Merge pull request #112 from TomHarrop/dorado

Dorado basecaller

tools/dorado/.shed.yml

@@ -0,0 +1,23 @@
+---
+auto_tool_repositories:
+  name_template: "{{ tool_id }}"
+  description_template: "{{ tool_name }} from the dorado suite"
+categories:
+  - Sequence Analysis
+description: Dorado is a high-performance, easy-to-use, open source basecaller for Oxford Nanopore reads.
+exclude:
+  - tool_test_output.html
+  - tool_test_output.json
+homepage_url: https://github.com/nanoporetech/dorado
+long_description: >
+  Dorado is a high-performance, easy-to-use, open source basecaller for Oxford Nanopore reads.
+name: dorado
+owner: galaxy-australia
+remote_repository_url: https://github.com/usegalaxy-au/tools-au/tree/main/tools/dorado
+suite:
+  name: suite_dorado
+  description: >
+    Dorado is a high-performance, easy-to-use, open source basecaller for Oxford Nanopore reads.
+  long_description: >
+    Dorado is a high-performance, easy-to-use, open source basecaller for Oxford Nanopore reads.
+type: unrestricted

tools/dorado/README.md

@@ -0,0 +1,48 @@
+
+## Tool versions
+
+Dorado is distributed on
+[DockerHub](https://hub.docker.com/r/nanoporetech/dorado/tags) by nanoporetech.
+The containers are identified by sha256 hash, but not tagged with a version.
+
+We can still use the containers and display the dorado version by hard-coding
+both dorado version and container hash into the wrapper (see `macros.xml`).
+Unfortunately you have to pull a >6 GB container and run `dorado --version` just
+to check the tool version. This also prevents auto-updates of this wrapper.
+
+You can update the list of models at the same time (see
+below). **You must do this when you update the wrapper**.
+
+## Basecalling models
+
+The models are bundled in the container at `/models` and made available by the
+`dorado_models.loc` file. 
+
+The columns are `value`, `container_hash`, `name` and  `path`.
+
+To update the list, modify `tool-data/dorado_models.loc.sample`.
+
+Because models can be added and removed, models are listed **per container** in
+the loc file.
+
+Here's some code to **append** the models from the container with hash
+`1c65eb070a9fc1d88710c4dc09b06541f96fdd28`  to the loc file.
+
+```bash
+export DORADO_HASH="1c65eb070a9fc1d88710c4dc09b06541f96fdd28"
+
+apptainer exec "docker://nanoporetech/dorado:sha${DORADO_HASH}" \
+    ls /models | \
+    awk -v hash="${DORADO_HASH}" '{print hash "_" $0 "\t" hash "\t" $0 "\t/models/" $0}' \
+    >> tool-data/dorado_models.loc.sample
+```
+
+The loc file doesn't have a header, so you can keep it sorted.
+
+```bash
+cp tool-data/dorado_models.loc.sample \
+    tool-data/dorado_models.loc.sample.old &&
+sort -t$'\t' -k1,1V tool-data/dorado_models.loc.sample.old \
+    > tool-data/dorado_models.loc
+```
+

