Mercurial > repos > peterjc > mira4_assembler
changeset 6:626d5cfd01aa draft
Uploaded v0.0.1 preview 6, support for fragment length (using mira4_validator.py)
| author | peterjc |
|---|---|
| date | Mon, 21 Oct 2013 12:01:47 -0400 |
| parents | ffefb87bd414 |
| children | 902f01c1084b |
| files | tools/mira4/README.rst tools/mira4/mira4_de_novo.xml tools/mira4/mira4_mapping.xml tools/mira4/mira4_validator.py |
| diffstat | 4 files changed, 189 insertions(+), 43 deletions(-) [+] |
line wrap: on
line diff
--- a/tools/mira4/README.rst Tue Oct 15 12:07:34 2013 -0400 +++ b/tools/mira4/README.rst Mon Oct 21 12:01:47 2013 -0400 @@ -40,9 +40,10 @@ * http://toolshed.g2.bx.psu.edu/view/peterjc/mira_datatypes -There are just three Galaxy files to install: +There are four Galaxy files to install: -* mira4.py (the Python script) +* mira4.py (the Python wrapper script) +* mira4_validator.py (the Python parameter validation script) * mira4_de_novo.xml (the Galaxy tool definition for de novo usage) * mira4_mapping.xml (the Galaxy tool definition for mapping usage) @@ -81,7 +82,7 @@ For making the "Galaxy Tool Shed" http://toolshed.g2.bx.psu.edu/ tarball use the following command from the Galaxy root folder:: - $ tar -czf mira4_wrapper.tar.gz tools/mira4/README.rst tools/mira4/mira4_de_novo.xml tools/mira4/mira4_mapping.xml tools/mira4/mira4.py tools/mira4/tool_dependencies.xml test-data/tvc_mini.fastq test-data/tvc_contigs.fasta test-data/tvc_map_ref_strain.fasta test-data/tvc_map_same_strain.fasta + $ tar -czf mira4_wrapper.tar.gz tools/mira4/README.rst tools/mira4/mira4_de_novo.xml tools/mira4/mira4_mapping.xml tools/mira4/mira4.py tools/mira4/mira4_validator.py tools/mira4/tool_dependencies.xml test-data/tvc_mini.fastq test-data/tvc_contigs.fasta test-data/tvc_map_ref_strain.fasta test-data/tvc_map_same_strain.fasta Check this worked:: @@ -90,6 +91,7 @@ tools/mira4/mira4_de_novo.xml tools/mira4/mira4_mapping.xml tools/mira4/mira4.py + tools/mira4/mira4_validator.py tools/mira4/tool_dependencies.xml test-data/tvc_mini.fastq test-data/tvc_contigs.fasta
--- a/tools/mira4/mira4_de_novo.xml Tue Oct 15 12:07:34 2013 -0400 +++ b/tools/mira4/mira4_de_novo.xml Mon Oct 21 12:01:47 2013 -0400 @@ -1,7 +1,6 @@ <tool id="mira_4_0_de_novo" name="MIRA v4.0 de novo assember" version="0.0.1"> <description>Takes Sanger, Roche 454, Solexa/Illumina, Ion Torrent and PacBio reads</description> <requirements> - <requirement type="python-module">Bio</requirement> <requirement type="binary">mira</requirement> <requirement type="package" version="4.0">MIRA</requirement> </requirements> @@ -29,17 +28,39 @@ <option value="text">Synthetic reads (database entries, consensus sequences, artifical reads, etc)</option> <!-- TODO reference/backbone as an entry here? --> </param> - <param name="segment_placement" type="select" label="Pairing type (segment placing)"> - <option value="">None (e.g. single end sequencing)</option> - <option value="FR">---> <--- (e.g. Sanger capillary or Solexa/Illumina paired-end library)</option> - <option value="RF"><--- ---> (e.g. Solexa/Illumina mate-pair library)</option> - <option value="SB">2---> 1---> (e.g. Roche 454 paired-end libraries or IonTorrent long-mate; see note)</option> - <option value="?">Unknown or not relevant (e.g. primer walking with Sanger capillary sequencing)</option> - </param> + <conditional name="segments"> + <param name="type" type="select" label="Are these paired reads?"> + <option value="paired">Paired reads</option> + <option value="none">Single reads or not relevant (e.g. primer walking with Sanger capillary sequencing)</option> + </param> + <when value="paired"> + <param name="placement" type="select" label="Pairing type (segment placing)"> + <option value="FR">---> <--- (e.g. Sanger capillary or Solexa/Illumina paired-end library)</option> + <option value="RF"><--- ---> (e.g. Solexa/Illumina mate-pair library)</option> + <option value="SB">2---> 1---> (e.g. Roche 454 paired-end libraries or IonTorrent long-mate; see note)</option> + </param> + <!