[WIP] odgi update + new tools

tools/odgi/build.xml

@@ -1,18 +1,14 @@
-<tool id="odgi_build" name="odgi build" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="21.05">
+<tool id="odgi_build" name="odgi build" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="24.1">
     <description>construct a dynamic succinct variation graph</description>
     <macros>
         <import>macros.xml</import>
     </macros>
     <expand macro="requirements"/>
     <command detect_errors="exit_code"><![CDATA[
 cp '$gfa' input.gfa &&
-
-odgi build
--g input.gfa
--o '$odgi_output'
-$sort
-#if $to_gfa:
-    --to-gfa > '$gfa_output'
+odgi build -g input.gfa -o '$odgi_output' $sort
+#if $to_gfa
+&& odgi view -i '$odgi_output' $to_gfa > '$gfa_output'
 #end if
     ]]></command>
     <inputs>
@@ -31,19 +27,29 @@ $sort
         </data>
     </outputs>
     <tests>
-        <test>
+        <test expect_num_outputs="2">
             <param name="gfa" value="t.gfa" />
             <param name="sort" value="true" />
             <param name="to_gfa" value="true" />
             <output name="odgi_output" file="note5_out.og" />
             <output name="gfa_output" file="note5.gfa" />
         </test>
     </tests>
-    <help><![CDATA[
-optimized dynamic genome/graph implementation (odgi)
-----------------------------------------------------
+    <help format="markdown"><![CDATA[
+@HELP_HEADER@
+
+`odgi build` converts a GFA file into an ODGI graph, optionally topologically sorting nodes and emitting a companion GFA view.
+
+### Inputs
+
+- **Construct the graph from this GFA input file**: Source GFA file to ingest.
+- **Topological sort the graph**: Apply generalized topological sorting before writing the `.og`.
+- **Output additional GFA file**: Re-export the resulting graph in GFA format.
+
+### Outputs
 
-Construct a dynamic succinct variation graph (build)
+- **odgi**: The constructed graph in ODGI format.
+- **GFA**: Optional GFA view of the constructed graph.
     ]]></help>
     <expand macro="citations">
     </expand>

tools/odgi/cover.xml

@@ -0,0 +1,63 @@
+<tool id="odgi_cover" name="odgi cover" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="24.1">
+    <description>generate path covers for graph components</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+odgi cover
+--idx '$idx'
+--out '$output'
+#if str($hogwild_depth)
+    --hogwild-depth $hogwild_depth
+#end if
+#if str($num_paths_per_component)
+    --num-paths-per-component $num_paths_per_component
+#end if
+#if str($node_windows_size)
+    --node-windows-size $node_windows_size
+#end if
+#if str($min_node_coverage)
+    --min-node-coverage $min_node_coverage
+#end if
+$ignore_paths
+#if str($save_coverages)
+    --write-node-coverages '$node_coverages'
+#end if
+@THREADS@
+    ]]></command>
+    <inputs>
+        <expand macro="graph_input_param"/>
+        <param argument="--hogwild-depth" type="float" optional="true" min="0" label="Cover graph to this average depth" help="Randomly cover the graph until reaching the specified average depth."/>
+        <param argument="--num-paths-per-component" type="integer" optional="true" min="1" label="Number of paths to generate per component"/>
+        <param argument="--node-windows-size" type="integer" optional="true" min="2" label="Node window size when extending paths"/>
+        <param argument="--min-node-coverage" type="integer" optional="true" min="1" label="Minimum node coverage to reach"/>
+        <param argument="--ignore-paths" type="boolean" truevalue="--ignore-paths" falsevalue="" checked="false" label="Ignore embedded paths when initializing coverage"/>
+        <param argument="--save-coverages" type="boolean" truevalue="true" falsevalue="" checked="false" label="Write node coverages table" help="Generate a tabular file with per-node coverage statistics."/>
+    </inputs>
+    <outputs>
+        <data name="output" format="odgi" label="${tool.name} on ${on_string}"/>
+        <data name="node_coverages" format="tabular" label="${tool.name} on ${on_string}: node coverages">
+            <filter>save_coverages</filter>
+        </data>
+    </outputs>
+    <tests/>
+    <help format="markdown"><![CDATA[
+@HELP_HEADER@
+
+`odgi cover` generates path covers for each connected component of the graph.
+
+### Inputs
+
+- **Input odgi graph**: Graph to cover.
+- **Average depth / Paths per component / Node window size / Minimum node coverage**: Control coverage generation heuristics.
+- **Ignore embedded paths**: Start covering from scratch ignoring existing paths.
+- **Write node coverages table**: Emit per-node coverage values.
+
+### Outputs
+
+- **odgi**: Graph augmented with generated cover paths.
+- **node coverages**: Optional tabular coverage summary.
+    ]]></help>
+    <expand macro="citations"/>
+</tool>

