[PATCH 4/4] gnu: Add prokka.

[Ben Woodcroft]

* gnu/packages/bioinformatics.scm (prokka): New variable.
---
 gnu/packages/bioinformatics.scm | 110 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 110 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index 4541791..c4f421c 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -62,6 +62,7 @@
   #:use-module (gnu packages image)
   #:use-module (gnu packages imagemagick)
   #:use-module (gnu packages java)
+  #:use-module (gnu packages less)
   #:use-module (gnu packages linux)
   #:use-module (gnu packages logging)
   #:use-module (gnu packages machine-learning)
@@ -3743,6 +3744,115 @@ for sequences to be aligned and then, simultaneously 
with the alignment,
 predicts the locations of structural units in the sequences.")
     (license license:gpl2+)))
 
+(define-public prokka
+  ;; There has been many commits since the last released version 1.11 so we
+  ;; package from git.
+  (let ((commit "1caf2394850998f89a3782cc8846dc51978faac2"))
+    (package
+      (name "prokka")
+      (version (string-append "1.11-1." (string-take commit 8)))
+      (source
+       (origin
+         (method git-fetch)
+         (uri (git-reference
+               (url "https://github.com/tseemann/prokka.git";)
+               (commit commit)))
+         (file-name (string-append name "-" version "-checkout"))
+         (sha256
+          (base32
+           "1h8dpmrcfw5z3w9sydgv4439v1wdidyq4nx247fyqahvn88rfsnj"))
+         (modules '((guix build utils)))
+         ;; Remove bundled code.
+         (snippet '(begin
+                     (delete-file-recursively "binaries")
+                     (delete-file-recursively "perl5")
+                     #t))))
+      (build-system gnu-build-system)
+      (arguments
+       `(#:phases
+         (modify-phases %standard-phases
+           (delete 'configure)
+           (replace 'build
+             (lambda _
+               (zero? (system* "bin/prokka" "--setupdb"))))
+           (replace 'install
+             (lambda* (#:key outputs #:allow-other-keys)
+               ;; Set the path to the 'db' to refer to the share directory.
+               (substitute* "bin/prokka"
+                 (("^my \\$DBDIR = .*")
+                  (string-append "my $DBDIR = '"
+                                 (assoc-ref outputs "out")
+                                 "/share/prokka/db';\n")))
+               (let* ((out (assoc-ref outputs "out"))
+                      (bin (string-append out "/bin"))
+                      (prokka (string-append bin "/prokka"))
+                      (share (string-append out "/share/prokka"))
+                      (path (getenv "PATH"))
+                      (perl5lib (getenv "PERL5LIB")))
+                 (mkdir-p share)
+                 (copy-recursively "db" (string-append share "/db"))
+                 (install-file "bin/prokka" bin)
+                 (install-file "bin/prokka-genbank_to_fasta_db" bin)
+                 (wrap-program prokka
+                   `("PATH" ":" prefix
+                     (,(string-append path ":" out "/bin"))))
+                 (for-each
+                  (lambda (binary)
+                    (wrap-program binary
+                      `("PERL5LIB" ":" prefix
+                        (,(string-append perl5lib ":" out
+                                         "/lib/perl5/site_perl")))))
+                  (list prokka (string-append
+                                bin "/prokka-genbank_to_fasta_db"))))
+               #t))
+           (delete 'check)
+           (add-after 'install 'post-install-check
+             (lambda* (#:key inputs #:allow-other-keys)
+               ;; There are no tests, instead we run a sanity check on an
+               ;; entire genome.
+               (zero? (system* "bin/prokka"
+                               "--noanno"
+                               "--outdir" "example-out"
+                               (assoc-ref inputs "example-genome"))))))))
+      (native-inputs
+       `(("example-genome"
+          ,(origin
+             (method url-fetch)
+             (uri "http://www.ebi.ac.uk/ena/data/view/CP002565&display=fasta";)
+             (file-name (string-append "ena-genome-CP002565.fasta"))
+             (sha256
+              (base32
+               "0dv3m29kgyssjc96zbmb5khkrk7cy7a66bsjk2ricwc302g5hgfy"))))))
+      (inputs
+       `(("perl" ,perl)
+         ("bioperl" ,bioperl-minimal)
+         ("blast+" ,blast+)
+         ("hmmer" ,hmmer)
+         ("aragorn" ,aragorn)
+         ("prodigal" ,prodigal)
+         ("parallel" ,parallel)
+         ("infernal" ,infernal)
+         ("barrnap" ,barrnap)
+         ("minced" ,minced)
+         ("tbl2asn" ,ncbi-tools)
+         ("grep" ,grep)
+         ("sed" ,sed)
+         ("less" ,less)
+         ("perl-time-piece" ,perl-time-piece)
+         ("perl-xml-simple" ,perl-xml-simple)
+         ("perl-digest-md5" ,perl-digest-md5)))
+      (home-page "http://www.vicbioinformatics.com/software.prokka.shtml";)
+      (synopsis "Rapid prokaryotic genome annotation")
+      (description
+       "Prokka is a software tool for the rapid annotation of prokaryotic
+genomes.  It produces GFF3, GBK and SQN files that are ready for editing in
+Sequin and ultimately submitted to Genbank/DDJB/ENA. ")
+      (license (list license:gpl2
+                     ;; Available under various licenses.
+                     (license:non-copyleft
+                      "file://doc"
+                      "See license files in the doc directory."))))))
+
 (define-public proteinortho
   (package
     (name "proteinortho")
-- 
2.10.2