tools/dorado/dorado.xml

@@ -0,0 +1,186 @@
+<tool id="dorado" name="Dorado" version="@VERSION@+galaxy0" python_template_version="3.5" profile="21.05">
+    <description>basecaller for raw Oxford Nanopore data</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="xrefs"/>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+
+ln -s '$pod5_file' ./reads.pod5
+
+&&
+
+dorado basecaller
+--trim '${trim}'
+#if $kit_name
+    --kit-name '${kit_name}'
+#end if
+'${model.fields.path}'
+reads.pod5 
+> calls.bam
+
+&& 
+
+dorado summary
+calls.bam 
+> summary.tsv
+
+    ]]></command>
+    <inputs>
+        <!-- FIXME: add pod5 datatype to Galaxy and change here.
+        https://github.com/galaxyproject/galaxy/pull/18419 -->
+        <param name="pod5_file" type="data" format="binary" label="Raw pod5 file" help="Only pod5 is supported. You can convert fast5 to pod5 with the fast5 to pod5 tool."/>
+        <param name="model" type="select" label="Basecalling model. See the Help section for info on model names.">
+            <options from_data_table="dorado_models">
+                <!-- only allow models that shipped in this container -->
+                <filter type="static_value" column="1" value="@CONTAINER_HASH@"/>
+            </options>
+        </param>
+        <param type="select" argument="--trim" label="DNA adapter and primer trimming" help="Detect and remove any adapter and/or primer sequences from the beginning and end of DNA reads. Note that if you intend to demultiplex the reads, trimming adapters and primers could interfere with correct demultiplexing.">
+            <option value="all" selected="true">Any. Trim any detected adapters or primers.</option>
+            <option value="primers"> Primers. Trim any detected adapters or primers, but if barcoding is enabled the barcode sequences will not be trimmed.</option>
+            <option value="adapters"> Adapters. Trim any detected adapters, but primers will not be trimmed, and if barcoding is enabled then barcodes will not be trimmed either.</option>
+            <option value="none"> None. Nothing will be trimmed.</option>
+        </param>
+        <param type="select" argument="--kit-name" optional="true" label="Enable barcoding with the selected kit name." help="Reads are classified into their barcode groups during basecalling. The classification will be reflected in the read group name as well as in the BC tag of the output record.">
+            <option value="EXP-NBD103">EXP-NBD103</option>
+            <option value="EXP-NBD104">EXP-NBD104</option>
+            <option value="EXP-NBD114">EXP-NBD114</option>
+            <option value="EXP-NBD196">EXP-NBD196</option>
+            <option value="EXP-PBC001">EXP-PBC001</option>
+            <option value="EXP-PBC096">EXP-PBC096</option>
+            <option value="SQK-16S024">SQK-16S024</option>
+            <option value="SQK-16S114-24">SQK-16S114-24</option>
+            <option value="SQK-LWB001">SQK-LWB001</option>
+            <option value="SQK-MLK111-96-XL">SQK-MLK111-96-XL</option>
+            <option value="SQK-MLK114-96-XL">SQK-MLK114-96-XL</option>
+            <option value="SQK-NBD111-24">SQK-NBD111-24</option>
+            <option value="SQK-NBD111-96">SQK-NBD111-96</option>
+            <option value="SQK-NBD114-24">SQK-NBD114-24</option>
+            <option value="SQK-NBD114-96">SQK-NBD114-96</option>
+            <option value="SQK-PBK004">SQK-PBK004</option>
+            <option value="SQK-PCB109">SQK-PCB109</option>
+            <option value="SQK-PCB110">SQK-PCB110</option>
+            <option value="SQK-PCB111-24">SQK-PCB111-24</option>
+            <option value="SQK-PCB114-24">SQK-PCB114-24</option>
+            <option value="SQK-RAB201">SQK-RAB201</option>
+            <option value="SQK-RAB204">SQK-RAB204</option>
+            <option value="SQK-RBK001">SQK-RBK001</option>
+            <option value="SQK-RBK004">SQK-RBK004</option>
+            <option value="SQK-RBK110-96">SQK-RBK110-96</option>
+            <option value="SQK-RBK111-24">SQK-RBK111-24</option>
+            <option value="SQK-RBK111-96">SQK-RBK111-96</option>
+            <option value="SQK-RBK114-24">SQK-RBK114-24</option>
+            <option value="SQK-RBK114-96">SQK-RBK114-96</option>
+            <option value="SQK-RLB001">SQK-RLB001</option>
+            <option value="SQK-RPB004">SQK-RPB004</option>
+            <option value="SQK-RPB114-24">SQK-RPB114-24</option>
+            <option value="TWIST-16-UDI">TWIST-16-UDI</option>
+            <option value="TWIST-96A-UDI">TWIST-96A-UDI</option>
+            <option value="VSK-PTC001">VSK-PTC001</option>
