ua->get('http://api.metagenomics.anl.gov/1/metadata/template')->content);
+  my $data = $json->decode($ua->get(($Conf::api_url || 'http://api.metagenomics.anl.gov/1/').'metadata/template')->content);
 
   my $template = { "name" => "mgrast",
 		   "label" => "MG-RAST",
diff --git a/src/MGRAST/tools/BulkLoader/BulkLoader.java b/src/MGRAST/tools/BulkLoader/BulkLoader.java
new file mode 100644
index 00000000..5973a7f0
--- /dev/null
+++ b/src/MGRAST/tools/BulkLoader/BulkLoader.java
@@ -0,0 +1,199 @@
+import java.io.*;
+import java.math.BigDecimal;
+import java.net.HttpURLConnection;
+import java.net.URL;
+import java.text.ParseException;
+import java.text.SimpleDateFormat;
+import java.util.List;
+import java.util.Arrays;
+import java.util.ArrayList;
+
+import com.opencsv.CSVReader;
+
+import org.apache.cassandra.config.Config;
+import org.apache.cassandra.dht.Murmur3Partitioner;
+import org.apache.cassandra.exceptions.InvalidRequestException;
+import org.apache.cassandra.io.sstable.CQLSSTableWriter;
+
+public class BulkLoader {
+    
+    static String filename;
+    static String outdir;
+    static String schema;
+    static String insert;
+    
+    public static void main(String[] args) throws IOException {
+        if (args.length < 4) {
+            String concat = Arrays.toString(args);
+            System.out.println("Expecting 4 arguments - <keyspace>, <table>, <csv_file>, <output dir>");
+            System.out.println("Got: " + concat.substring(1, concat.length() -1));
+            System.exit(1);
+        }
+        
+        int lineNumber = 0;
+        long start = System.currentTimeMillis();
+        String keyspace = args[0];
+        String table = args[1];
+        filename = args[2];
+        outdir = args[3];
+        
+        System.out.println("keyspace: "+keyspace);
+        System.out.println("table: "+table);
+        System.out.println("filename: "+filename);
+        System.out.println("outdir: "+outdir);
+        
+        // Schema and Insert for bulk load
+        if (table.equals("id_annotation")) {
+            schema = String.format("CREATE TABLE %s.%s (" +
+                                        "id int, " +
+                                        "source text, " +
+                                        "md5 text, " +
+                                        "is_protein boolean, " +
+                                        "single text, " +
+                                        "lca list<text>, " +
+                                        "accession list<text>, " +
+                                        "function list<text>, " +
+                                        "organism list<text>, " +
+                                        "PRIMARY KEY (id, source) " +
+                                    ")", keyspace, table);
+            insert = String.format("INSERT INTO %s.%s (" +
+                                        "id, source, md5, is_protein, single, lca, accession, function, organism" +
+                                    ") VALUES (" +
+                                        "?, ?, ?, ?, ?, ?, ?, ?, ?" +
+                                    ")", keyspace, table);
+        } else if (table.equals("md5_annotation")) {
+            schema = String.format("CREATE TABLE %s.%s (" +
+                                        "md5 text, " +
+                                        "source text, " +
+                                        "is_protein boolean, " +
+                                        "single text, " +
+                                        "lca list<text>, " +
+                                        "accession list<text>, " +
+                                        "function list<text>, " +
+                                        "organism list<text>, " +
+                                        "PRIMARY KEY (md5, source) " +
+                                    ")", keyspace, table);
+            insert = String.format("INSERT INTO %s.%s (" +
+                                        "md5, source, is_protein, single, lca, accession, function, organism" +
+                                    ") VALUES (" +
+                                        "?, ?, ?, ?, ?, ?, ?, ?" +
+                                    ")", keyspace, table);
+        } else if (table.equals("job_md5s")) {
+            schema = String.format("CREATE TABLE %s.%s (" +
+                                        "version int, " +
+                                        "job int, " +
+                                        "source text, " +
+                                        "md5 text, " +
+                                        "data frozen<md5_info>, " +
+                                        "accession list<text>, " +
+                                        "function list<text>, " +
+                                        "organism list<text>, " +
+                                        "PRIMARY KEY ((version, job), source, md5) " +
+                                    ")", keyspace, table);
+            insert = String.format("INSERT INTO %s.%s (" +