-- min/max validation is done via the <code> tag --> + <param name="min_size" type="integer" optional="true" min="0" value="" + label="Minimum size of 'good' DNA templates in the library preparation" + help="Optional, but if used you must also supply a maximum value." /> + <param name="max_size" type="integer" optional="true" min="0" value="" + label="Maximum size of 'good' DNA templates in the library preparation" + help="Optional, but if used you must also supply a minimum value." /> + <param name="naming" type="select" label="Pair naming convention"> + <option value="solexa">Solexa/Illumina (using '/1' and '/2' suffixes)</option> + <option value="FR">Forward/Reverse scheme (using '.f*' and '.r*' suffixes)</option> + <option value="tigr">TIGR scheme (using 'TF*' and 'TR*' suffixes)</option> + <option value="sanger">Sanger scheme (see notes)</option> + <option value="stlouis">St. Louis scheme (see notes)</option> + </param> + </when> + <when value="none" /><!-- no further questions --> + </conditional> <param name="filenames" type="data" format="fastq,mira" multiple="true" required="true" label="Read file(s)" help="Multiple files allowed, for example paired reads can be given as two files (MIRA looks at read names to identify pairs)." /> </repeat> </inputs> + <code file="mira4_validator.py" /> <outputs> <data name="out_fasta" format="fasta" label="MIRA de novo contigs (FASTA)" /> <data name="out_maf" format="mira" label="MIRA de novo assembly" /> @@ -67,12 +88,20 @@ readgroup technology = ${rg.technology} +##Record the segment placement (if any) +#if str($rg.segments.type) == "paired" +segmentplacement = ${rg.segments.placement} +segmentnaming = ${rg.segments.naming} +#if str($rg.segments.min_size) != "" or str($rg.segments.max_size) != "" +##If our min/max validation failed I trust MIRA to give an error message... +templatesize = $rg.segments.min_size $rg.segments.max_size +#end if +#end if +#if str($rg.segments.type) == "none" +segmentplacement = ? +#end if ##MIRA will accept multiple filenames on one data line, or multiple data lines #for $f in $rg.filenames -#if str($rg.segment_placement) != "" -##Record the segment placement (if any) -segmentplacement = ${rg.segment_placement} -#end if ##Must now map Galaxy datatypes to MIRA file types... #if $f.ext.startswith("fastq") ##MIRA doesn't like fastqsanger etc, just plain old fastq: @@ -120,18 +149,35 @@ It is particularly suited to small genomes such as bacteria. -**Notes** + +**Notes on paired reads** .. class:: warningmark -Note that the raw data for Roche 454 and Ion Torrent paired-end libraries -sequences a circularised fragment such that the raw data starts with the -end of the fragment, a linker, then the start of the fragment. This means -both the start and end are sequenced from the same strand, and thus should -be given to MIRA as orientation "2---> 1--->". However, in order to -use this data with traditional tools expecting Sanger capillary style -libraries which expect "---> <---" your FASTQ files may have been -pre-processed to mimic this by reverse complementing one of the pair. +MIRA uses read naming conventions to identify paired read partners +(and does not care about their order in the input files). In most cases, +the Solexa/Illumina setting is fine. For Sanger capillary sequencing, +you may need to rename your reads to match one of the standard conventions +supported by MIRA. For Roche 454 or Ion Torrent the appropriate settings +depend on how the FASTQ file was produced: + +* If using Roche's ``sffinfo`` or older versions of ``sff_extract`` + to convert SFF files to FASTQ, your reads will probably have the + ``---> <---`` orientation and use the ``.f`` and ``.r`` + suffixes (FR naming). + +* If using a recent version of ``sff_extract``, then the ``/1`` and ``/2`` + suffixes are used (Solexa/Illumina style naming) and the original + ``2---> 1--->`` orientation is preserved. + +The reason for this is the raw data for Roche 454 and Ion Torrent paired-end +libraries sequences a circularised fragment such that the raw data begins +with the end of the fragment, a linker, then the start of the fragment. +This means both the start and end are sequenced from the same strand, and +have the orientation ``2---> 1--->``. However, in order to use the data +with traditional tools expecting Sanger capillary style ``---> <---`` +orientation it was common to reverse complement one of the pair to mimic this. + **Citation**