+            <option value="VSK-VMK001">VSK-VMK001</option>
+            <option value="VSK-VMK004">VSK-VMK004</option>
+            <option value="VSK-VPS001">VSK-VPS001</option>
+        </param>
+    </inputs>
+    <outputs>
+        <data format="unsorted.bam" name="out_bam" label="Reads from ${on_string} basecalled by ${tool.name} with model ${model.fields.name}" from_work_dir="calls.bam"/>
+        <data format="tsv" name="out_tsv" label="${tool.name} sequencing summary for ${on_string}" from_work_dir="summary.tsv"/>
+    </outputs>
+    <tests>
+        <!-- test 1 -->
+        <test expect_num_outputs="2">
+            <param name="pod5_file" value="FAL00375_473bf0ed_0.ten_reads.pod5"/>
+            <param name="model" value="dna_r9.4.1_e8_fast@v3.4"/>
+            <param name="trim" value="all"/>
+            <output name="out_bam" ftype="unsorted.bam">
+                <assert_contents>
+                    <has_size size="10000" delta="1000"/>
+                </assert_contents>
+            </output>
+            <output name="out_tsv" ftype="tsv">
+                <assert_contents>
+                    <has_text text="00777c4b-cbd6-4a79-8647-bbe5f5f3f3bf"/>
+                </assert_contents>
+            </output>
+        </test>
+        <!-- test 2: trim parameter -->
+        <test expect_num_outputs="2">
+            <param name="pod5_file" value="FAL00375_473bf0ed_0.ten_reads.pod5"/>
+            <param name="model" value="dna_r9.4.1_e8_fast@v3.4"/>
+            <param name="trim" value="adapters"/>
+            <output name="out_bam" ftype="unsorted.bam">
+                <assert_contents>
+                    <has_size size="10000" delta="1000"/>
+                </assert_contents>
+            </output>
+            <output name="out_tsv" ftype="tsv">
+                <assert_contents>
+                    <has_text text="0072b26f-f37c-4517-afa7-621543ac2187"/>
+                </assert_contents>
+            </output>
+        </test>
+        <!-- test 3: barcode detection -->
+        <test expect_num_outputs="2">
+            <param name="pod5_file" value="SQK-RBK114_BC01_BC04_unclassified.pod5"/>
+            <param name="model" value="dna_r10.4.1_e8.2_400bps_hac@v4.3.0"/>
+            <param name="trim" value="all"/>
+            <param name="kit_name" value="SQK-RBK114-96"/>
+            <output name="out_bam" ftype="unsorted.bam">
+                <assert_contents>
+                    <has_size size="10000" delta="1000"/>
+                </assert_contents>
+            </output>
+            <output name="out_tsv" ftype="tsv">
+                <assert_contents>
+                    <has_size size="1103e241-dd7f-43bc-ae19-9a3c6326ad83"/>
+                    <has_text text="SQK-RBK114-96_barcode04"/>
+                </assert_contents>
+            </output>
+        </test>
+    </tests>
+    <help><![CDATA[
+Basecall raw Nanopore data using Oxford Nanopore’s open source
+`dorado <https://github.com/nanoporetech/dorado/>`__ basecaller.
+
+The input is pod5 format. If you have older data in fast5 format, you
+can convert them using the ``fast5 to pod5`` convert tool.
+
+Basecalling models
+------------------
+
+**TLDR: to decide which model to use, see Oxford Nanopore’s** `table of
+basecalling
+models <https://github.com/nanoporetech/dorado/?tab=readme-ov-file#decoding-dorado-model-names>`__.
+
+The names of Dorado models are structured with each segment
+corresponding to a different aspect of the model separated by
+underscores.
+
+For example, the model ``dna_r10.4.1_e8.2_400bps_hac@v4.3.0`` can be
+decoded as follows:
+
+Analyte Type (``dna``):
+   -  For DNA sequencing, it is represented as dna. If you are using a
+      Direct RNA Sequencing Kit, this will be rna002 or rna004,
+      depending on the kit.
+Pore Type (``r10.4.1``):
+   -  The type of flow cell used.
+Chemistry Type (``e8.2``):
+   -  The chemistry type, which corresponds to the kit used for
+      sequencing. For example, Kit 14 chemistry is denoted by e8.2 and
+      Kit 10 or Kit 9 are denoted by e8.
+Translocation Speed (``400bps``):
+   -  The speed of translocation selected at the run setup in MinKNOW
+Model Type (``hac``):
+   -  The size of the model, where larger models yield more accurate
+      basecalls but take more time. The three types of models are fast,
+      hac, and sup. The fast model is the quickest, sup is the most
+      accurate, and hac provides a balance between speed and accuracy.
+Model Version Number (``v4.3.0``):
+   -  The version of the model. Model updates are regularly released,
+      and higher version numbers typically signify greater accuracy.
+
+    ]]></help>
+</tool>

