Mercurial > repos > iuc > gemini_gene_wise
changeset 5:4b26f6c99227 draft
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/gemini commit 62ed732cba355e695181924a8ed4cce49ca21c59
line wrap: on
line diff
--- a/gemini_gene_wise.xml Fri Dec 14 12:55:02 2018 -0500 +++ b/gemini_gene_wise.xml Fri Jan 11 17:48:39 2019 -0500 @@ -1,5 +1,5 @@ -<tool id="gemini_@BINARY@" name="GEMINI @BINARY@" version="@VERSION@.1"> - <description>Custom genotype filtering by gene</description> +<tool id="gemini_@BINARY@" name="GEMINI @BINARY@" version="@VERSION@"> + <description>Discover per-gene variant patterns across families</description> <macros> <import>gemini_macros.xml</import> <token name="@BINARY@">gene_wise</token> @@ -11,34 +11,62 @@ <![CDATA[ gemini @BINARY@ - #if int($min_filters) > 0: - --min_filters $min_filters + --min-filters $min_filters #end if - #set $multiline_sql_expr = $gt_filter - #set $cmdln_param = "--gt-filter" - @MULTILN_SQL_EXPR_TO_CMDLN@ + #for $filter in $filter_by_genotype: + #set $multiline_sql_expr = str($filter.gt_filter) + #if $filter.is_required: + #set $cmdln_param = "--gt-filter-required" + #else: + #set $cmdln_param = "--gt-filter" + #end if + @MULTILN_SQL_EXPR_TO_CMDLN@ + #end for + #set $report = $oformat.report @COLUMN_SELECT@ - @CMDLN_SQL_FILTER_FILTER_OPTION@ + #set $where_clause_elements = [] + #set $filter_cmdln_param = '--filter' + #for $cond in $constraint: + #if str($cond.filter).strip(): + #silent $where_clause_elements.append(str($cond.filter).strip()) + #if $cond.overwrite_default_filter: + #set $filter_cmdln_param = '--where' + #end if + #end if + #end for - "${ infile }" - > "${ outfile }" + @PARSE_REGION_ELEMENTS@ + #if $region_elements: + #silent $where_clause_elements.append(" OR ".join($region_elements)) + #end if + #set $filter = " AND ".join($where_clause_elements) + #if str($filter): + $filter_cmdln_param '$filter' + #end if + + '$infile' + > '$outfile' ]]> </command> <inputs> <expand macro="infile" /> - - <param name="gt_filter" type="text" area="True" size="5x50" label="Restrictions to apply to genotype values" help="(--gt-filter)"> - <expand macro="sanitize_query" /> - </param> - <param name="min_filters" type="integer" value="0" min="0" label="Minimum number of filters" help="(--min-filters)" /> - <expand macro="column_filter" /> - <expand macro="filter" /> - - + <expand macro="gt_filter" default_repeat="1" min_repeat="1" max_repeat="999"> + <param name="is_required" type="boolean" checked="False" + label="Make this an obligate filter that a variant has to pass to be considered" + help="By default, a variant has to pass a minimum number of genotype filters (set below) to get reported. By making a filter required, you ensure that variants that fail this one filter are always excluded. Required filters that a variant passes do not count towards its number of passed (regular) filters" /> + </expand> + <param name="min_filters" type="integer" value="1" min="1" label="Minimum number of filters" help="(--min-filters)" /> + <expand macro="region_filter" /> + <expand macro="insert_constraint"> + <expand macro="overwritable_where_default" default_where="exonic, high impact variants (SQL clause: is_exonic = 1 and impact_severity != 'LOW')" /> + </expand> + <section name="oformat" title="Output - included information" expanded="true"> + <expand macro="column_filter" /> + </section> </inputs> <outputs> <data name="outfile" format="tabular" /> @@ -46,10 +74,12 @@ <tests> <test> <param name="infile" value="gemini_amend_input.db" ftype="gemini.sqlite" /> - <param name="gt_filter" value="((gt_depths).(*).(>=1).(all))" /> + <repeat name="filter_by_genotype"> + <param name="gt_filter" value="((gt_depths).(*).(>=1).(all))" /> + </repeat> <output name="outfile"> <assert_contents> - <has_line_matching expression="chrom	start	end	gene.*" /> + <has_line_matching expression="variant_id	gene.