[med-svn] [reapr] 01/02: add manpages

Sascha Steinbiss sascha at steinbiss.name
Fri Feb 5 20:48:34 UTC 2016


This is an automated email from the git hooks/post-receive script.

sascha-guest pushed a commit to branch master
in repository reapr.

commit 25e622236f56ae7bfb632e01a8a4fcb0426dcbd1
Author: Sascha Steinbiss <sascha at steinbiss.name>
Date:   Fri Feb 5 20:44:25 2016 +0000

    add manpages
---
 debian/control                             |  3 +-
 debian/man_src/reapr-break.1.ronn          | 43 +++++++++++++++++++++
 debian/man_src/reapr-facheck.1.ronn        | 21 ++++++++++
 debian/man_src/reapr-gapresize.1.ronn      | 15 ++++++++
 debian/man_src/reapr-perfectfrombam.1.ronn | 34 +++++++++++++++++
 debian/man_src/reapr-perfectmap.1.ronn     | 18 +++++++++
 debian/man_src/reapr-pipeline.1.ronn       | 38 +++++++++++++++++++
 debian/man_src/reapr-plots.1.ronn          | 15 ++++++++
 debian/man_src/reapr-preprocess.1.ronn     | 10 +++++
 debian/man_src/reapr-score.1.ronn          | 52 +++++++++++++++++++++++++
 debian/man_src/reapr-seqrename.1.ronn      | 14 +++++++
 debian/man_src/reapr-smaltmap.1.ronn       | 52 +++++++++++++++++++++++++
 debian/man_src/reapr-stats.1.ronn          | 39 +++++++++++++++++++
 debian/man_src/reapr-summary.1.ronn        | 20 ++++++++++
 debian/man_src/reapr.1.ronn                | 61 ++++++++++++++++++++++++++++++
 debian/manpages                            |  1 +
 debian/rules                               | 18 +++++++--
 17 files changed, 450 insertions(+), 4 deletions(-)