+                                        "version, job, source, md5, data, accession, function, organism" +
+                                    ") VALUES (" +
+                                        "?, ?, ?, ?, ?, ?, ?, ?" +
+                                    ")", keyspace, table);
+        } else {
+            System.out.println("Unsupported table type: " + table);
+            System.exit(1);
+        }
+        
+        // magic!
+        Config.setClientMode(true);
+        
+        // Create output directory that has keyspace and table name in the path
+        File outputDir = new File(outdir + File.separator + keyspace + File.separator + table);
+        if (!outputDir.exists() && !outputDir.mkdirs()) {
+            throw new RuntimeException("Cannot create output directory: " + outputDir);
+        }
+        
+        // Prepare SSTable writer
+        CQLSSTableWriter.Builder builder = CQLSSTableWriter.builder();
+        // set output directory
+        builder.inDirectory(outputDir)
+            // set target schema
+            .forTable(schema)
+            // set CQL statement to put data
+            .using(insert)
+            // set partitioner if needed - default is Murmur3Partitioner so set if you use different one.
+            .withPartitioner(new Murmur3Partitioner());
+        CQLSSTableWriter writer = builder.build();
+        
+        // set cvs reader / parser
+        try (
+            CSVReader csvReader = new CSVReader(new FileReader(filename));
+        ) {
+            // Write to SSTable while reading data
+            String[] line;
+            while ((line = csvReader.readNext()) != null) {
+                // We use Java types here based on
+                // http://www.datastax.com/drivers/java/2.0/com/datastax/driver/core/DataType.Name.html#asJavaClass%28%29
+                if (table.equals("id_annotation")) {
+                    writer.addRow(Integer.parseInt(line[0]),
+                                  line[1],
+                                  line[2],
+                                  Boolean.valueOf(line[3]),
+                                  line[4],
+                                  parseList(line[5]),
+                                  parseList(line[6]),
+                                  parseList(line[7]),
+                                  parseList(line[8]));
+                } else if (table.equals("md5_annotation")) {
+                    writer.addRow(line[0],
+                                  line[1],
+                                  Boolean.valueOf(line[2]),
+                                  line[3],
+                                  parseList(line[4]),
+                                  parseList(line[5]),
+                                  parseList(line[6]),
+                                  parseList(line[7]));
+                }
+                // Print nK
+                lineNumber += 1;
+                if (lineNumber % 10000 == 0) {
+                    System.out.println((lineNumber / 1000) + "K");
+                }
+            }
+        } catch (InvalidRequestException | IOException e) {
+            e.printStackTrace();
+        }
+        
+        try {
+            writer.close();
+        } catch (IOException ignore) {}
+        
+        // done
+        long end = System.currentTimeMillis();
+        System.out.println("Successfully parsed " + lineNumber + " lines.");
+        System.out.println("Execution time was " + ((end-start) / 1000) + " seconds.");
+        System.exit(0);
+    }
+    
+    public static List<String> parseList (String listStr) {
+        List<String> aList = new ArrayList<String>();
+        listStr = listStr.trim();
+        if (listStr.isEmpty()) {
+            return aList;
+        }
+        // remove leading and trailing brackets
+        listStr = listStr.substring(1, listStr.length() - 1).trim();
+        if (listStr.isEmpty()) {
+            return aList;
+        }
+        // split by comma
+        String[] parts = listStr.split(",");
+        for (int i=0; i<parts.length; i++) {
+            // add to list, remove leading and trailing single-quotes
+            String item = parts[i].trim();
+            if (item.length() > 1) {
+                aList.add( item.substring(1, item.length() - 1).trim() );
+            } else {
+                aList.add( new String() );
+            }
+        }
+        return aList;
+    }
+    
+}
diff --git a/src/MGRAST/tools/BulkLoader/BulkLoader.sh b/src/MGRAST/tools/BulkLoader/BulkLoader.sh
new file mode 100755
index 00000000..12a5a655
--- /dev/null
+++ b/src/MGRAST/tools/BulkLoader/BulkLoader.sh
@@ -0,0 +1,58 @@
+#!/bin/sh
+
+HELP=0
+CASS_DIR=''
+KEYSPACE=''
+TABLE=''
+INFILE=''
+OUTDIR=''
+JAVA=`which java`
+
+while getopts hc:k:t:i:o: option; do
+    case "${option}"
+        in
+            h) HELP=1;;
+            c) CASS_DIR=${OPTARG};;
+            k) KEYSPACE=${OPTARG};;
+            t) TABLE=${OPTARG};;
+            i) INFILE=${OPTARG};;
+            o) OUTDIR=${OPTARG};;
+    esac