tools/odgi/depth.xml

@@ -0,0 +1,88 @@
+<tool id="odgi_depth" name="odgi depth" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="24.1">
+    <description>compute path depth statistics</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+odgi depth
+--input '$input'
+#if str($subset_paths)
+    --subset-paths '$subset_paths'
+#end if
+#if str($path)
+    --path '$path'
+#end if
+#if str($paths)
+    --paths '$paths'
+#end if
+#if str($graph_pos)
+    --graph-pos '$graph_pos'
+#end if
+#if str($graph_pos_file)
+    --graph-pos-file '$graph_pos_file'
+#end if
+#if str($path_pos)
+    --path-pos '$path_pos'
+#end if
+#if str($path_pos_file)
+    --path-pos-file '$path_pos_file'
+#end if
+#if str($bed_input)
+    --bed-input '$bed_input'
+#end if
+$graph_depth_table
+$graph_depth_vec
+$path_depth
+$self_depth
+$summarize
+#if str($windows_in)
+    --windows-in '$windows_in'
+#end if
+#if str($windows_out)
+    --windows-out '$windows_out'
+#end if
+@THREADS@
+> '$output'
+    ]]></command>
+    <inputs>
+        <expand macro="graph_input_param_input"/>
+        <param argument="--subset-paths" type="data" format="txt" optional="true" label="Subset of paths for depth calculation" help="Text file with one path name per line."/>
+        <param argument="--path" type="text" optional="true" label="Single path to analyze"/>
+        <param argument="--paths" type="data" format="txt" optional="true" label="Report depth for these paths" help="Text file with one path name per line."/>
+        <param argument="--graph-pos" type="text" optional="true" label="Graph position (node[,offset[,strand]])"/>
+        <param argument="--graph-pos-file" type="data" format="txt" optional="true" label="File of graph positions"/>
+        <param argument="--path-pos" type="text" optional="true" label="Path position (path[,offset[,strand]])"/>
+        <param argument="--path-pos-file" type="data" format="txt" optional="true" label="File of path positions"/>
+        <param argument="--bed-input" type="data" format="bed" optional="true" label="BED intervals to evaluate"/>
+        <param argument="--graph-depth-table" type="boolean" truevalue="--graph-depth-table" falsevalue="" checked="false" label="Write node depth/unique depth table"/>
+        <param argument="--graph-depth-vec" type="boolean" truevalue="--graph-depth-vec" falsevalue="" checked="false" label="Write graph depth vector"/>
+        <param argument="--path-depth" type="boolean" truevalue="--path-depth" falsevalue="" checked="false" label="Write per-base path depth"/>
+        <param argument="--self-depth" type="boolean" truevalue="--self-depth" falsevalue="" checked="false" label="Write self depth per base"/>
+        <param argument="--summarize" type="boolean" truevalue="--summarize" falsevalue="" checked="false" label="Summarize depth distribution"/>
+        <param argument="--windows-in" type="text" optional="true" label="Depth windows inside range (LEN:MIN:MAX)"/>
+        <param argument="--windows-out" type="text" optional="true" label="Depth windows outside range (LEN:MIN:MAX)"/>
+    </inputs>
+    <outputs>
+        <expand macro="standard_tabular_output"/>
+    </outputs>
+    <tests/>
+    <help format="markdown"><![CDATA[
+@HELP_HEADER@
+
+`odgi depth` reports depth metrics for selected paths, positions, or intervals.
+
+### Inputs
+
+- **Input odgi graph**: Graph to profile.
+- **Subset/path/path lists**: Limit the analysis to specific paths.
+- **Positions / BED intervals**: Query exact positions or ranges.
+- **Depth summary toggles**: Emit tables or vectors describing depth per node or path.
+- **Depth windows**: Report intervals within or outside depth ranges.
+
+### Outputs
+
+- **Tabular**: Depth metrics written in tab-separated form.
+    ]]></help>
+    <expand macro="citations"/>
+</tool>