diff --git a/debian/control b/debian/control
index 1f34b2a..c7010b8 100644
--- a/debian/control
+++ b/debian/control
@@ -7,7 +7,8 @@ Build-Depends: debhelper (>= 9),
                libbamtools-dev,
                libtabixpp-dev,
                libhts-dev (>= 1.3),
-               zlib1g-dev
+               zlib1g-dev,
+               ruby-ronn
 Standards-Version: 3.9.6
 Vcs-Browser: https://anonscm.debian.org/cgit/debian-med/reapr.git
 Vcs-Git: https://anonscm.debian.org/git/debian-med/reapr.git
diff --git a/debian/man_src/reapr-break.1.ronn b/debian/man_src/reapr-break.1.ronn
new file mode 100644
index 0000000..b40baf8
--- /dev/null
+++ b/debian/man_src/reapr-break.1.ronn
@@ -0,0 +1,43 @@
+reapr-break(1) -- make broken assembly via REAPR
+================================================
+
+## SYNOPSIS
+
+`reapr break` [options] <assembly.fa> <errors.gff.gz> <outfileprefix>
+
+## DESCRIPTION
+
+<errors.gff.gz> is the errors GFF file made when running `score`.
+
+## OPTIONS
+
+* `-a`:
+  Agressive breaking: break contigs at any FCD or low_frag error, as
+  opposed to the default of replacing with Ns. Contigs are broken at the
+  midpoint of each error. Also see option `-t`. Incompatible with `-b`.
+
+* `-b`:
+  Ignore FCD and low fragment coverage errors that do not contain
+  a gap (the default is to replace these with Ns). Incompatible with `-a`.
+
+* `-e` <float>:
+  Minimum FCD error [0]
+
+* `-l` <int>:
+  Minimum sequence length to output [100]
+
+* `-m` <int>:
+  Max sequence length to write to the bin. Sequences longer
+  than this are written to the main assembly output. This is to stop
+  long stretches of sequence being lost [999]
+
+* `-t` <int>:
+  When `-a` is used, use this option to specify how many bases
+  are trimmed off the end of each new contig around a break.
+  `-t` N means that, at an FCD error, a contig is broken at the middle
+  coordinate of the error, then N bases are
+  trimmed off each new contig end [0]
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-facheck.1.ronn b/debian/man_src/reapr-facheck.1.ronn
new file mode 100644
index 0000000..3d8fb11
--- /dev/null
+++ b/debian/man_src/reapr-facheck.1.ronn
@@ -0,0 +1,21 @@
+reapr-facheck(1) -- checks IDs in fasta file
+============================================
+
+## SYNOPSIS
+
+`reapr facheck` <in.fa> [out_prefix]
+
+## DESCRIPTION
+
+Checks that the names in the fasta file are ok.  Things like
+trailing whitespace or characters |':- could break the pipeline.
+
+If out_prefix is not given, it will die when a bad ID is found.
+No output means everything is OK.
+
+If out_prefix is given, writes a new fasta file out_prefix.fa with new
+names, and out_prefix.info which has the mapping from old name to new name.
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-gapresize.1.ronn b/debian/man_src/reapr-gapresize.1.ronn
new file mode 100644
index 0000000..cdef108
--- /dev/null
+++ b/debian/man_src/reapr-gapresize.1.ronn
@@ -0,0 +1,15 @@
+reapr-gapresize(1) -- calculates gap sizes based on read mapping
+================================================================
+
+## SYNOPSIS
+
+`reapr gapresize` [options] <assembly.fasta> <in.bam> <avg-fragment-length> <max-fragment-length> <outfileprefix>
+
+## OPTIONS
+
+* `-g` <int>:
+  Only consider gaps of at least this length [1]
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-perfectfrombam.1.ronn b/debian/man_src/reapr-perfectfrombam.1.ronn
new file mode 100644
index 0000000..b2413c5
--- /dev/null
+++ b/debian/man_src/reapr-perfectfrombam.1.ronn
@@ -0,0 +1,34 @@
+reapr-perfectfrombam(1) -- generate perfect mapping plots from a bam file
+=========================================================================
+
+## SYNOPSIS
+
+`reapr perfectfrombam` [options] <in.bam> <outfileprefix> <mininsert> <maxinsert> <repetitive-max-qual> <perfect-min-qual> <perfect-min-alignment-score>
+
+## DESCRIPTION
+
+Alternative to using `perfectmap`, for large genomes.
+
+Takes a BAM, which must have AS:... tags in each line. Makes file
+of perfect mapping depth, for use with the REAPR pipeline.  Recommended to
+use `perfectmap` instead, unless your genome is large (more than ~300MB),
