#!/usr/bin/perl
use SOAP::Lite;

#connect to the SubLoc SOAP server via the WSDL file
$wsdl = 'http://www.bioinfo.tsinghua.edu.cn/~tigerchen/SubLoc.wsdl';
$serv = SOAP::Lite->service($wsdl);
$matched = 0;

###################################################################
########### id_search by SWISSPROT ID
$res = $serv -> id_search("SWISSPROT","O03978");
# get the number of matched entries: $#$res
$matched = $#$res+1;
print "$matched matched entries are found!\n";
# get all entries' ID SQ LC CX DE OS
# SQ=sequence, LC=location, CX=cross database id, DE=description, OS=organism
for(my $index=0;$index<=$#$res;$index++){
    print $res->[$index]->{ID};
    print $res->[$index]->{SQ};
    print $res->[$index]->{DE};
    print $res->[$index]->{LC};
    print $res->[$index]->{OS};
    print $res->[$index]->{CX};
}
########### id_search by DBSubLoc ID
$res = $serv -> id_search("DBSubLoc","10043258");
# get the number of matched entries: $#$res
$matched = $#$res+1;
print "$matched matched entries are found!\n";
# get all entries' ID SQ LC CX DE OS
# SQ=sequence, LC=location, CX=cross database id, DE=description, OS=organism
for(my $index=0;$index<=$#$res;$index++){
    print $res->[$index]->{ID};
    print $res->[$index]->{SQ};
    print $res->[$index]->{DE};
    print $res->[$index]->{LC};
    print $res->[$index]->{OS};
    print $res->[$index]->{CX};
}
########### id_search by GO ID
# Caution! Search by GO ID may return many many entries, please be patient, thanks
$res = $serv -> id_search("GO","0005739");
# get the number of matched entries: $#$res
$matched = $#$res+1;
print "$matched matched entries are found!\n";
# get all entries' ID SQ LC CX DE OS
# SQ=sequence, LC=location, CX=cross database id, DE=description, OS=organism
for(my $index=0;$index<=$#$res;$index++){
    print $res->[$index]->{ID};
    print $res->[$index]->{SQ};
    print $res->[$index]->{DE};
    print $res->[$index]->{LC};
    print $res->[$index]->{OS};
    print $res->[$index]->{CX};
}


#######################################################################
############### name_search
$res=$serv -> name_search("p53");
# get the number of matched entries: $#$res
$matched = $#$res +1;
print "$matched matched entries are found!\n";
# get all entries' ID SQ LC CX DE OS
# SQ=sequence, LC=location, CX=cross database id, DE=description, OS=organism
for(my $index=0;$index<=$#$res;$index++){
    print $res->[$index]->{ID};
    print $res->[$index]->{SQ};
    print $res->[$index]->{DE};
    print $res->[$index]->{LC};
    print $res->[$index]->{OS};
    print $res->[$index]->{CX};
}

##########################################################################
############### blast_search one sequence
$seq="TKRFFNKNNRLNKGYAKTFSINEPDNNFYRKKFEHILPPVDLISEYESIYPGTLQELMHMAQKEQAHKHAIDLKNLKIQERIAKLTRICLLIFGIGLVVLIFLKLLK";
$res = $serv -> blast_search("all","full",$seq,1);
# we provide two kinds of output formats: original BLAST output and parsed results by BioPerl's BPlite, users could choose one they like.
# get the original output of BLAST program, you could use BioPerl to parse it quite easily. There are many example scripts on the Internet.
print $res->{plain};
# or you can get the parsed result (using BioPerl's BPlite)
# the parsed output's fields, every field is separated by a TAB.
#HIT_NUM    SCORE   BITS    PERCENT EXPECT  SUBJECT_NAME    IDENTITIES	MATCH_LENGTH	QUERY_START QUERY_END	SUBJECT_START	SUBJECT_END
print $res->{parsed};
# the parsed output's fields
#HIT_NUM    SCORE   BITS    PERCENT EXPECT  SUBJECT_NAME    IDENTITIES	MATCH_LENGTH	QUERY_START QUERY_END	SUBJECT_START	SUBJECT_END

###################################################################
############### SubLoc prediction by SVM
# for more details about the method, please see:
#Hua, S. and Sun, Z. (2001). Support vector machine approach for protein subcellular localization prediction. Bioinformatics 17, 721-8.
# for prokaryotic organisms 
$seq="TKRFFNKNNRLNKGYAKTFSINEPDNNFYRKKFEHILPPVDLISEYESIYPGTLQELMHMAQKEQAHKHAIDLKNLKIQERIAKLTRICLLIFGIGLVVLIFLKLLK";
$pre= $serv -> pro_predict($seq);
# get sequence
print $pre->{Seq}."\n";
# get predicted location
print $pre->{Prediction}."\n";
# get expected accuracy
print $pre->{ExpectAcc}."\n";
# get reliability index
print $pre->{RI}."\n";

# for eukaryotic organisms
$pre= $serv -> eu_predict($seq);
# get sequence
print $pre->{Seq}."\n";
# get predicted location
print $pre->{Prediction}."\n";
# get expected accuracy
print $pre->{ExpectAcc}."\n";
# get reliability index
print $pre->{RI}."\n";

######################################################################
################# SubLoc prediction by PSORT
# for more details about PSORT, please refer:
# Nakai, K. and Horton, P. (1999). PSORT: a program for detecting sorting signals in proteins and
# predicting their subcellular localization. Trends Biochem Sci 24, 34-6.

$pre= $serv -> psort_predict($seq);
# get predicted location
print $pre->{Prediction}."\n";
# get all output information of PSORT
print $pre->{Detail}."\n";


########################################################################
################## Submit new entries by users
$ID="test";
$DE="pseudo-entry for test";
$OS="human";
$LC="cytoplasmic";
$CX="unknown";
$SQ="XXXXXXXXXXXXXXXXXX";
$feed = $serv -> feed_entry($ID,$DE,$OS,$LC,$CX,$SQ);
print $feed;

## if you have a fasta file containing many sequences, you may use following code:
## need BioPerl, if not installed, please visit www.bioperl.org to download and install it.
use Bio::SeqIO;
$filename='/path/to/your/fasta/file';
my $fasta = Bio::SeqIO -> new('-fh'=>$filename,'-format'=>'fasta');
while (my $seq = $fasta->next_seq){
	# do what you want to do, for example:
	$pre=$serv -> psort_predict($seq);
	# .......
	$blast=$serv-> blast_search('all','full',$seq,1);
	# .......
}