*" /> </assert_contents> </output> </test> @@ -58,7 +88,81 @@ <![CDATA[ **What it does** -The gemini query tool allows querying by variant and the inheritance tools described above enable querying by gene for fixed inheritance patterns. The gene_wise tool allows querying by gene with custom genotype filters to bridge the gap between these tools. +This tool extends the *GEMINI inheritance pattern* tool in that it lets you +search for custom gene-wise inheritance patterns of variants, instead of fixed +ones. + +See also: the `command line tool documentation +<https://gemini.readthedocs.io/en/latest/content/tools.html#gene-wise-custom-genotype-filtering-by-gene>`__ + +----- + +*Genotype filters* + +The syntax for specifying a genotype filter (``--gt-filter`` command line +option) is the same as for the *GEMINI query* tool and is described `here +<https://gemini.readthedocs.io/en/latest/content/querying.html#gt-filter-filtering-on-genotypes>`__. + +The difference with the *gene_wise* tool is that it lets you specify multiple +such filters and, if you do, every filter can be met by a **different variant** +as long as all of them are in the **same gene**. + +This is useful if your analysis includes several families that you suspect +(based on a shared phenotype) to have the same gene affected, but not +necessarily through the same variant. In this case, you can formulate one filter +per family like, for example:: + + gt_types.fam1_kid == HET and gt_types.fam1_mom == HOM_REF and gt_types.fam1_dad == HOM_REF + + gt_types.fam2_kid == HET + + gt_types.fam3_kid == HET + +, which would allow you to find a causal gene that's affected by different +(dominant) variants in children from three different families. Note that the +first filter combines three conditions applied to family 1, which, thus, must +be met by the same variant site. + +*Regular and required filters* (``--gt-filter`` *vs* ``--gt-filter-required``) +and the *Minimum number of filters* + +For every single genotype filter you define you can specify whether it should +be applied as a regular or as a required filter. The difference is that, if a +variant doesn't pass a required filter it is excluded from further analysis. +Of the regular filters, a gene and its variants only have to pass a threshold +number defined by *Minimum number of filters* (``--min-filters``). Imagine, +with the above filters you had specified ``--min-filters`` as ``2``, then a +gene for which the child in family 3 carries one copy of a variant allele and +the child in family 3 carries a copy of a different allele would be reported +no matter if any other allele in that gene passes the first filter, *etc.*. + +----- + +*Region filters* + +They let you restrict your analysis to parts of the genome, which can be useful +if you have prior knowledge of the approximate location of the causative gene. + +If you specify more then one region filter, they get combined with a logical +*OR*, meaning variants and genes falling in *any* of the regions are reported. + +----- + +*Additional constraints on variants* + +These get translated directly into the WHERE clause of an SQL query and, thus, +have to be expressed in valid SQL syntax. Of particular interest, here, is the +fact that, by default, the *gene-wise* tool applies the WHERE clause: +``is_exonic = 1 and impact_severity != 'LOW'``, which means the tool only +considers variants in exons that are not of *LOW* impact severity (*i.e.*, not +silent mutations). While this can be a good and biologically justifiable +setting, you can overwrite it if you need. + +Note that in SQL syntax tests for equality use a single ``=``, while genotype +filters (discussed above) are following Python syntax and use ``==`` for the +same purpose. Also note that non-numerical values need to be enclosed in +single-quotes, *e.g.* ``'LOW'``, but numerical values must *NOT* be. + ]]> </help> <expand macro="citations"/>
