Add WDL implementation for Salmon

cspell.json

@@ -93,6 +93,7 @@
         "excludelist",
         "bklist",
         "probelist",
+        "incompat",
         // genomics / sequencing terms
         "phred",
         "mapq",
@@ -111,6 +112,7 @@
         "exonic",
         "intragenic",
         "transcriptomic",
+        "transcriptome",
         "endedness",
         "unstranded",
         "Unstranded",

sprocket.toml

@@ -1,5 +1,4 @@
 [check]
-all_lint_rules = true
 except = ["ContainerUri", "TodoComment", "UnusedInput"]
 deny_notes = true
 

test/fixtures/salmon/README.md

@@ -0,0 +1,3 @@
+# Salmon test fixtures
+
+`salmon_index.tar.gz` - built with Salmon 2.6.0 by running the `index` task against the existing `reference/gencode.v50.BCR_ABL1.transcripts.fa.gz` fixture (a small protein-coding transcriptome subset covering the BCR and ABL1 genes).

tools/CHANGELOG.md

@@ -4,6 +4,12 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](http://keepachangelog.com/).
 
+## 2026 September
+
+### Added
+
+- Added WDL implementation for Salmon (`index` and `quant` tasks) [#326](https://github.com/stjudecloud/workflows/pull/326)
+
 ## 2026 August
 