tools/odgi/layout.xml

@@ -0,0 +1,114 @@
+<tool id="odgi_layout" name="odgi layout" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="24.1">
+    <description>compute 2D path-guided layouts</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+odgi layout
+--idx '$idx'
+--out '$layout_output'
+#if $tsv
+    $tsv '$layout_tsv'
+#end if
+#if str($path_index)
+    --path-index '$path_index'
+#end if
+#if str($path_sgd_use_paths)
+    --path-sgd-use-paths '$path_sgd_use_paths'
+#end if
+#if str($layout_initialization)
+    --layout-initialization '$layout_initialization'
+#end if
+#if str($path_sgd_min_term_updates_paths)
+    --path-sgd-min-term-updates-paths $path_sgd_min_term_updates_paths
+#end if
+#if str($path_sgd_min_term_updates_nodes)
+    --path-sgd-min-term-updates-nodes $path_sgd_min_term_updates_nodes
+#end if
+#if str($path_sgd_delta)
+    --path-sgd-delta $path_sgd_delta
+#end if
+#if str($path_sgd_eta)
+    --path-sgd-eta $path_sgd_eta
+#end if
+#if str($path_sgd_eta_max)
+    --path-sgd-eta-max $path_sgd_eta_max
+#end if
+#if str($path_sgd_iter_max)
+    --path-sgd-iter-max $path_sgd_iter_max
+#end if
+#if str($path_sgd_iteration_max_learning_rate)
+    --path-sgd-iteration-max-learning-rate $path_sgd_iteration_max_learning_rate
+#end if
+#if str($path_sgd_zipf_theta)
+    --path-sgd-zipf-theta $path_sgd_zipf_theta
+#end if
+#if str($path_sgd_zipf_space)
+    --path-sgd-zipf-space $path_sgd_zipf_space
+#end if
+#if str($path_sgd_zipf_space_max)
+    --path-sgd-zipf-space-max $path_sgd_zipf_space_max
+#end if
+#if str($path_sgd_zipf_space_quantization_step)
+    --path-sgd-zipf-space-quantization-step $path_sgd_zipf_space_quantization_step
+#end if
+#if str($path_sgd_snapshot)
+    --path-sgd-snapshot '$path_sgd_snapshot'
+#end if
+@THREADS@
+    ]]></command>
+    <inputs>
+        <expand macro="graph_input_param"/>
+        <param argument="--tsv" type="boolean" truevalue="--tsv" falsevalue="" checked="false" label="Export TSV layout alongside binary layout"/>
+        <param argument="--path-index" type="data" format="txt" optional="true" label="Precomputed path index (.xp)"/>
+        <param argument="--path-sgd-use-paths" type="data" format="txt" optional="true" label="Paths to sample for PG-SGD"/>
+        <param argument="--layout-initialization" type="select" optional="true" label="Layout initialization mode">
+            <option value="d" selected="true">Node rank in X, Gaussian noise in Y (default)</option>
+            <option value="r">Uniform noise in X and Y</option>
+            <option value="u">Node rank in X, uniform noise in Y</option>
+            <option value="g">Gaussian noise in X and Y</option>
+            <option value="h">Hilbert curve initialization</option>
+        </param>
+        <param argument="--path-sgd-min-term-updates-paths" type="float" optional="true" min="0" label="Minimum term updates (paths multiple)"/>
+        <param argument="--path-sgd-min-term-updates-nodes" type="float" optional="true" min="0" label="Minimum term updates (nodes multiple)"/>
+        <param argument="--path-sgd-delta" type="float" optional="true" min="0" label="Maximum displacement threshold (bp)"/>
+        <param argument="--path-sgd-eta" type="float" optional="true" min="0" label="Final learning rate"/>
+        <param argument="--path-sgd-eta-max" type="float" optional="true" min="0" label="Initial learning rate"/>
+        <param argument="--path-sgd-iter-max" type="integer" optional="true" min="1" label="Maximum PG-SGD iterations"/>
+        <param argument="--path-sgd-iteration-max-learning-rate" type="integer" optional="true" min="0" label="Iteration where learning rate peaks"/>
+        <param argument="--path-sgd-zipf-theta" type="float" optional="true" label="Zipf theta"/>
+        <param argument="--path-sgd-zipf-space" type="integer" optional="true" min="1" label="Zipf space size"/>
+        <param argument="--path-sgd-zipf-space-max" type="integer" optional="true" min="1" label="Zipf space max before quantization"/>
+        <param argument="--path-sgd-zipf-space-quantization-step" type="integer" optional="true" min="1" label="Zipf quantization step size"/>
+        <param argument="--path-sgd-snapshot" type="text" optional="true" label="Snapshot prefix for intermediate layouts"/>
+    </inputs>
+    <outputs>
+        <data name="layout_output" format="binary" label="${tool.name} on ${on_string}"/>
+        <data name="layout_tsv" format="tabular" label="${tool.name} on ${on_string}: coordinates">
+            <filter>tsv</filter>
+        </data>
+    </outputs>
+    <tests/>
+    <help format="markdown"><![CDATA[
+@HELP_HEADER@
+
+`odgi layout` computes 2D coordinates for ODGI graphs using path-guided stochastic gradient descent (PG-SGD).
+The binary `.lay` output can be fed into visualization or downstream statistics; optionally export a TSV table of node coordinates.
+
+### Inputs
+
+- **Input odgi graph**: Sorted, id-compacted graph to lay out.
+- **Export TSV layout alongside binary layout**: Produce a human-readable coordinate table as well as the binary `.lay`.
+- **Precomputed path index**: Reuse a `.xp` index to accelerate layout.
+- **Layout initialization mode**: Choose how the initial coordinates are seeded.
+- **PG-SGD parameters**: Min term updates, learning rates, iteration cap, and Zipf sampling controls fine-tune convergence.
+- **Snapshot prefix**: Write intermediate layouts for debugging or animation.
+
+### Outputs
+
+- **Binary**: Primary `.lay` file containing node coordinates.
+- **coordinates**: Optional TSV with node ID, X, Y triples.
+    ]]></help>
+    <expand macro="citations"/>
+</tool>