--- a/gemini_macros.xml Fri Dec 14 12:55:02 2018 -0500 +++ b/gemini_macros.xml Fri Jan 11 17:48:39 2019 -0500 @@ -1,15 +1,12 @@ <macros> <!-- gemini version to be used --> - <token name="@VERSION@">0.18.1</token> + <token name="@VERSION@">0.20.1</token> <!-- minimal annotation files version required by this version of gemini --> - <token name="@DB_VERSION@">181</token> + <token name="@DB_VERSION@">200</token> <xml name="requirements"> <requirements> <requirement type="package" version="@VERSION@">gemini</requirement> - <requirement type="package" version="0.2.6">tabix</requirement> - <!-- for conda useage --> - <!--requirement type="package" version="1.3.1">htslib</requirement--> <yield /> </requirements> </xml> @@ -24,9 +21,17 @@ <exit_code range=":-1" /> <regex match="Error:" /> <regex match="Exception:" /> + <yield /> </stdio> </xml> + <xml name="citations"> + <citations> + <citation type="doi">10.1371/journal.pcbi.1003153</citation> + <yield /> + </citations> + </xml> + <xml name="annotation_dir"> <param name="annotation_databases" type="select" label="Choose a gemini annotation source"> <options from_data_table="gemini_versioned_databases"> @@ -36,31 +41,36 @@ </param> </xml> - <xml name="add_header_column"> - <param name="header" type="boolean" truevalue="--header" falsevalue="" checked="False" - label="Add a header of column names to the output" help="(--header)"/> - </xml> - - <xml name="radius"> - <param name="radius" type="integer" value="3" label="Set filter for Breadth-first search (BFS) in the Protein-Protein Interaction network" help="(-r)" > - <validator type="in_range" min="0"/> + <xml name="infile"> + <param name="infile" type="data" format="gemini.sqlite" label="GEMINI database" help="Only files with version @VERSION@ are accepted." > + <options options_filter_attribute="metadata.gemini_version" > + <filter type="add_value" value="@VERSION@" /> + </options> </param> </xml> - <xml name="variant_mode"> - <param name="variant_mode" type="boolean" truevalue="--var" falsevalue="" checked="False" - label="Returns variant info (e.g. impact, biotype) for interacting genes" help="(--var)"/> + + <xml name="add_header_column"> + <param argument="--header" name="header" type="boolean" truevalue="--header" falsevalue="" checked="True" + label="Add a header of column names to the output" /> </xml> - <xml name="column_filter"> + <xml name="column_filter" token_help="" token_minimalset="variant_id, gene"> <conditional name="report"> - <param name="report_selector" type="select" label="Columns to include in the report" - help="By default, this tool reports all columns in the variants table. One may choose to report only a subset of the columns."> - <option value="all" selected="True">all</option> - <option value="column_filter">User given columns</option> + <param name="report_selector" type="select" + label="Set of columns to include in the variant report table" + help="@HELP@"> + <option value="minimal">Minimal (report only a preconfigured minimal set of columns)</option> + <option value="full">Full (report all columns defined in the GEMINI database variants table)</option> + <option value="custom">Custom (report user-specified columns)</option> </param> - <when value="all"/> - <when value="column_filter"> - <param name="columns" type="select" display="checkboxes" multiple="True" label="Choose columns to include in the report" help="(--columns)"> + <when value="full" /> + <when value="minimal"> + <param name="columns" type="hidden" value="@MINIMALSET@" /> + <param name="extra_cols" type="hidden" value="" /> + </when> + <when value="custom"> + <param name="columns" type="select" display="checkboxes" multiple="true" optional="true" + label="Choose columns to include in the report" help="(--columns)"> <option value="gene">gene</option> <option value="chrom">chrom</option> <option value="start">start</option> @@ -69,27 +79,23 @@ <option value="alt">alt</option> <option value="impact">impact</option> <option value="impact_severity">impact_severity</option> - <option value="max_aaf_all">alternative allele frequency</option> + <option value="max_aaf_all">alternative allele frequency (max_aaf_all)</option> </param> - <param name="extra_cols" type="text" label="Additional columns." help="Separate by whitespace"/> + <param name="extra_cols" type="text" + label="Additional columns (comma-separated)" + help="Column must be specified by the exact name they have in the GEMINI database, e.g., is_exonic or num_hom_alt, but, for genotype columns, GEMINI wildcard syntax is supported. The order of columns in the list is maintained in the output."