 ### Changed

tools/salmon.wdl

@@ -0,0 +1,292 @@
+version 1.3
+
+task index {
+    meta {
+        description: "Builds a Salmon index from a transcriptome FASTA file, for use in quantification"
+        outputs: {
+            index_tar_gz: "A gzipped TAR file containing the Salmon index files."
+        }
+    }
+
+    parameter_meta {
+        transcripts_fasta: "FASTA format file containing the reference transcriptome to index. Can be gzipped."
+        decoys_fasta: {
+            description: "Optional FASTA file containing decoy genome sequences to improve mapping specificity. Can be gzipped.",
+            group: "Common",
+        }
+        index_name: {
+            description: "Name for the output index, in compressed archive format. The suffix `.tar.gz` will be added.",
+            group: "Common",
+        }
+        use_all_cores: {
+            description: "Use all cores? Recommended for cloud environments.",
+            group: "Resources",
+        }
+        ncpu: {
+            description: "Number of cores to allocate for task",
+            group: "Resources",
+        }
+        modify_disk_size_gb: {
+            description: "Add to or subtract from dynamic disk space allocation. Default disk size is determined by the size of the inputs. Specified in GB.",
+            group: "Resources",
+        }
+        modify_memory_gb: {
+            description: "Add to or subtract from dynamic memory allocation. Default memory is determined by the size of the inputs. Specified in GB.",
+            group: "Resources",
+        }
+    }
+
+    input {
+        File transcripts_fasta
+        File? decoys_fasta
+        String index_name = "salmon_index"
+        Boolean use_all_cores = false
+        Int ncpu = 4
+        Int modify_disk_size_gb = 0
+        Int modify_memory_gb = 0
+    }
+
+    String salmon_index_filename = index_name + ".tar.gz"
+
+    Float transcripts_fasta_size = size(transcripts_fasta, "GB")
+    Float decoys_fasta_size = size(decoys_fasta, "GB")
+    Int disk_size_gb = ceil(transcripts_fasta_size * 4) + ceil(decoys_fasta_size * 4) + 10 + modify_disk_size_gb
+
+    command <<<
+        set -euo pipefail
+
+        n_cores=~{ncpu}
+        if ~{use_all_cores}; then
+            n_cores=$(nproc)
+        fi
+
+        transcripts_name="~{basename(transcripts_fasta, ".gz")}"
+        gunzip -c "~{transcripts_fasta}" > "$transcripts_name" || ln -sf "~{transcripts_fasta}" "$transcripts_name"
+        fasta="$transcripts_name"
+
+        decoys_name="~{if defined(decoys_fasta) then basename(select_first([decoys_fasta]), ".gz") else ""}"
+        if [ -n "$decoys_name" ]; then
+            gunzip -c "~{decoys_fasta}" > "$decoys_name" || ln -sf "~{decoys_fasta}" "$decoys_name"
+            grep "^>" "$decoys_name" | cut -d " " -f1 | sed "s/^>//" > decoys.txt
+            cat "$transcripts_name" "$decoys_name" > combined.fasta
+            fasta=combined.fasta
+        fi
+
+        salmon index \
+            -t "$fasta" \
+            -i "~{index_name}" \
+            ~{if defined(decoys_fasta) then "-d decoys.txt" else ""} \
+            -p "$n_cores"
+
+        tar -czf "~{salmon_index_filename}" "~{index_name}"
+
+        rm -f "$transcripts_name" "$decoys_name" combined.fasta
+    >>>
+
+    output {
+        File index_tar_gz = salmon_index_filename
+    }
+
+    requirements {
+        cpu: ncpu
+        # Based on n of 1 test with GRCh38 and GENCODE v50.
+        memory: "~{if defined(decoys_fasta) then 52 else 5}  + 4 + modify_memory_gb} GB"
+        disks: "~{disk_size_gb} GB"
+        container: "quay.io/biocontainers/salmon:2.6.0--hfa8f182_0"
+        maxRetries: 1
+    }
+}
+
+task quant {
+    meta {
+        description: "Runs `salmon quant` in mapping-based mode to quantify transcript-level expression from RNA-Seq reads, using a pre-built Salmon index."
+        outputs: {
+            quant_results_tar_gz: "A gzipped TAR file containing the Salmon quantification output directory, including `quant.sf`.",
+            quant_sf: "The raw `quant.sf` file, renamed to `<prefix>.quant.sf`, provided alongside the tarballed output."
+        }
+    }
+
+    parameter_meta {
+        index_tar_gz: "A gzipped TAR file containing the Salmon index files. Suitable as the output of the `index` task."
+        read_one_fastqs_gz: "An array of gzipped FASTQ files containing read one information"
+        read_two_fastqs_gz: {
+            description: "An array of gzipped FASTQ files containing read two information. Omit for single-end reads.",
+            group: "Common",
+        }
+        lib_type: {
+            description: "Salmon library type describing the relative orientation and strandedness of paired reads.",
+            help: "Use `A` to let Salmon auto-detect the library type - recommended for most users.",
+            external_help: "https://combine-lab.github.io/salmon/guides/library-types/",
+            group: "Common",
+        }
+        prefix: {
+            description: "Prefix for the Salmon quantification output. The extension `.tar.gz` will be added.",
+            group: "Common",
+        }
+        num_bootstraps: {
+            description: "Compute bootstrapped abundance estimates.",
+            help: "This is done by resampling (with replacement) from the counts assigned to the fragment equivalence classes, and then re-running the optimization procedure for each such sample.",
+            group: "Salmon Options",
+        }
+        incompat_prior: {
+            description: "This parameter governs the a priori probability that a fragment mapping is nonetheless the correct mapping.",
+            help: "Specifically, this is for a fragment mapping or aligning to the reference in a manner incompatible with the prescribed library type.",
+            group: "Salmon Options",
+        }
+        range_factorization_bins: {
+            description: "The range-factorization feature allows using a data-driven likelihood factorization.",
+            help: "This can improve quantification accuracy on certain classes of difficult transcripts.",
+            group: "Salmon Options",
+        }
+        fld_mean: {
+            description: "Allows the user to set the expected mean fragment length of the sequencing library.",
+            help: "Since the empirical fragment length distribution cannot be estimated from the mappings of single-end reads, this is only important when running Salmon with single-end reads.",
+            group: "Salmon Options",
+        }
+        fld_sd: {
+            description: "Allows the user to set the expected standard deviation of the fragment length distribution.",
+            help: "Since the empirical fragment length distribution cannot be estimated from the mappings of single-end reads, this is only important when running Salmon with single-end reads.",
+            group: "Salmon Options",