+done
+
+USAGE="Usage: BulkLoader.sh [-h] -c <cassandra dir> -k <keyspace> -t <table> -i <csv_file> -o <output dir>"
+
+# check options
+if [ $HELP -eq 1 ]; then
+    echo $USAGE
+    exit
+fi
+if [ -z "$KEYSPACE" ] || [ -z "$TABLE" ] || [ -z "$INFILE" ]; then
+    echo "[error] missing parameter"
+    echo $USAGE
+    exit
+fi
+if [ ! -f "$INFILE" ]; then
+    echo "[error] file $INFILE does not exist"
+    echo $USAGE
+    exit
+fi
+if [ -z "$OUTDIR" ]; then
+    OUTDIR=/data/sstable
+fi
+if [ -z "$CASS_DIR" ]; then
+    CASS_DIR=/opt/cassandra
+fi
+
+# set classpath
+CLASSPATH=".:$CASS_DIR/conf/cassandra.yaml:$CASS_DIR/lib/*"
+
+# Compile
+echo "compile: javac -cp $CLASSPATH BulkLoader.java"
+javac -cp $CLASSPATH BulkLoader.java
+
+# Import
+echo
+echo "run: $JAVA -ea -cp $CLASSPATH -Xms20G -Xmx20G -Dlog4j.configuration=log4j-tools.properties BulkLoader $KEYSPACE $TABLE $INFILE $OUTDIR"
+$JAVA -ea -cp $CLASSPATH -Xms20G -Xmx20G -Dlog4j.configuration=log4j-tools.properties BulkLoader $KEYSPACE $TABLE $INFILE $OUTDIR
+
diff --git a/src/MGRAST/tools/BulkLoader/opencsv-3.4.jar b/src/MGRAST/tools/BulkLoader/opencsv-3.4.jar
new file mode 100644
index 00000000..2b1edc9e
Binary files /dev/null and b/src/MGRAST/tools/BulkLoader/opencsv-3.4.jar differ
diff --git a/src/MGRAST/Schema/dump_analysis_for_cass.pl b/src/MGRAST/tools/dump_analysis_for_cass.pl
similarity index 71%
rename from src/MGRAST/Schema/dump_analysis_for_cass.pl
rename to src/MGRAST/tools/dump_analysis_for_cass.pl
index 9b9c1ff8..9853c4c0 100755
--- a/src/MGRAST/Schema/dump_analysis_for_cass.pl
+++ b/src/MGRAST/tools/dump_analysis_for_cass.pl
@@ -129,53 +129,42 @@ sub process_batch {
         next unless ($data->{$mid});
         next if (int($ea) > 0);
         $ea = $ea * -1;
-        my $md5 = $data->{$mid}[0]{md5};
-        my $acc = {};
-        my $fun = {};
-        my $org = {};
+        my $md5  = $data->{$mid}[0]{md5};
+        my $srcs = {};
+        my $md5_info = {
+            "abundance"  => $abund,
+            "exp_avg"    => $ea,
+            "ident_avg"  => $ia,
+            "len_avg"    => $la,
+            "exp_stdv"   => $es,
+            "ident_stdv" => $is,
+            "len_stdv"   => $ls,
+            "seek"       => defined($seek) ? $seek : "",
+            "length"     => defined($len) ? $len : "",
+            "is_protein" => $prot ? 'true' : 'false'
+        };
+        
         foreach my $ann (@{$data->{$mid}}) {
-            push @{ $acc->{$ann->{source}} }, cescape($ann->{accession} || "");
-            push @{ $fun->{$ann->{source}} }, cescape($ann->{function} || "");
-            push @{ $org->{$ann->{source}} }, cescape($ann->{organism} || "");
+            next unless ($ann->{source});
+            # source => [[ accession, function, organism ]]
+            $ann->{accession} =~ s/\'/''/g;
+            $ann->{function}  =~ s/\'/''/g;
+            $ann->{organism}  =~ s/\'/''/g;
+            push @{$srcs->{$ann->{source}}}, [ $ann->{accession}, $ann->{function}, $ann->{organism} ];
         }
-        my $out = [ $version, $job,
-                    $ea, $ia, $la, $md5,
-                    $es, $is, $ls,
-                    $abund, $seek, $len,
-                    ($prot ? 'true' : 'false'),
-                    cstring($acc),
-                    cstring($fun),
-                    cstring($org)
-                  ];
-        push @output, $out;
-    }
-    
-    return @output;
-}
-
-sub cescape {
-    my ($text) = @_;
-    $text =~ s/\'/''/g;
-    $text =~ s/\"/\\"/g;
-    return $text;
-}
-
-sub cstring {
-    my ($obj) = @_;
-    my $str = "{";
-    foreach my $key (keys %$obj) {
-        my $has_data = 0;
-        foreach my $v (@{$obj->{$key}}) {
-            if ($v) {
-                $has_data = 1;
-                last;
-            }
-        }
-        if ($has_data) {
-            $str .= "'".$key."':[".join(",", map {"'".$_."'"} @{$obj->{$key}})."],";
+        foreach my $src (keys %$srcs) {
+            my @acc = map { $_->[0] } @{$srcs->{$src}};
+            my @fun = map { $_->[1] } @{$srcs->{$src}};
+            my @org = map { $_->[2] } @{$srcs->{$src}};
+            my $acc = "[".join(",", map { "'".$_."'" } @acc)."]";
+            my $fun = "[".join(",", map { "'".$_."'" } @fun)."]";
+            my $org = "[".join(",", map { "'".$_."'" } @org)."]";
+            $acc =~ s/\"/\\"/g;
+            $fun =~ s/\"/\\"/g;
+            $org =~ s/\"/\\"/g;
+            my $out = [ $version, $job, $src, $md5, $json->encode($md5_info), $acc, $fun, $org ];
+            push @output, $out;
         }
     }
-    chop $str;
-    $str .= "}";
-    return $str;
+    return @output;
 }
diff --git a/src/MGRAST/Schema/dump_annotation_for_cass.pl b/src/MGRAST/tools/dump_annotation_for_cass.pl
similarity index 93%
rename from src/MGRAST/Schema/dump_annotation_for_cass.pl
rename to src/MGRAST/tools/dump_annotation_for_cass.pl
index 81675dd4..26963867 100755
--- a/src/MGRAST/Schema/dump_annotation_for_cass.pl
+++ b/src/MGRAST/tools/dump_annotation_for_cass.pl
@@ -126,7 +126,7 @@
 