> + <expand macro="sanitize_query" /> + </param> </when> </conditional> </xml> - <xml name="filter"> - <conditional name="filter"> - <param name="filter_selector" type="select" label="Apply additional constraints" - help="By default, this tool will report all variants regardless of their putative functional impact. In order to apply additional constraints on the variants returned, you can this optional filter."> - <option value="no">No additional constraints</option> - <option value="yes">Apply additional constraints</option> - </param> - <when value="no"/> - <when value="yes"> - <param name="filter" type="text" label="Contraints in SQL syntax" help="Conditions applied here will become WHERE clauses in the query issued to the GEMINI database. E.g. alt='G' or impact_severity = 'HIGH'. (--filter)"> - <expand macro="sanitize_query" /> - </param> - </when> - </conditional> + <xml name="filter" token_argument="--filter"> + <param argument="@ARGUMENT@" name="filter" type="text" + label="Additional constraints expressed in SQL syntax" + help="Constraints defined here will become the WHERE clause of the SQL query issued to the GEMINI database. E.g. alt='G' or impact_severity = 'HIGH'."> + <expand macro="sanitize_query" /> + </param> </xml> <xml name="sanitize_query"> @@ -103,10 +109,90 @@ </sanitizer> </xml> + <xml name="lenient" token_argument="--lenient" token_truevalue="--lenient" token_help="The exact consequence of this setting depends on the type of inheritance pattern you are looking for (see the tool help below)."> + <param argument="@ARGUMENT@" name="lenient" type="boolean" truevalue="@TRUEVALUE@" falsevalue="" checked="False" + label="Include hits with less convincing inheritance patterns" + help= "@HELP@" /> + </xml> + + <xml name="unaffected"> + <param argument="--allow-unaffected" name="allow_unaffected" type="boolean" truevalue="--allow-unaffected" falsevalue="" checked="False" + label="Report candidates shared by unaffected samples" + help="Activating this option will enable the reporting of variants as candidate causative even if they are shared by unaffected samples in the family tree. The default will only report variants that are unique to affected samples."/> + </xml> + + <xml name="min_kindreds" token_label="Minimum number of families with a candidate variant for a gene to be reported" token_help="This is the number of families required to have a variant fitting the inheritance model in the same gene in order for the gene and its variants to be reported. For example, we may only be interested in candidates where at least 4 families have a variant (with a fitting inheritance pattern) in that gene."> + <param argument="--min-kindreds" name="min_kindreds" type="integer" value="1" min="1" + label="@LABEL@" + help="@HELP@" /> + </xml> + + <xml name="insert_constraint" token_max_repeat="1"> + <repeat name="constraint" title="Additional constraints on variants" default="0" max="@MAX_REPEAT@"> + <expand macro="filter" /> + <yield /> + </repeat> + </xml> + + <xml name="overwritable_where_default" token_default_where=""> + <param name="overwrite_default_filter" type="boolean" checked="false" + label="Overwrite the default constraint of this tool" + help="By default, this tool restricts its analysis to @DEFAULT_WHERE@ and this constraint is applied on top of any constraint expressed above. With this option here selected, your custom constraint, if given, will overwrite the default instead." /> + </xml> + + <xml name="gt_filter" token_default_repeat="0" token_min_repeat="0" token_max_repeat="1"> + <repeat name="filter_by_genotype" title="Genotype filter expression" default="@DEFAULT_REPEAT@" min="@MIN_REPEAT@" max="@MAX_REPEAT@"> + <param argument="--gt-filter" name="gt_filter" type="text" value="" area="True" size="5x50" + label="Restrictions to apply to genotype values" help=""> + <expand macro="sanitize_query" /> + <validator type="expression" message="Genotype filter expression cannot be empty">value.strip()</validator> + </param> + <yield /> + </repeat> + </xml> + + <xml name="sample_filter"> + <repeat name="filter_by_sample" title="Sample filter expression" default="0" max="1"> + <param argument="--sample-filter" name="sample_filter" type="text" area="True" size="5x50" + label="SQL