From 9f661ebc21fd09708d824c79914b4d7d25f292b8 Mon Sep 17 00:00:00 2001
From: elserj <elserj@localhost>
Date: Thu, 23 Sep 2010 21:54:22 +0000
Subject: [PATCH] Added species db loading script to svn.  This script will
 parse the fasta files and put the gene_id, gene_header, and sequences into a
 database

svn path=/; revision=50
---
 interactome_scripts/load_species_db.pl | 103 +++++++++++++++++++++++++
 1 file changed, 103 insertions(+)
 create mode 100755 interactome_scripts/load_species_db.pl

diff --git a/interactome_scripts/load_species_db.pl b/interactome_scripts/load_species_db.pl
new file mode 100755
index 0000000..e5aad70
--- /dev/null
+++ b/interactome_scripts/load_species_db.pl
@@ -0,0 +1,103 @@
+#!/usr/bin/perl
+
+###############################################################
+#  Written by Justin Elser 2/22/10                            #
+#                                                             #
+#  This program takes all fasta files in the current working  #
+#    directory and puts them in a database named protein      #
+#       sequences                                             #
+#                                                             #
+#  Initial version 0.1                                        #
+#     Shouldn't have to change much once this is done         #
+#         except to maybe add more species detection          #
+#         which is actually done in the pulled in subs        #
+#                                                             #
+###############################################################
+
+use strict;
+use warnings;
+
+use DBI;
+use Term::Screen::ReadLine;
+
+# pull in the find_species and find_gene common subroutines
+require "$ENV{HOME}/scripts/jaiswallab/interactome_scripts/find_species.pl";
+
+
+# define the database handle to be used 
+
+my $screen = Term::Screen::ReadLine->new();
+	# clear the screen
+	$screen->clrscr;
+	# ask for username
+	$screen->at(0,0)->puts("Username: ");
+	my $username = $screen->readline(ROW => 0, COL=>11);
+
+	# ask for password, replace character presses with stars
+	$screen->at(1,0)->puts("Password: ");
+	my $password = $screen->readline(ROW => 1, COL => 11, PASSWORD => 1);
+
+	$screen->at(2,0);
+	undef $screen;
+
+my $dbh = DBI->connect('DBI:mysql:protein_sequences;host=floret.cgrb.oregonstate.edu', $username, $password,
+	{ RaiseError=> 1, AutoCommit=>1 }
+	) or die "Failed to connect to database: $DBI::errstr";
+
+
+my @files = glob("*.fa");
+
+foreach my $file (@files) {
+	my $species = find_species($file);
+	print "on species $species\n";
+	
+	# create the database table
+	my $safe_table = $dbh->quote_identifier($species);
+	
+	$dbh->do("drop table if exists $safe_table");
+	$dbh->do("CREATE TABLE $safe_table (
+        	`gene_id` VARCHAR( 255 ) NOT NULL ,
+		`gene_header` TEXT NOT NULL ,
+		`sequence` TEXT NOT NULL ,
+		UNIQUE ( `gene_id` )
+		) TYPE = MYISAM");
+		
+	# and the statement handler to do the inserts
+	my $insert_sth = $dbh->prepare("insert into $safe_table (gene_id, gene_header, sequence) values (?,?,?)");
+	
+	# create the variables
+	my %seq_hash; # keys is $gene, values are $seq and $gene_header
+	my $gene;
+	my $seq;
+	
+	open(in_file,$file);
+	
+	while(<in_file>) {
+		my $gene_header;
+		my $line = $_;
+		chomp $line;
+		if($line =~ /^\>/) {
+			if(defined($gene)) {
+				$seq_hash{$gene}->{'sequence'} = $seq;
+				$seq_hash{$gene}->{'gene_header'} = $gene_header;
+			}
+			$gene_header = $line;
+			$gene = find_gene($gene_header,$species);
+			$gene =~ s/^\>//;
+			$seq = "";
+		}else{
+			$seq = "$seq"."$line";
+		}
+	}
+	
+	# add the last gene to the hash
+	$seq_hash{$gene}->{'sequence'} = $seq;
+	$seq_hash{$gene}->{'gene_header'} = $gene_header;
+
+	foreach my $key (keys %seq_hash) {
+		$insert_sth->execute($key,$seq_hash{$key}->{'gene_header'},$seq_hash{$key}->{'sequence'});
+	}
+}
+
+
+
-- 
2.34.1