Velvet is a de novo genomic assembler specially designed for short read

sequencing technologies, such as Solexa or 454, developed by Daniel Zerbino
and Ewan Birney at the European Bioinformatics Institute (EMBL-EBI).

Citation:

Velvet: algorithms for de novo short read assembly using de Bruijn graphs.
D.R. Zerbino and E. Birney. Genome Research 18: 821-829 (2008)

WWW: http://www.ebi.ac.uk/~zerbino/velvet/

PR:		140147
Submitted by:	Motomichi Matsuzaki <mzaki@m.u-tokyo.ac.jp>

6 files changed, 243 insertions, 0 deletions

diff --git a/biology/velvet/Makefile b/biology/velvet/Makefile
new file mode 100644
index 000000000000..5fcdca3ac0a5
--- /dev/null
+++ b/biology/velvet/Makefile
@@ -0,0 +1,89 @@
+# New ports collection makefile for:	velvet
+# Date created:		27 Oct 2009
+# Whom:			Motomichi Matsuzaki <mzaki@m.u-tokyo.ac.jp>
+#
+# $FreeBSD$
+#
+
+PORTNAME=	velvet
+PORTVERSION=	0.7.55
+CATEGORIES=	biology
+MASTER_SITES=	http://www.ebi.ac.uk/~zerbino/velvet/
+DISTNAME=	${PORTNAME}_${PORTVERSION}
+EXTRACT_SUFX=	.tgz
+
+MAINTAINER=	mzaki@m.u-tokyo.ac.jp
+COMMENT=	Sequence assembler for very short reads
+
+USE_GMAKE=	yes
+ALL_TARGET=	default
+
+MAXKMERLENGTH?=	31
+MAKE_ENV+=	MAXKMERLENGTH=${MAXKMERLENGTH}
+
+USE_PERL5_RUN=	yes
+
+BINARIES=	velvetg velveth
+SCRIPTS=	contrib/VelvetOptimiser/VelvetOptimiser.pl \
+		contrib/afg_handling/asmbly_splitter.pl \
+		contrib/afg_handling/snp_view.pl \
+		contrib/estimate-exp_cov/velvet-estimate-exp_cov.pl \
+		contrib/fasta2agp/fasta2agp.pl \
+		contrib/extractContigReads/extractContigReads.pl \
+		contrib/observed-insert-length.pl/observed-insert-length.pl \
+		contrib/shuffleSequences_fasta/shuffleSequences_fasta.pl
+PERLMOD_DIRS=	contrib/VelvetOptimiser/VelvetOpt
+PERLMOD_FILES=	contrib/VelvetOptimiser/VelvetOpt/Assembly.pm \
+		contrib/VelvetOptimiser/VelvetOpt/Utils.pm \
+		contrib/VelvetOptimiser/VelvetOpt/gwrap.pm \
+		contrib/VelvetOptimiser/VelvetOpt/hwrap.pm
+
+.if !defined(WITHOUT_PYTHON)
+USE_PYTHON=	yes
+SCRIPTS+=	contrib/layout/graph2.py
+.endif
+
+.for f in ${BINARIES} ${SCRIPTS}
+PLIST_FILES+=	bin/${f:T}
+.endfor
+.for d in ${PERLMOD_DIRS}
+PLIST_DIRS+=	%%SITE_PERL%%/${d:T}
+.endfor
+.for f in ${PERLMOD_FILES}
+PLIST_FILES+=	%%SITE_PERL%%/${f:H:T}/${f:T}
+.endfor
+
+PORTDOCS=	ChangeLog Manual.pdf
+
+post-patch:
+.for f in ${SCRIPTS}
+	${REINPLACE_CMD} -e s,#!/usr/bin/perl,${PERL}, ${WRKSRC}/${f}
+.endfor
+.if !defined(WITHOUT_PYTHON)
+	(IFS=''; ${REINPLACE_CMD} -e `${ECHO} '1i\'; ${ECHO} '#!${PYTHON_CMD}'` ${WRKSRC}/contrib/layout/graph2.py)
+.endif
+
+post-build:
+	${REINPLACE_CMD} -e s,%%MAXKMERLENGTH%%,${MAXKMERLENGTH},g ${WRKSRC}/contrib/VelvetOptimiser/VelvetOptimiser.pl
+
+do-install:
+.for f in ${BINARIES}
+	${INSTALL_PROGRAM} ${WRKSRC}/${f} ${PREFIX}/bin
+.endfor
+.for f in ${SCRIPTS}
+	${INSTALL_SCRIPT} ${WRKSRC}/${f} ${PREFIX}/bin/${f:T}
+.endfor
+.for d in ${PERLMOD_DIRS}
+	${MKDIR} ${PREFIX}/${SITE_PERL_REL}/${d:T}
+.endfor
+.for f in ${PERLMOD_FILES}
+	${INSTALL_DATA} ${WRKSRC}/${f} ${PREFIX}/${SITE_PERL_REL}/${f:H:T}/${f:T}
+.endfor
+.if !defined(NOPORTDOCS)
+	${MKDIR} ${DOCSDIR}
+.for f in ${PORTDOCS}
+	${INSTALL_MAN} ${WRKSRC}/${f} ${DOCSDIR}
+.endfor
+.endif
+
+.include <bsd.port.mk>
diff --git a/biology/velvet/distinfo b/biology/velvet/distinfo
new file mode 100644
index 000000000000..9e4392f835cd
--- /dev/null
+++ b/biology/velvet/distinfo
@@ -0,0 +1,3 @@
+MD5 (velvet_0.7.55.tgz) = ea99c0b157b4195c313f93bd9f77ecb3
+SHA256 (velvet_0.7.55.tgz) = 315331f418c2de89c6c871cea3759d947320c0adbc763fea290ec78ece69b614
