Re: [PATCH] Add preseq.

[Ricardo Wurmus]

>>> Why leave the headers?  Isn’t it risky to leave headers that may be
>>> different from the Samtools that we’re actually linking against?
>>
>> The problem is with our Samtools package(s).  They don't install
>> headers.  I'd better update them to install the headers; then I could
>> just delete the complete "samtools" directory.
>
> Yes that would be best, if that’s fine with you.

Attached is a patch to add a build phase to samtools to install the
headers into $out/include/samtools/.

The second patch is just the updated patch for adding preseq.

>From b1ba27153faca9bdf46d42046c43f8c60f945368 Mon Sep 17 00:00:00 2001
From: Ricardo Wurmus <address@hidden>
Date: Thu, 2 Jul 2015 15:53:47 +0200
Subject: [PATCH 1/2] gnu: samtools: Install headers.

* gnu/packages/bioinformatics.scm (samtools)[arguments]: Add phase
  "install-headers" to install header files.
---
 gnu/packages/bioinformatics.scm | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 4a1804a..ce2f55b 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -1824,6 +1824,10 @@ distribution, coverage uniformity, strand specificity, 
etc.")
        ;; systems.
        #:tests? ,(string=? (or (%current-system) (%current-target-system))
                            "x86_64-linux")
+       #:modules ((ice-9 ftw)
+                  (ice-9 regex)
+                  (guix build gnu-build-system)
+                  (guix build utils))
        #:make-flags (list "LIBCURSES=-lncurses"
                           (string-append "prefix=" (assoc-ref %outputs "out")))
        #:phases
@@ -1848,7 +1852,18 @@ distribution, coverage uniformity, strand specificity, 
etc.")
            (let ((lib (string-append (assoc-ref outputs "out") "/lib")))
              (mkdir-p lib)
              (copy-file "libbam.a" (string-append lib "/libbam.a"))))
-         (alist-delete 'configure %standard-phases)))))
+         (alist-cons-after
+          'install 'install-headers
+          (lambda* (#:key outputs #:allow-other-keys)
+            (let ((include (string-append (assoc-ref outputs "out")
+                                          "/include/samtools/")))
+              (mkdir-p include)
+              (for-each (lambda (file)
+                          (copy-file file (string-append include
+                                                         (basename file))))
+                        (scandir "." (lambda (name) (string-match "\\.h$" 
name))))
+              #t))
+          (alist-delete 'configure %standard-phases))))))
     (native-inputs `(("pkg-config" ,pkg-config)))
     (inputs `(("ncurses" ,ncurses)
               ("perl" ,perl)
-- 
2.1.0

>From e3bd0a861121c441b44d77d30f2aaddb766b49c2 Mon Sep 17 00:00:00 2001
From: Ricardo Wurmus <address@hidden>
Date: Tue, 23 Jun 2015 10:54:00 +0200
Subject: [PATCH 2/2] gnu: Add preseq.

* gnu/packages/bioinformatics.scm (preseq): New variable.
---
 gnu/packages/bioinformatics.scm                    | 56 ++++++++++++++++++++++
 .../patches/preseq-1.0.2-install-to-PREFIX.patch   | 37 ++++++++++++++
 .../patches/preseq-1.0.2-link-with-libbam.patch    | 33 +++++++++++++
 3 files changed, 126 insertions(+)
 create mode 100644 gnu/packages/patches/preseq-1.0.2-install-to-PREFIX.patch
 create mode 100644 gnu/packages/patches/preseq-1.0.2-link-with-libbam.patch

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index ce2f55b..27b847e 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -2137,6 +2137,62 @@ subsequent visualization, annotation and storage of 
results.")
     ;; LGPLv2.1+
     (license (list license:gpl2 license:lgpl2.1+))))
 