filter to use to filter the sample table" help=""> + <expand macro="sanitize_query" /> + <validator type="expression" message="Sample filter expression cannot be empty">value.strip()</validator> + </param> + <param argument="--in" name="in" type="select" + label="A variant must be in either all, none or any samples passing the sample-query filter" + help=""> + <option value="">Return a variant if it is found in any sample passing the sample filter. (default) </option> + <option value="--in all">Return a variant if it is found in ALL samples passing the sample filter. (all)</option> + <option value="--in none">Return a variant if it is found in NO sample passing the sample filter. (none)</option> + <option value="--in only">Return a variant if it is found in any sample passing the sample filter, and in NO sample NOT passing it. (only)</option> + <option value="--in only all">Return a variant if is found in ALL samples passing the sample filter, and in NO sample NOT passing it. (only all)</option> + </param> + <expand macro="min_kindreds" + label="Minimum number of families in which a variant must pass the sample filter" help=""/> + <param argument="--family-wise" name="family_wise" type="boolean" truevalue="--family-wise" falsevalue="" checked="False" + label="Apply the sample-filter on a family-wise basis" help="If a variant passes the sample filter in at least the minimum number of families specified above it is retained." /> + </repeat> + </xml> + + <xml name="region_filter"> + <repeat name="regions" title="Region Filter" default="0" min="0" + help="Filter variant sites by their position in the genome. If multiple Region Filters are specified, all variants that fall in ONE of the regions are reported."> + <param name="chrom" type="text" label="Chromosome"> + <validator type="expression" message="A chromosome identifier is required when specifying a region filter">value.strip()</validator> + </param> + <param name="start" type="text" label="Region Start"> + <validator type="expression" message="an integer number is required">not value or value.isdigit()</validator> + </param> + <param name="stop" type="text" label="Region End"> + <validator type="expression" message="an integer number is required">not value or value.isdigit()</validator> + </param> + </repeat> + </xml> + <token name="@PROVIDE_ANNO_DATA@"><![CDATA[ mkdir gemini && - ln -s "${annotation_databases.fields.path}/gemini/data" gemini/data && - export GEMINI_CONFIG="${annotation_databases.fields.path}" && + ln -s '${annotation_databases.fields.path}/gemini/data' gemini/data && + export GEMINI_CONFIG='${annotation_databases.fields.path}' && ]]></token> <token name="@MULTILN_SQL_EXPR_TO_CMDLN@"> @@ -119,67 +205,50 @@ #end if </token> - <token name="@CMDLN_SQL_FILTER_FILTER_OPTION@"> - #if str($filter.filter_selector) == 'yes' and $filter.filter: - --filter '${ str( $filter.filter ) }' + <token name="@SET_COLS@"> + #if str($report.report_selector) == 'full': + #set cols = "*" + #else: + #if $report.columns and str($report.columns) != '': + #set $cols = str($report.columns) + #else + #set $cols = '' + #end if + #if str($report.extra_cols).strip(): + #if $cols: + #set $cols = $cols + ', ' + str($report.extra_cols) + #else: + #set $cols = str($report.extra_cols) + #end if + #end if + #if not $cols: + #set $cols = "variant_id, gene" + #end if #end if </token> <token name="@COLUMN_SELECT@"> - #if $report.report_selector != 'all': - --columns "${report.columns} - #if str($report.extra_cols).strip() - #echo ','+','.join(str($report.extra_cols).split()) - #end if - " + @SET_COLS@ + #if $cols != "*" + --columns '$cols' #end if </token> - <xml name="family"> - <param name="families" type="text" value="" label="Comma seperated list of families to restrict the analysis to." help="e.g. Family1,Family3 (--families)"/> - </xml> - - <xml name="lenient"> - <param name="lenient" type="boolean" truevalue="--lenient" falsevalue="" checked="False" label="Loosen the restrictions on family structure"/> - </xml> - - <xml name="unaffected"> - <param name="allow_unaffected" type="boolean" truevalue="--allow-unaffected" falsevalue="" checked="False" label="Report