+SIZE (velvet_0.7.55.tgz) = 3475289
diff --git a/biology/velvet/files/patch-Makefile b/biology/velvet/files/patch-Makefile
new file mode 100644
index 000000000000..554927aa6fd8
--- /dev/null
+++ b/biology/velvet/files/patch-Makefile
@@ -0,0 +1,38 @@
+--- Makefile.orig	2009-09-02 22:21:01.000000000 +0900
++++ Makefile	2009-10-27 22:05:49.000000000 +0900
+@@ -1,18 +1,18 @@
+-CC = gcc
+-CFLAGS = -Wall
++CC ?= gcc
++CFLAGS += -Wall
+ DEBUG = -g
+-LDFLAGS = -lm
+-OPT = -O3
+-MAXKMERLENGTH=31
+-CATEGORIES=2
++#LDFLAGS = -lm
++#OPT = -O3
++MAXKMERLENGTH?=31
++CATEGORIES?=2
+ DEF = -D MAXKMERLENGTH=$(MAXKMERLENGTH) -D CATEGORIES=$(CATEGORIES)
+ 
+ Z_LIB_DIR=third-party/zlib-1.2.3
+ Z_LIB_FILES=$(Z_LIB_DIR)/*.o
+ 
+ # Mac OS users: uncomment the following lines
+-# Z_LIB_FILES=
+-# LDFLAGS = -lm -lz
++Z_LIB_FILES=
++LDFLAGS += -lm -lz
+ # CFLAGS = -Wall -m64
+ 
+ # Sparc/Solaris users: uncomment the following line
+@@ -32,7 +32,7 @@
+ 	-rm obj/*.o obj/dbg/*.o 
+ 
+ zlib : 
+-	cd $(Z_LIB_DIR); ./configure; make; rm minigzip.o; rm example.o
++#	cd $(Z_LIB_DIR); ./configure; make; rm minigzip.o; rm example.o
+ 
+ velveth : $(OBJ) 
+ 	$(CC) $(CFLAGS) $(OPT) $(LDFLAGS) -o velveth obj/tightString.o obj/run.o obj/recycleBin.o obj/splay.o obj/splayTable.o obj/readSet.o obj/crc.o obj/utility.o obj/kmer.o $(Z_LIB_FILES)
diff --git a/biology/velvet/files/patch-contrib-VelvetOptimiser-VelvetOptimiser.pl b/biology/velvet/files/patch-contrib-VelvetOptimiser-VelvetOptimiser.pl
new file mode 100644
index 000000000000..90bd168cc57e
--- /dev/null
+++ b/biology/velvet/files/patch-contrib-VelvetOptimiser-VelvetOptimiser.pl
@@ -0,0 +1,92 @@
+--- contrib/VelvetOptimiser/VelvetOptimiser.pl.orig	2009-09-14 23:44:19.000000000 +0900
++++ contrib/VelvetOptimiser/VelvetOptimiser.pl	2009-10-30 22:13:45.000000000 +0900
+@@ -44,9 +44,10 @@
+ my @hashvals;
+ my %assemblies;
+ my $readfile;
++my $MAXKMERLENGTH = %%MAXKMERLENGTH%%;
+ my $logfile = "logfile.txt";
+ my $hashs = 19;
+-my $hashe = 129;
++my $hashe = $MAXKMERLENGTH;
+ my $ass_num = 1;
+ my $interested = 1;
+ 
+@@ -311,26 +312,26 @@
+     my $ass = shift;
+     print STDERR "\tPlease type in the insert length for the short reads: ";
+     #my $len = <>;
+-    my $len = "200"; #do not ask for insert size!
++    my $len = "auto"; #do not ask for insert size!
+     chomp($len);
+-    while($len =~ /\D+/){
+-        print STDERR "\tThe length needs to be a number, please re-enter: ";
+-        $len = <>;
+-        chomp($len);
+-    }
++    #while($len =~ /\D+/){
++    #    print STDERR "\tThe length needs to be a number, please re-enter: ";
++    #    $len = <>;
++    #    chomp($len);
++    #}
+     print STDERR strftime("%b %e %H:%M:%S", localtime), " Running assembly with short insert length $len\n";
+     print OUT strftime("%b %e %H:%M:%S", localtime), " Running assembly with short insert length $len\n";
+ 
+     #re-write the pstringg with the new velvetg command..
+-    my $vg = $ass->{pstringg};
+-    if($vg =~ /ins_length /){
+-        $vg =~ s/ins_length\s+\d+/ins_length $len/;
+-    }
+-    else {
+-        $vg .= " -ins_length $len";
+-    }
++    #my $vg = $ass->{pstringg};
++    #if($vg =~ /ins_length /){
++    #    $vg =~ s/ins_length\s+\d+/ins_length $len/;
++    #}
++    #else {
++    #    $vg .= " -ins_length $len";
++    #}
+ 
+-    $ass->{pstringg} = $vg;
++    #$ass->{pstringg} = $vg;
+     my $worked = VelvetOpt::gwrap::objectVelvetg($ass);
+     if($worked){
+         $ass->getAssemblyDetails();