+(define-public preseq
+  (package
+    (name "preseq")
+    (version "1.0.2")
+    (source (origin
+              (method url-fetch)
+              (uri
+               (string-append "http://smithlabresearch.org/downloads/preseq-";
+                              version ".tar.bz2"))
+              (sha256
+               (base32 "0r7sw07p6nv8ygvc17gd78lisbw5336v3vhs86b5wv8mw3pwqksc"))
+              (patches (list (search-patch 
"preseq-1.0.2-install-to-PREFIX.patch")
+                             (search-patch 
"preseq-1.0.2-link-with-libbam.patch")))
+              (modules '((guix build utils)))
+              (snippet
+               ;; Remove bundled samtools.
+               '(delete-file-recursively "preseq-master/samtools"))))
+    (build-system gnu-build-system)
+    (arguments
+     `(#:tests? #f ;no "check" target
+       #:phases
+       (modify-phases %standard-phases
+         (add-after
+          'unpack 'enter-dir
+          (lambda _
+            (chdir "preseq-master")
+            #t))
+         (add-after
+          'unpack 'use-samtools-headers
+          (lambda _
+            (substitute* '("smithlab_cpp/SAM.cpp"
+                           "smithlab_cpp/SAM.hpp")
+              (("sam.h") "samtools/sam.h"))
+            #t))
+         (delete 'configure))
+       #:make-flags (list (string-append "PREFIX="
+                                         (assoc-ref %outputs "out"))
+                          (string-append "LIBBAM="
+                                         (assoc-ref %build-inputs "samtools")
+                                         "/lib/libbam.a"))))
+    (inputs
+     `(("gsl" ,gsl)
+       ("samtools" ,samtools-0.1)
+       ("zlib" ,zlib)))
+    (home-page "http://smithlabresearch.org/software/preseq/";)
+    (synopsis "Program for analyzing library complexity")
+    (description
+     "The preseq package is aimed at predicting and estimating the complexity
+of a genomic sequencing library, equivalent to predicting and estimating the
+number of redundant reads from a given sequencing depth and how many will be
+expected from additional sequencing using an initial sequencing experiment.
+The estimates can then be used to examine the utility of further sequencing,
+optimize the sequencing depth, or to screen multiple libraries to avoid low
+complexity samples.")
+    (license license:gpl3+)))
+
 (define-public sra-tools
   (package
     (name "sra-tools")
diff --git a/gnu/packages/patches/preseq-1.0.2-install-to-PREFIX.patch 
b/gnu/packages/patches/preseq-1.0.2-install-to-PREFIX.patch
new file mode 100644
index 0000000..f8318ae
--- /dev/null
+++ b/gnu/packages/patches/preseq-1.0.2-install-to-PREFIX.patch
@@ -0,0 +1,37 @@
+This patch has been submitted upstream: 
https://github.com/smithlabcode/preseq/pull/15
+
+From 65387b0d766e6c06a15cf8d8f9183d9a4ba644bb Mon Sep 17 00:00:00 2001
+From: Ricardo Wurmus <address@hidden>
+Date: Thu, 25 Jun 2015 15:01:53 +0200
+Subject: [PATCH] Install to PREFIX
+
+---
+ Makefile | 8 ++++++--
+ 1 file changed, 6 insertions(+), 2 deletions(-)
+
+diff --git a/Makefile b/Makefile
+index 4d0367a..e334f2f 100644
+--- a/preseq-master/Makefile
++++ b/preseq-master/Makefile
+@@ -22,6 +22,10 @@ ifndef ROOT
+ ROOT = $(shell pwd)
+ endif
+ 
++ifndef PREFIX
++PREFIX = $(ROOT)
++endif
++
+ ifndef SMITHLAB_CPP
+ SMITHLAB_CPP=$(ROOT)/smithlab_cpp/
+ endif
+@@ -91,8 +95,8 @@ endif
+       $(CXX) $(CXXFLAGS) -o $@ $^ $(INCLUDEARGS) $(LIBS)
+ 
+ install: $(PROGS)
+-      @mkdir -p $(ROOT)/bin
+-      @install -m 755 $(PROGS) $(ROOT)/bin
++      @mkdir -p $(PREFIX)/bin
++      @install -m 755 $(PROGS) $(PREFIX)/bin
+ 
+ clean:
+       @-rm -f $(PROGS) *.o *~
diff --git a/gnu/packages/patches/preseq-1.0.2-link-with-libbam.patch 
b/gnu/packages/patches/preseq-1.0.2-link-with-libbam.patch
new file mode 100644
index 0000000..035a3b1
--- /dev/null
+++ b/gnu/packages/patches/preseq-1.0.2-link-with-libbam.patch
@@ -0,0 +1,33 @@
+This patch has been submitted upstream: 
https://github.com/smithlabcode/preseq/pull/15
+
+From a937a54d56e8a6ff06237471e568154d1fa5f267 Mon Sep 17 00:00:00 2001
+From: Ricardo Wurmus <address@hidden>
+Date: Thu, 25 Jun 2015 15:02:35 +0200
+Subject: [PATCH] Link with libbam.a if provided via LIBBAM
+
+---
+ Makefile | 7 +++++++
+ 1 file changed, 7 insertions(+)
+
+diff --git a/Makefile b/Makefile
+index e334f2f..de310e4 100644
+--- a/preseq-master/Makefile
++++ b/preseq-master/Makefile
+@@ -83,10 +83,17 @@ $(PROGS): $(addprefix $(SMITHLAB_CPP)/, \
+ preseq: continued_fraction.o load_data_for_complexity.o moment_sequence.o
+ 
+ ifdef SAMTOOLS_DIR
++ifdef LIBBAM
++LIBS += -pthread
++bam2mr preseq: $(addprefix $(SMITHLAB_CPP)/, SAM.o) \
++        $(LIBBAM)
++else
+ bam2mr preseq: $(addprefix $(SMITHLAB_CPP)/, SAM.o) \
+         $(addprefix $(SAMTOOLS_DIR)/, sam.o bam.o bam_import.o bam_pileup.o \
+         faidx.o bam_aux.o kstring.o knetfile.o sam_header.o razf.o bgzf.o)
+ endif
++endif # SAMTOOLS_DIR
++
+ 
+ %.o: %.cpp %.hpp
+       $(CXX) $(CXXFLAGS) -c -o $@ $< $(INCLUDEARGS)
-- 
2.1.0