candidates that also impact samples labeled as unaffected." help="(--allow-unaffected)"/> - </xml> - - <xml name="min_kindreds"> - <param name="min_kindreds" type="integer" value="1" label="The min. number of kindreds that must have a candidate variant in a gene" help="default: 1 (--min-kindreds)" /> - </xml> - - <xml name="min_sequence_depth"> - <param name="d" type="integer" value="0" min="0" label="The minimum aligned sequence depth (genotype DP) required for each sample" - help="default: 0 (-d)" /> - </xml> - - <xml name="min_gq"> - <param name="min_gq" type="integer" value="0" label="the minimum genotype quality required for each sample in a family" help="default: 0 (--min-gq)"> - <validator type="in_range" min="0"/> - </param> - </xml> - - <xml name="gt_pl_max"> - <param name="gt_pl_max" type="integer" value="-1" min="-1" label="The maximum phred-scaled genotype likelihod (PL) allowed for each sample in a family" help="default: -1 (not set) (--gt-pl-max)" /> - </xml> - - <xml name="citations"> - <citations> - <citation type="doi">10.1371/journal.pcbi.1003153</citation> - <yield /> - </citations> - </xml> - - <xml name="infile"> - <param name="infile" type="data" format="gemini.sqlite" label="GEMINI database" help="Only files with version @VERSION@ are accepted." > - <options options_filter_attribute="metadata.gemini_version" > - <filter type="add_value" value="@VERSION@" /> - </options> - <validator type="expression" message="This version of Gemini will only work with Gemini files that are for version @VERSION@.">value is not None and value.metadata.gemini_version == "@VERSION@"</validator> - </param> - </xml> - + <token name="@PARSE_REGION_ELEMENTS@"><![CDATA[ + #set $region_elements = [] + #for $r in $regions: + ## The actual chromosome name needs to be single-quoted + ## in SQL, so we need to quote the single quotes like the + ## sanitize_query macro would if the whole was a parameter. + #set $r_elements = ["chrom = '\"'\"'%s'\"'\"'" % str($r.chrom).strip()] + #if str($r.start).strip(): + #silent $r_elements.append("start >= %d" % int($r.start)) + #end if + #if str($r.stop).strip(): + #silent $r_elements.append("end <= %d" % int($r.stop)) + #end if + #silent $region_elements.append("(%s)" % " AND ".join($r_elements)) + #end for + ]]> + </token> </macros>
--- a/repository_dependencies.xml Fri Dec 14 12:55:02 2018 -0500 +++ b/repository_dependencies.xml Fri Jan 11 17:48:39 2019 -0500 @@ -1,4 +1,4 @@ <?xml version="1.0" ?> <repositories description="This requires the GEMINI data manager definition to install all required annotation databases."> - <repository changeset_revision="fe5a9a7d95b0" name="data_manager_gemini_database_downloader" owner="iuc" toolshed="https://toolshed.g2.bx.psu.edu"/> + <repository changeset_revision="f57426daa04d" name="data_manager_gemini_database_downloader" owner="iuc" toolshed="https://toolshed.g2.bx.psu.edu"/> </repositories> \ No newline at end of file
--- a/test-data/gemini_versioned_databases.loc Fri Dec 14 12:55:02 2018 -0500 +++ b/test-data/gemini_versioned_databases.loc Fri Jan 11 17:48:39 2019 -0500 @@ -1,3 +1,3 @@ ## GEMINI versioned databases #DownloadDate dbkey DBversion Description Path -1999-01-01 hg19 181 GEMINI annotations (test snapshot) ${__HERE__}/test-cache +1999-01-01 hg19 200 GEMINI annotations (test snapshot) ${__HERE__}/test-cache
--- a/test-data/test-cache/gemini-config.yaml Fri Dec 14 12:55:02 2018 -0500 +++ b/test-data/test-cache/gemini-config.yaml Fri Jan 11 17:48:39 2019 -0500 @@ -2,12 +2,14 @@ versions: ALL.wgs.phase3_shapeit2_mvncall_integrated_v5a.20130502.sites.tidy.vcf.gz: 4 ESP6500SI.all.snps_indels.tidy.v2.vcf.gz: 2 - ExAC.r0.3.sites.vep.tidy.vcf.gz: 3 + ExAC.r0.3.sites.vep.tidy.vcf.gz: 4 GRCh37-gms-mappability.vcf.gz: 2 - clinvar_20160203.tidy.vcf.gz: 5 + clinvar_20170130.tidy.vcf.gz: 5 cosmic-v68-GRCh37.tidy.vcf.gz: 3 - dbsnp.b141.20140813.hg19.tidy.vcf.gz: 4 + dbsnp.b147.20160601.tidy.vcf.gz: 1 detailed_gene_table_v75: 2 geno2mp.variants.tidy.vcf.gz: 1 + gnomad.exomes.r2.0.1.sites.no-VEP.nohist.tidy.vcf.gz: 2 hg19.rmsk.bed.gz: 2 summary_gene_table_v75: 2 + whole_genome_SNVs.tsv.compressed.gz: 2