+@@ -347,8 +348,8 @@
+ my $usage = "\nVelvetOptimiser.pl: A script to run the Velvet assembler and optimise its output. Simon Gladman - CSIRO 2008, 2009.\n\n";
+ $usage .= "Usage: VelvetOptimiser.pl <-f 'velveth parameters'> [-s <hash start>] [-e <hash end>] [-a <yes>]\n\n";
+ $usage .= "Where:\t<-f 'velveth parameters'> is the parameter line normally passed to velveth in quotes.\n";
+-$usage .= "\t-s <hash start> The hash value you want velvet to start looking from. Default: 19. MUST BE ODD > 0 & <=31!\n";
+-$usage .= "\t-e <hash end> The hash value you want velvet to stop looking at. Default: 31. MUST BE ODD AND > START & <= 31!\n";
++$usage .= "\t-s <hash start> The hash value you want velvet to start looking from. Default: 19. MUST BE ODD > 0 & <=$MAXKMERLENGTH!\n";
++$usage .= "\t-e <hash end> The hash value you want velvet to stop looking at. Default: $MAXKMERLENGTH. MUST BE ODD AND > START & <= $MAXKMERLENGTH!\n";
+ $usage .= "\t-a <yes> The final optimised assembly will include read tracking and amos file outputs (however, intermediate assemblies won't.)\n";
+ $usage .= "\nIf the optimizer requires an insert length for some paired end data, it will ask for it when it gets to the optimization step.\n";
+ 
+@@ -390,9 +391,9 @@
+     $hashs = $opts{'s'};
+        print STDERR "\n\t\t's-Param' is >$hashs<\n";
+     unless($hashs =~ /^\d+$/){ die "\tFatal error! Start hash not a number!\n$usage";}
+-    if($hashs > 129){
+-        print STDERR "\tStart hash value too high.  New start hash value is 129.\n";
+-        $hashs = 129;
++    if($hashs > $MAXKMERLENGTH){
++        print STDERR "\tStart hash value too high.  New start hash value is $MAXKMERLENGTH.\n";
++        $hashs = $MAXKMERLENGTH;
+     }
+     if(!&isOdd($hashs)){
+         $hashs = $hashs - 1;
+@@ -404,9 +405,9 @@
+ if($opts{'e'}){
+     $hashe = $opts{'e'};
+     unless($hashe =~ /^\d+$/){ die "\tFatal error! End hash not a number!\n$usage";}
+-    if($hashe > 129 || $hashe < 1){
+-        print STDERR "\tEnd hash value not in workable range.  New end hash value is 129.\n";
+-        $hashe = 129;
++    if($hashe > $MAXKMERLENGTH || $hashe < 1){
++        print STDERR "\tEnd hash value not in workable range.  New end hash value is $MAXKMERLENGTH.\n";
++        $hashe = $MAXKMERLENGTH;
+     }
+     if($hashe < $hashs){
+         print STDERR "\tEnd hash value lower than start hash value.  New end hash value = $hashs.\n";
diff --git a/biology/velvet/files/patch-src-run.c b/biology/velvet/files/patch-src-run.c
new file mode 100644
index 000000000000..7ecd02708351
--- /dev/null
+++ b/biology/velvet/files/patch-src-run.c
@@ -0,0 +1,11 @@
+--- src/run.c.orig	2009-09-16 20:11:59.000000000 +0900
++++ src/run.c	2009-10-27 22:27:27.000000000 +0900
+@@ -152,7 +152,7 @@
+ 						   double_strand);
+ 
+ 	destroySplayTable(splayTable);
+-	closedir(dir);
++	if (dir) closedir(dir);
+ 	free(filename);
+ 	free(buf);
+ 
diff --git a/biology/velvet/pkg-descr b/biology/velvet/pkg-descr
new file mode 100644
index 000000000000..da5c56fa485a
--- /dev/null
+++ b/biology/velvet/pkg-descr
@@ -0,0 +1,10 @@
+Velvet is a de novo genomic assembler specially designed for short read
+sequencing technologies, such as Solexa or 454, developed by Daniel Zerbino
+and Ewan Birney at the European Bioinformatics Institute (EMBL-EBI).
+
+Citation:
+
+Velvet: algorithms for de novo short read assembly using de Bruijn graphs.
+D.R. Zerbino and E. Birney. Genome Research 18: 821-829 (2008)
+
+WWW: http://www.ebi.ac.uk/~zerbino/velvet/