tools/dorado/dorado_pod5_convert.xml

@@ -0,0 +1,60 @@
+<tool id="dorado_pod5_convert" name="fast5 to pod5" version="@VERSION@+galaxy0" python_template_version="3.5" profile="21.05">
+    <description>converter for raw Oxford Nanopore data</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="xrefs"/>
+    <expand macro="requirements"/>
+    <command><![CDATA[
+
+mkdir input_fast5
+
+&&
+
+tar -xf '$fast5_in' -C input_fast5
+
+&&
+
+pod5 convert fast5
+--threads \${GALAXY_SLOTS:-4}
+--recursive
+--output output.pod5
+input_fast5
+
+    ]]>
+    </command>
+    <inputs>
+        <param name="fast5_in" type="data" format="fast5.tar" label="Oxford Nanopore raw data in fast5 format in a tar archive."/>
+    </inputs>
+    <outputs>
+        <!-- FIXME: add pod5 datatype to Galaxy and change here.
+        https://github.com/galaxyproject/galaxy/pull/18419 -->
+        <data format="binary" name="pod5_out" label="${on_string} converted to pod5 from fast5" from_work_dir="output.pod5"/>
+    </outputs>
+    <tests>
+        <test expect_num_outputs="1">
+            <param name="fast5_in" ftype="fast5.tar" value="FAL00375_473bf0ed_0.ten_reads.0_0.fast5.tar"/>
+            <output name="pod5_out" ftype="binary">
+                <assert_contents>
+                    <has_size value="519736" delta="50000"/>
+                </assert_contents>
+            </output>
+        </test>
+        <test expect_num_outputs="1">
+            <param name="fast5_in" ftype="fast5.tar.gz" value="reads_in_directories.tar.gz"/>
+            <output name="pod5_out" ftype="binary">
+                <assert_contents>
+                    <has_size value="950136" delta="90000"/>
+                </assert_contents>
+            </output>
+        </test>
+    </tests>
+    <help><![CDATA[
+        Convert fast5 to `pod5 <https://github.com/nanoporetech/pod5-file-format>`__ for basecalling with Dorado.
+
+        Combine all your fast5 files into a single tar archive, and optionally
+        compress the archive with Gzip, Bzip2 or XZ, before uploading it to
+        Galaxy.
+
+            ]]></help>
+</tool>

tools/dorado/macros.xml

@@ -0,0 +1,15 @@
+<macros>
+    <!-- UPDATING: pull the latest container and check the version. Update both tokens. You MUST also update the model list. See README.md for more.  -->
+    <token name="@VERSION@">0.7.1+80da5f5</token>
+    <token name="@CONTAINER_HASH@">1c65eb070a9fc1d88710c4dc09b06541f96fdd28</token>
+    <xml name="requirements">
+        <requirements>
+            <container type="docker">nanoporetech/dorado:sha@CONTAINER_HASH@</container>
+        </requirements>
+    </xml>
+    <xml name="xrefs">
+        <xrefs>
+            <xref type="bio.tools">dorado</xref>
+        </xrefs>
+    </xml>
+</macros>

tools/dorado/test-data/dorado_models.loc

@@ -0,0 +1 @@
+../tool-data/dorado_models.loc.sample