--- a/tools/mira4/mira4_mapping.xml Tue Oct 15 12:07:34 2013 -0400 +++ b/tools/mira4/mira4_mapping.xml Mon Oct 21 12:01:47 2013 -0400 @@ -1,7 +1,6 @@ <tool id="mira_4_0_mapping" name="MIRA v4.0 mapping" version="0.0.1"> <description>Maps Sanger, Roche 454, Solexa/Illumina, Ion Torrent and PacBio reads</description> <requirements> - <requirement type="python-module">Bio</requirement> <requirement type="binary">mira</requirement> <requirement type="package" version="4.0">MIRA</requirement> </requirements> @@ -38,13 +37,27 @@ <option value="pcbiohq">PacBio high quality (corrected)</option> <option value="text">Synthetic reads (database entries, consensus sequences, artifical reads, etc)</option> </param> - <param name="segment_placement" type="select" label="Pairing type (segment placing)"> - <option value="">None (e.g. single end sequencing)</option> - <option value="FR">---> <--- (e.g. Sanger capillary or Solexa/Illumina paired-end library)</option> - <option value="RF"><--- ---> (e.g. Solexa/Illumina mate-pair library)</option> - <option value="SB">2---> 1---> (e.g. Roche 454 paired-end libraries or IonTorrent long-mate; see note)</option> - <option value="?">Unknown or not relevant (e.g. primer walking with Sanger capillary sequencing)</option> - </param> + <conditional name="segments"> + <param name="type" type="select" label="Are these paired reads?"> + <option value="paired">Paired reads</option> + <option value="none">Single reads or not relevant (e.g. primer walking with Sanger capillary sequencing)</option> + </param> + <when value="paired"> + <param name="placement" type="select" label="Pairing type (segment placing)"> + <option value="FR">---> <--- (e.g. Sanger capillary or Solexa/Illumina paired-end library)</option> + <option value="RF"><--- ---> (e.g. Solexa/Illumina mate-pair library)</option> + <option value="SB">2---> 1---> (e.g. Roche 454 paired-end libraries or IonTorrent long-mate; see note)</option> + </param> + <param name="naming" type="select" label="Pair naming convention"> + <option value="solexa">Solexa/Illumina (using '/1' and '/2' suffixes)</option> + <option value="FR">Forward/Reverse scheme (using '.f*' and '.r*' suffixes)</option> + <option value="tigr">TIGR scheme (using 'TF*' and 'TR*' suffixes)</option> + <option value="sanger">Sanger scheme (see notes)</option> + <option value="stlouis">St. Louis scheme (see notes)</option> + </param> + </when> + <when value="none" /><!-- no further questions --> + </conditional> <param name="filenames" type="data" format="fastq,mira" multiple="true" required="true" label="Read file(s)" help="Multiple files allowed, for example paired reads can be given as two files (MIRA looks at read names to identify pairs)." /> </repeat> @@ -104,9 +117,13 @@ ##This is perhaps redundant as MIRA defaults to StrainX for the reads: strain = StrainX #end if -#if str($rg.segment_placement) != "" ##Record the segment placement (if any) -segmentplacement = ${rg.segment_placement} +#if str($rg.segments.type) == "paired" +segmentplacement = ${rg.segments.placement} +segmentnaming = ${rg.segments.naming} +#end if +#if str($rg.segments.type) == "none" +segmentplacement = ? #end if ##MIRA will accept multiple filenames on one data line, or multiple data lines #for $f in $rg.filenames @@ -160,18 +177,35 @@ It is particularly suited to small genomes such as bacteria. -**Notes** + +**Notes on paired reads** .. class:: warningmark -Note that the raw data for Roche 454 and Ion Torrent paired-end libraries -sequences a circularised