+since although very fast to run, 'reapr perfectmap' uses a lot of memory.
+
+A BAM file made by `smaltmap` is suitable input.
+
+Reads in pair pointing towards each other, with the given minimum
+alignment score and mapping quality and within the given insert size range
+are used to generate the coverage across the genome.
+
+Additionally, regions with repetitive coverage are called, by taking read
+pairs where at least one read of the pair (is mapped and) has mapping
+quality less than or equal to <repetitive max qual>.
+
+## OPTIONS
+
+* `-noclean`:
+  Use this to not delete the temporary bam file
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-perfectmap.1.ronn b/debian/man_src/reapr-perfectmap.1.ronn
new file mode 100644
index 0000000..1f2d610
--- /dev/null
+++ b/debian/man_src/reapr-perfectmap.1.ronn
@@ -0,0 +1,18 @@
+reapr-perfectmap(1) -- make perfect uniquely mapping plot files
+===============================================================
+
+## SYNOPSIS
+
+`reapr perfectmap` <assembly.fa> <reads_1.fastq> <reads_2.fastq> <avginsertsize> <outprefix>
+
+## DESCRIPTION
+
+Note: the reads can be gzipped.  If the extension is '.gz', then they
+are assumed to be gzipped and dealt with accordingly (i.e. called something
+like <reads_1.fastq.gz> <reads_2.fastq.gz>).
+
+IMPORTANT: all reads must be the same length.
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-pipeline.1.ronn b/debian/man_src/reapr-pipeline.1.ronn
new file mode 100644
index 0000000..cbe5f4f
--- /dev/null
+++ b/debian/man_src/reapr-pipeline.1.ronn
@@ -0,0 +1,38 @@
+reapr-pipeline(1) -- run the REAPR pipeline
+===========================================
+
+## SYNOPSIS
+
+`reapr pipeline` [options] <assembly.fa> <in.bam> <outdirectory> [perfectmapprefix]
+
+## DESCRIPTION
+
+where [perfectmapprefix] is optional and should be the prefix used when task
+`perfectmap` was run.
+
+It is assumed that reads in <in.bam> are 'innies', i.e. the correct orientation
+is reads in a pair pointing towards each other (---> <---).
+
+## OPTIONS
+
+* `-stats|fcdrate|score|break option=value`:
+  You can pass options to `stats`, `fcdrate`, `score` or `break`
+  if you want to change the default settings. These
+  can be used multiple times to use more than one option. e.g.:
+    `-stats i=100 -stats j=1000`
+  If an option has no value, use `1`. e.g.
+    `-break b=1`
+
+* `-fcdcut` <float>:
+  Set the fcdcutoff used when running score. Default is to
+  run fcdrate to determine the cutoff. Using this option will
+  skip fcdrate and use the given value.
+
+* `-x`:
+  By default, a bash script is written to run all
+  the pipeline stages. Using this option stops the
+  script from being run.
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-plots.1.ronn b/debian/man_src/reapr-plots.1.ronn
new file mode 100644
index 0000000..9bc71e0
--- /dev/null
+++ b/debian/man_src/reapr-plots.1.ronn
@@ -0,0 +1,15 @@
+reapr-plots(1) -- makes Artemis plot files for a given contig
+=============================================================
+
+## SYNOPSIS
+
+`reapr plots` [options] <in.stats.gz> <outprefix> <assembly.fa> <contigid>
+
+## OPTIONS
+
+* `-s` <prefix>:
+  This should be the outfiles prefix used when `score` task was run
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-preprocess.1.ronn b/debian/man_src/reapr-preprocess.1.ronn
new file mode 100644
index 0000000..5dbc843
--- /dev/null
+++ b/debian/man_src/reapr-preprocess.1.ronn
@@ -0,0 +1,10 @@
+reapr-preprocess(1) -- preprocess files for REAPR run
+===============================================================
+
+## SYNOPSIS
+
+`reapr preprocess` <assembly.fa> <in.bam> <outputdir>
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-score.1.ronn b/debian/man_src/reapr-score.1.ronn
new file mode 100644
index 0000000..ac07fcf
--- /dev/null
+++ b/debian/man_src/reapr-score.1.ronn
@@ -0,0 +1,52 @@
+reapr-score(1) -- calculate scores and assembly errors
+======================================================
+
+## SYNOPSIS
+
+`reapr score` [options] <assembly.fa.gaps.gz> <in.bam> <statsprefix> <FCDcutoff> <outprefix>
+
+## OPTIONS
+
+* `-f` <int>:
+  Minimum inner fragment coverage [1]
+
+* `-g` <int>:
+  Max gap length to call over [0.5 * outer_mean_insert_size]
+
+* `-l` <int>:
+  Length of window [100]
+
+* `-p` <int>:
+  Use perfect mapping reads score with given min coverage.
+  Incompatible with -P.
+
+* `-P` <int>:
+  Same as -p, but force the score to be zero at any position with
+  at least the given coverage of perfect mapping reads and which has an
+  OK insert plot, , i.e. perfect mapping reads + insert distribution
+  override all other tests when calculating the score.
+  Incompatible with -p.
+
+* `-q` <float>:
+  Max bad read ratio [0.33]
+
+* `-r` <int>:
+  Min read coverage [max(1, mean_read_cov - 4 * read_cov_stddev)]
+
+* `-R` <int>:
+  Repeat calling cutoff. -R N means call a repeat if fragment
+  coverage is >= N * (expected coverage).
+  Use -R 0 to not call repeats [2]
+
+* `-s` <int>:
+  Min score to report in errors file [0.4]
+
+* `-u` <int>:
+  FCD error window length for error calling [insert_size / 2]
+
+* `-w` <float>:
+  Min percentage of bases in window needed to call as bad [0.8]
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-seqrename.1.ronn b/debian/man_src/reapr-seqrename.1.ronn
new file mode 100644
index 0000000..cb315d2
--- /dev/null
+++ b/debian/man_src/reapr-seqrename.1.ronn
@@ -0,0 +1,14 @@
+reapr-seqrename(1) -- renames all sequences in a BAM file
+=====================================================
+
+## SYNOPSIS
+
+`reapr seqrename` <renamefile> <in.bam> <out.bam>
+
+## DESCRIPTION
+
+<renamefile> is the file `*.info` made by `facheck`, which contains the mapping of old name to new name.
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-smaltmap.1.ronn b/debian/man_src/reapr-smaltmap.1.ronn
new file mode 100644
index 0000000..4f14772
--- /dev/null
+++ b/debian/man_src/reapr-smaltmap.1.ronn
@@ -0,0 +1,52 @@
+reapr-smaltmap(1) -- map read pairs using SMALT
+===============================================
+
+## SYNOPSIS
+
+`reapr smaltmap` [options] <assembly.fa> <reads_1.fastq> <reads_2.fastq> <out.bam>
+
+## DESCRIPTION
+
+Maps read pairs to an asembly with SMALT, making a final BAM file that
+is sorted, indexed and has duplicates removed.
+
+The n-th read in <reads_1.fastq> should be the mate of the n-th read in
+<reads_2.fastq>.
+
+It is assumed that reads are 'innies', i.e. the correct orientation
+is reads in a pair pointing towards each other (---> <---).
+
+## OPTIONS
+
+  * `-k` <int>:
+    The -k option (kmer hash length) when indexing the genome
+    with 'smalt index' [13]
+
+  * `-s` <int>:
+    The -s option (step length) when indexing the genome
+     with 'smalt index' [2]
+
+  * `-m` <int>:
+   The -m option when mapping reads with 'smalt map' [not used by default]
+
+  * `-n` <int>:
+    The number of threads used when running 'smalt map' [1]
+
+  * `-y` <float>:
+    The -y option when mapping reads with 'smalt map'.
+  The default of 0.5 means that at least 50% of each read must map
+  perfectly. Depending on the quality of your reads, you may want to
+  increase this to be more stringent (or consider using -m) [0.5]
+
+  * `-x`:
+    Use this to just print the commands that will be run, instead of
+  actually running them
+
+  * `-u` <int>:
+      The -u option of 'smalt sample'. This is used to estimate the insert
+  size from a sample of the reads, by mapping every n^th read pair [1000]
+
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-stats.1.ronn b/debian/man_src/reapr-stats.1.ronn
new file mode 100644
index 0000000..e98b665
--- /dev/null
+++ b/debian/man_src/reapr-stats.1.ronn
@@ -0,0 +1,39 @@
+reapr-stats(1) -- generate REAPR stats from a BAM file
+======================================================
+
+## SYNOPSIS
+
+`reapr stats` [options] <preprocess-output-directory> <outprefix>
+
+## OPTIONS
+
+* `-f` <int>:
+  Insert size [ave from stats.txt]
+
+* `-i` <int>:
+  Minimum insert size [pc1 from stats.txt]
+
+* `-j` <int>:
+  Maximum insert size [pc99 from stats.txt]
+
+* `-m` <int>:
+  Maximum read length (this doesn't need to be exact, it just
+  determines memory allocation, so must be >= max read length) [2000]
+
+* `-p` <string>:
+  Name of .gz perfect mapping file made by `perfectmap`
+
+* `-q` <int>:
+  Ignore reads with mapping quality less than this [0]