tools/odgi/macros.xml

@@ -1,15 +1,33 @@
-<?xml version="1.0"?>
 <macros>
     <xml name="requirements">
         <requirements>
             <requirement type="package" version="@TOOL_VERSION@">odgi</requirement>
         </requirements>
     </xml>
-    <token name="@TOOL_VERSION@">0.3</token>
-    <token name="@VERSION_SUFFIX@">1</token>
+    <token name="@TOOL_VERSION@">0.9.3</token>
+    <token name="@VERSION_SUFFIX@">0</token>
+    <token name="@HELP_HEADER@">optimized dynamic genome/graph implementation (odgi)
+----------------------------------------------------</token>
+    <token name="@THREADS@">--threads=${GALAXY_SLOTS:-1}</token>
+    <xml name="graph_input_param">
+        <param argument="--idx" type="data" format="odgi" label="Input odgi graph"/>
+    </xml>
+    <xml name="graph_input_param_input">
+        <param argument="--input" type="data" format="odgi" label="Input odgi graph"/>
+    </xml>
+    <xml name="group_delim_params">
+        <param argument="--delim" type="text" optional="true" label="Delimiter to define path groups" help="Split path names at this character to group them before summarizing statistics or haplotypes."/>
+        <param argument="--delim-pos" type="integer" optional="true" min="1" label="Use the N-th occurrence of the delimiter" help="Specify which occurrence of the delimiter to use when extracting the group identifier."/>
+    </xml>
+    <xml name="standard_tabular_output">
+        <data name="output" format="tabular" label="${tool.name} on ${on_string}"/>
+    </xml>
+    <xml name="standard_text_output">
+        <data name="output" format="txt" label="${tool.name} on ${on_string}"/>
+    </xml>
     <xml name="citations">
         <citations>
-            <yield />
+            <citation type="doi">10.1093/bioinformatics/btac308</citation>
         </citations>
     </xml>
 </macros>