fragment such that the raw data starts with the -end of the fragment, a linker, then the start of the fragment. This means -both the start and end are sequenced from the same strand, and thus should -be given to MIRA as orientation "2---> 1--->". However, in order to -use this data with traditional tools expecting Sanger capillary style -libraries which expect "---> <---" your FASTQ files may have been -pre-processed to mimic this by reverse complementing one of the pair. +MIRA uses read naming conventions to identify paired read partners +(and does not care about their order in the input files). In most cases, +the Solexa/Illumina setting is fine. For Sanger capillary sequencing, +you may need to rename your reads to match one of the standard conventions +supported by MIRA. For Roche 454 or Ion Torrent the appropriate settings +depend on how the FASTQ file was produced: + +* If using Roche's ``sffinfo`` or older versions of ``sff_extract`` + to convert SFF files to FASTQ, your reads will probably have the + ``---> <---`` orientation and use the ``.f`` and ``.r`` + suffixes (FR naming). + +* If using a recent version of ``sff_extract``, then the ``/1`` and ``/2`` + suffixes are used (Solexa/Illumina style naming) and the original + ``2---> 1--->`` orientation is preserved. + +The reason for this is the raw data for Roche 454 and Ion Torrent paired-end +libraries sequences a circularised fragment such that the raw data begins +with the end of the fragment, a linker, then the start of the fragment. +This means both the start and end are sequenced from the same strand, and +have the orientation ``2---> 1--->``. However, in order to use the data +with traditional tools expecting Sanger capillary style ``---> <---`` +orientation it was common to reverse complement one of the pair to mimic this. + **Citation**
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tools/mira4/mira4_validator.py Mon Oct 21 12:01:47 2013 -0400 @@ -0,0 +1,64 @@ +#Called from the Galaxy Tool XML file +#import sys + +def validate_input(trans, error_map, param_values, page_param_map): + """Validates the min_size/max_size user input, before execution.""" + err_list = [] + for read_group in param_values["read_group"]: + err = dict() + segments = read_group["segments"] + if str(segments["type"]) != "paired": + err_list.append(dict()) + continue + + min_size = str(segments["min_size"]).strip() + max_size = str(segments["max_size"]).strip() + #sys.stderr.write("DEBUG min_size=%r, max_size=%r\n" % (min_size, max_size)) + + #Somehow Galaxy seems to turn an empty field into string "None"... + if min_size=="None": + min_size = "" + if max_size=="None": + max_size = "" + + if min_size=="" and max_size=="": + #Both missing is good + pass + elif min_size=="": + err["min_size"] = "Minimum size required if maximum size given" + elif max_size=="": + err["max_size"] = "Maximum size required if minimum size given" + + if min_size: + try: + min_size_int = int(min_size) + if min_size_int < 0: + err["min_size"] = "Minumum size must not be negative (%i)" % min_size_int + min_size = None # Avoid doing comparison below + except ValueError: + err["min_size"] = "Minimum size is not an integer (%s)" % min_size + min_size = None # Avoid doing comparison below + + if max_size: + try: + max_size_int = int(max_size) + if max_size_int< 0: + err["max_size"] = "Maximum size must not be negative (%i)" % max_size_int + max_size = None # Avoid doing comparison below + except ValueError: + err["max_size"] = "Maximum size is not an integer (%s)" % max_size + max_size = None # Avoid doing comparison below + + if min_size and max_size and min_size_int > max_size_int: + msg = "Minimum size must be less than maximum size (%i vs %i)" % (min_size_int, max_size_int) + err["min_size"] = msg + err["max_size"] = msg + + if err: + err_list.append({"segments":err}) + else: + err_list.append(dict()) + + if any(err_list): + #Return an error map only if any readgroup gave errors + error_map["read_group"] = err_list