+
+* `-s` <int>:
+  Calculate FCD error every n-th base
+  [ceil((fragment size) / 1000)]
+
+* `-u` <string>:
+  File containing list of chromosomes to look at
+  (one per line)
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr-summary.1.ronn b/debian/man_src/reapr-summary.1.ronn
new file mode 100644
index 0000000..b83c920
--- /dev/null
+++ b/debian/man_src/reapr-summary.1.ronn
@@ -0,0 +1,20 @@
+reapr-summary(1) -- make REAPR summary stats file
+=================================================
+
+## SYNOPSIS
+
+`reapr summary` [options] <assembly.fa> <scoreprefix> <breakprefix> <outfileprefix>
+
+## DESCRIPTION
+
+<scoreprefix> is the outfiles prefix used when `score` was run, and
+<breakprefix> is the outfiles prefix used when `break` was run.
+
+## OPTIONS
+
+* `-e` <float>:
+  Minimum FCD error [0]
+
+## SEE ALSO
+
+reapr(1)
\ No newline at end of file
diff --git a/debian/man_src/reapr.1.ronn b/debian/man_src/reapr.1.ronn
new file mode 100644
index 0000000..d8e13fe
--- /dev/null
+++ b/debian/man_src/reapr.1.ronn
@@ -0,0 +1,61 @@
+reapr(1) -- universal tool for genome assembly evaluation
+=========================================================
+
+## SYNOPSIS
+
+`reapr` <task> [options]
+
+## DESCRIPTION
+
+reapr is the main executable for the **REAPR** software. The actual
+functionality is provided by the subtools (`tasks'):
+
+### Common tasks
+
+* `facheck`:
+  checks IDs in fasta file
+
+* `smaltmap`:
+  map read pairs using SMALT: makes a BAM file to be used as input to the pipeline
+
+* `perfectmap`:
+  make perfect uniquely mapping plot files
+
+* `pipeline`:
+  runs the REAPR pipeline, using an assembly and mapped reads as input, and optionally results of perfectmap. (It runs facheck, preprocess, stats, fcdrate, score, summary and break)
+
+* `plots`:
+  makes Artemis plot files for a given contig, using results from stats (and optionally results from score)
+
+* `seqrename`:
+  renames all sequences in a BAM file: use this if you already mapped your reads but then found facheck failed - saves remapping the reads so that pipeline can be run
+
+### Advanced tasks
+
+* `preprocess`:
+  preprocess files: necessary for running stats
+
+* `stats`:
+  generates stats from a BAM file
+
+* `fcdrate`:
+  estimates FCD cutoff for score, using results from stats
+
+* `score`:
+  calculates scores and assembly errors, using results from stats
+
+* `summary`:
+  make summary stats file, using results from score
+
+* `break`:
+  makes broken assembly, using results from score
+
+* `gapresize`:
+  experimental, calculates gap sizes based on read mapping
+
+* `perfectfrombam`:
+  generate perfect mapping plots from a bam file (alternative to using perfectmap for large genomes)
+
+## SEE ALSO
+
+reapr-facheck(1), reapr-smaltmap(1), reapr-perfectmap(1) , reapr-pipeline(1), reapr-plots(1), reapr-seqrename(1), reapr-preprocess(1), reapr-stats(1), reapr-fcdrate(1), reapr-score(1), reapr-summary(1), reapr-break(1), reapr-gapresize(1), reapr-perfectfrombam(1)
\ No newline at end of file
diff --git a/debian/manpages b/debian/manpages
new file mode 100644
index 0000000..13cdaf4
--- /dev/null
+++ b/debian/manpages
@@ -0,0 +1 @@
+debian/man/*.1
diff --git a/debian/rules b/debian/rules
index 91c5142..cd67a89 100755
--- a/debian/rules
+++ b/debian/rules
@@ -1,10 +1,11 @@
 #!/usr/bin/make -f
 
-# DH_VERBOSE := 1
-
 %:
 	dh $@
 
+mandir := $(CURDIR)/debian/man
+debfolder := $(CURDIR)/debian
+
 override_dh_auto_build:
 	cd src && $(MAKE)
 
@@ -13,4 +14,15 @@ override_dh_auto_install:
 	cp src/reapr.pl debian/reapr/usr/lib/reapr/reapr.pl
 	chmod -x src/*.cpp
 	find src -executable -exec cp {} debian/reapr/usr/lib/reapr \;
-	dh_auto_install --
\ No newline at end of file
+	dh_auto_install --
+
+override_dh_installman:
+	mkdir -p $(mandir)
+	ronn $(debfolder)/man_src/*.ronn
+	cp $(debfolder)/man_src/*.? $(mandir)
+	dh_installman --
+
+override_dh_auto_clean:
+	rm -rf $(debfolder)/man_src/*.html $(debfolder)/man_src/*.?
+	rm -rf $(mandir)
+	dh_auto_clean --
\ No newline at end of file

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-med/reapr.git



More information about the debian-med-commit mailing list