 print STDERR "Dumping protein data ...\n";
 foreach my $md5 (@$pmd5s) {
-    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.source, a.id, f.name, o.name FROM md5_protein a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN organisms_ncbi o ON a.organism = o._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
+    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.name, a.id, f.name, o.name FROM md5_protein a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN organisms_ncbi o ON a.organism = o._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
     next unless ($data && @$data);
     my $mid  = $data->[0][0];
     my $srcs = {};
@@ -167,7 +167,7 @@
 
 print STDERR "Dumping rna data ...\n";
 foreach my $md5 (@$rmd5s) {
-    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.source, a.id, f.name, o.name FROM md5_rna a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN organisms_ncbi o ON a.organism = o._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
+    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.name, a.id, f.name, o.name FROM md5_rna a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN organisms_ncbi o ON a.organism = o._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
     next unless ($data && @$data);
     my $mid  = $data->[0][0];
     my $srcs = {};
@@ -208,7 +208,7 @@
 
 print STDERR "Dumping ontology data ...\n";
 foreach my $md5 (@$omd5s) {
-    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.source, a.id, f.name FROM md5_ontology a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
+    my $data = $dbh->selectall_arrayref("SELECT DISTINCT m._id, s.name, a.id, f.name FROM md5_ontology a INNER JOIN md5s m ON a.md5 = m.md5 LEFT OUTER JOIN functions f ON a.function = f._id LEFT OUTER JOIN sources s ON a.source = s._id where a.md5='$md5'");
     next unless ($data && @$data);
     my $mid  = $data->[0][0];
     my $srcs = {};
@@ -228,7 +228,7 @@
         $acc =~ s/\"/\\"/g;
         $fun =~ s/\"/\\"/g;
         print IDUMP join(",", map { '"'.$_.'"' } ($mid, $src, $md5, "true", "", "[]", $acc, $fun, ""))."\n";
-        print MDUMP join(",", map { '"'.$_.'"' } ($md5, $src, "true", "", "[]", $acc, $fun. ""))."\n";
+        print MDUMP join(",", map { '"'.$_.'"' } ($md5, $src, "true", "", "[]", $acc, $fun, ""))."\n";
     }
 }
 
diff --git a/src/MGRAST/Schema/test_cass.py b/src/MGRAST/tools/test_cass.py
similarity index 100%
rename from src/MGRAST/Schema/test_cass.py
rename to src/MGRAST/tools/test_cass.py
diff --git a/src/PPO/DBObjectCache.pm b/src/PPO/DBObjectCache.pm
index 3493b89b..94f67102 100644
--- a/src/PPO/DBObjectCache.pm
+++ b/src/PPO/DBObjectCache.pm
@@ -137,7 +137,7 @@ sub object_to_cache {
   }
 
   my $id = $object->_id();
-  unless (ref $self->{'_cache'}->{$backend}->{$db}->{$class}->{$id}) {
+  unless ($id && ref($self->{'_cache'}->{$backend}->{$db}->{$class}->{$id})) {
     $self->{'_cache'}->{$backend}->{$db}->{$class}->{$id} = $object;
 
     if (scalar(@{$self->{'recent'}}) > SIZE) {
@@ -216,8 +216,7 @@ sub delete_object {
   my $class = $object->_class();
   my $id = $object->_id();
 
-  if (exists $self->{'_cache'}->{$backend}->{$db}->{$class} and
-      exists $self->{'_cache'}->{$backend}->{$db}->{$class}->{$id}) {
-    delete $self->{'_cache'}->{$backend}->{$db}->{$class}->{$id};
+  if ($id && exists($self->{'_cache'}->{$backend}->{$db}->{$class}) && exists($self->{'_cache'}->{$backend}->{$db}->{$class}->{$id})) {
+      delete $self->{'_cache'}->{$backend}->{$db}->{$class}->{$id};
   }
 }