UploadToDatabase

#!/bin/bash
#**************************************************************
#  EPPIC Pre-Computation Script
#  Author : Kumaran Baskaran
#  Date   : 05/07/2013
#***************************************************************
set -e
help="\t\tUsage : $0 \n
	\t\t [-d <dir-path> : Directory contains downloaded files\n
	\t\t -j <file-path> : eppic jar file path\n
	\t\t -n <name> : database name (usually uniprot relese name)]\n
\t\tExample : ./Download files -o ~/Downloads "

if (($# == 0)); then
	echo "`date +%d/%m/%y-%H:%M:%S` ERROR: Script requires arguments" >&2
	echo -e $help
	exit 1;
fi

while getopts :d:j:n:h opt
do
	case $opt in 
	d) DOWNLOAD=$OPTARG;;
	j) EPPIC_JAR=$OPTARG;;
	n) DATABASE_NAME=$OPTARG;;
	h) echo -e $help;;
	\?) echo "`date +%d/%m/%y-%H:%M:%S` ERROR: Invalid option: -$OPTARG" >&2
		exit 1;;
	:) echo "`date +%d/%m/%y-%H:%M:%S` ERROR: Option -$OPTARG requires an argument." >&2
		exit 1;;
	esac
done

if [ -z $DOWNLOAD ] || [ -z $EPPIC_JAR ] || [ -z $DATABASE_NAME ]
then
	echo -e "\n`date +%d/%m/%y-%H:%M:%S` ERROR: ---------Some Options not specified correctly -------" >&2
	echo -e $help
	exit 1
fi

if [ ! -e $EPPIC_JAR ]
then
	echo -e "\t `date +%d/%m/%y-%H:%M:%S` ERROR: Couldn't find $EPPIC_JAR ..........." >&2
	exit 1
fi


if [ ! -e $DOWNLOAD/uniref100.xml.gz ]
then 
	echo "`date +%d/%m/%y-%H:%M:%S` ERROR: Couldn't find $DOWNLOAD/uniref100.xml.gz" >&2
	exit 1
fi

if [ ! -e $DOWNLOAD/taxonomy-all.tab ]
then 
	echo "`date +%d/%m/%y-%H:%M:%S` ERROR: Couldn't find $DOWNLOAD/taxonomy-all.tab" >&2
	exit 1
fi

echo "`date +%d/%m/%y-%H:%M:%S` INFO: Creating output tab and cluster members tab file.."
cmd="java -cp $EPPIC_JAR owl.core.connections.UnirefXMLParser"

$cmd $DOWNLOAD/uniref100.xml.gz $DOWNLOAD/uniref100.tab $DOWNLOAD/uniref100.clustermembers.tab || exit 1

echo "`date +%d/%m/%y-%H:%M:%S` INFO: Creating a new database named $DATABASENAME .. "

mysql <<EOF
DROP DATABASE IF EXISTS $DATABASE_NAME;
CREATE DATABASE $DATABASE_NAME;
EOF

#---------------------------------------------------------------
# Create uniprot table

table=uniprot

# note 1: uniprot ids are 6 char, but for isoforms they can be 8 or 9 chars (hyphen and number additionally)
# note 2: primary keys are created later in indexing step so that load data is a lot quicker
# note 3: in first versions of uniprot there are duplicate uniprot ids, e.g. Q9WHW0 in version 1.0
#      Because of that we need to create the uniprot table with a primary key in uniparc id so that 
#      on loading the duplicates will be eliminated. Also a primary key on uniprot_clusters (member)
#      The other indexes are created in another step


echo "`date +%d/%m/%y-%H:%M:%S` INFO: Creating $table tables.."
mysql $DATABASE_NAME <<EOF
DROP TABLE IF EXISTS $table;
DROP TABLE IF EXISTS ${table}_clusters;
CREATE TABLE $table (
 id varchar(23),
 uniprot_id varchar(15),
 uniparc_id char(13) PRIMARY KEY,
 tax_id int,
 sequence text
);
CREATE TABLE ${table}_clusters (
 representative varchar(15),
 member varchar(15) PRIMARY KEY,
 tax_id int
);

EOF

#------------------------------------------------------
# Load Data in uniprot table

echo "`date +%d/%m/%y-%H:%M:%S` INFO: Loading Data in to $table table "
mysql --enable-local $DATABASE_NAME <<EOF
LOAD DATA LOCAL INFILE '$DOWNLOAD/uniref100.tab' INTO TABLE $table;
SHOW WARNINGS;
LOAD DATA LOCAL INFILE '$DOWNLOAD/uniref100.clustermembers.tab' INTO TABLE ${table}_clusters;
SHOW WARNINGS;
EOF
#-------------------------------------------------------------
##################################################
# Creating Uniprot Database Indices
#--------------------------------------------------------------
echo "Creating indices.."
mysql $DATABASE_NAME <<EOF
CREATE INDEX UNIPROTID_IDX ON uniprot (uniprot_id);
EOF
echo "`date +%d/%m/%y-%H:%M:%S` INFO: Finished loading data into database."