+        }
+        seq_bias: {
+            description: "Passing this flag will enable it to learn and correct for sequence-specific biases in the input data.",
+            group: "Salmon Options",
+        }
+        gc_bias: {
+            description: "Passing this flag will enable it to learn and correct for fragment-level GC biases in the input data.",
+            group: "Salmon Options",
+        }
+        pos_bias: {
+            description: "Passing this flag will enable modeling of a position-specific fragment start distribution.",
+            group: "Salmon Options",
+        }
+        use_em: {
+            description: "Use the \"standard\" EM algorithm to optimize abundance estimates instead of the variational Bayesian EM algorithm.",
+            group: "Salmon Options",
+        }
+        recover_orphans: {
+            description: "This flag enables orphan \"rescue\" for reads.",
+            group: "Salmon Options",
+        }
+        hard_filter: {
+            description: "This flag turns off soft filtering and range-factorized equivalence classes.",
+            help: "Removes all but the equally highest scoring mappings from the equivalence class label for each fragment.",
+            group: "Salmon Options",
+        }
+        allow_dovetail: {
+            description: "Dovetailing mappings and alignments are considered discordant and discarded by default.",
+            help: "If you wish to consider dovetailing mappings as concordant, you can do so by passing this flag.",
+            group: "Salmon Options",
+        }
+        dump_eq: {
+            description: "If passed, Salmon will write a file in the auxiliary directory, called eq_classes.txt.",
+            help: "Contains the equivalence classes and corresponding counts that were computed during quasi-mapping.",
+            group: "Salmon Options",
+        }
+        write_unmapped_names: {
+            description: "Passing this flag will tell Salmon to write out the names of reads (or mates in paired-end reads) that do not map to the transcriptome.",
+            group: "Salmon Options",
+        }
+        use_all_cores: {
+            description: "Use all cores? Recommended for cloud environments.",
+            group: "Resources",
+        }
+        ncpu: {
+            description: "Number of cores to allocate for task",
+            group: "Resources",
+        }
+        modify_disk_size_gb: {
+            description: "Add to or subtract from dynamic disk space allocation. Default disk size is determined by the size of the inputs. Specified in GB.",
+            group: "Resources",
+        }
+        modify_memory_gb: {
+            description: "Add to or subtract from dynamic memory allocation. Default memory is determined by the size of the inputs. Specified in GB.",
+            group: "Resources",
+        }
+    }
+
+    input {
+        File index_tar_gz
+        Array[File]+ read_one_fastqs_gz
+        Array[File]? read_two_fastqs_gz
+        String lib_type = "A"
+        String prefix = sub(basename(read_one_fastqs_gz[0]), "(([_.][rR](?:ead)?[12])((?:[_.-][^_.-]*?)*?))?\\.(fastq|fq)(\\.gz)?$", "")
+        Int num_bootstraps = 0
+        Float incompat_prior = 0.0
+        Int range_factorization_bins = 4
+        Int fld_mean = 250
+        Int fld_sd = 25
+        Boolean seq_bias = false
+        Boolean gc_bias = false
+        Boolean pos_bias = false
+        Boolean use_em = false
+        Boolean recover_orphans = false
+        Boolean hard_filter = false
+        Boolean allow_dovetail = false
+        Boolean dump_eq = false
+        Boolean write_unmapped_names = false
+        Boolean use_all_cores = false
+        Int ncpu = 4
+        Int modify_disk_size_gb = 0
+        Int modify_memory_gb = 0
+    }
+
+    Array[File] read_twos = select_first([read_two_fastqs_gz, []])
+
+    Float read_one_size = size(read_one_fastqs_gz, "GB")
+    Float read_two_size = size(read_twos, "GB")
+    Float index_size = size(index_tar_gz, "GB")
+    Int disk_size_gb = ceil((read_one_size + read_two_size + index_size) * 3) + 10 + modify_disk_size_gb
+    Int memory_gb = ceil(index_size * 4) + 8 + modify_memory_gb
+
+    command <<<
+        set -euo pipefail
+
+        n_cores=~{ncpu}
+        if ~{use_all_cores}; then
+            n_cores=$(nproc)
+        fi
+
+        mkdir salmon_index
+        tar -xzf "~{index_tar_gz}" -C salmon_index --strip-components 1
+
+        # shellcheck disable=SC2086
+        salmon quant \
+            -i salmon_index \
+            -l "~{lib_type}" \
+            ~{if length(read_twos) > 0 then "-1 " + sep(" ", squote(read_one_fastqs_gz)) + " -2 " + sep(" ", squote(read_twos)) else "-r " + sep(" ", squote(read_one_fastqs_gz))} \
+            -p "$n_cores" \
+            --numBootstraps ~{num_bootstraps} \
+            --incompatPrior ~{incompat_prior} \
+            --rangeFactorizationBins ~{range_factorization_bins} \
+            ~{if length(read_twos) == 0 then "--fldMean " + fld_mean else ""} \
+            ~{if length(read_twos) == 0 then "--fldSD " + fld_sd else ""} \
+            ~{if seq_bias then "--seqBias" else ""} \
+            ~{if gc_bias then "--gcBias" else ""} \
+            ~{if pos_bias then "--posBias" else ""} \
+            ~{if use_em then "--useEM" else ""} \
+            ~{if recover_orphans then "--recoverOrphans" else ""} \
+            ~{if hard_filter then "--hardFilter" else ""} \
+            ~{if allow_dovetail then "--allowDovetail" else ""} \
+            ~{if dump_eq then "--dumpEq" else ""} \
+            ~{if write_unmapped_names then "--writeUnmappedNames" else ""} \
+            -o "~{prefix}"
+
+        cp "~{prefix}/quant.sf" "~{prefix}.quant.sf"
+
+        tar -czf "~{prefix}.tar.gz" "~{prefix}"
+    >>>
+
+    output {
+        File quant_results_tar_gz = prefix + ".tar.gz"
+        File quant_sf = prefix + ".quant.sf"
+    }
+
+    requirements {
+        cpu: ncpu
+        memory: "~{memory_gb} GB"
+        disks: "~{disk_size_gb} GB"
+        container: "quay.io/biocontainers/salmon:2.6.0--hfa8f182_0"
+        maxRetries: 1
+    }
+}

tools/test/salmon.yaml

@@ -0,0 +1,48 @@
+index:
+  - name: builds_index_successfully
+    tags:
+      - reference
+    inputs:
+      transcripts_fasta:
+        - reference/gencode.v50.BCR_ABL1.transcripts.fa.gz
+    assertions:
+      outputs:
+        index_tar_gz:
+          - Name: salmon_index.tar.gz
+  - name: builds_index_with_decoys
+    tags:
+      - reference
+      - slow
+    inputs:
+      transcripts_fasta:
+        - reference/gencode.v50.BCR_ABL1.transcripts.fa.gz
+      decoys_fasta:
+        - reference/GRCh38.chrY_chrM.fa
+    assertions:
+      outputs:
+        index_tar_gz:
+          - Name: salmon_index.tar.gz
+
+quant:
+  - name: quantifies_paired_end_reads
+    inputs:
+      index_tar_gz:
+        - salmon/salmon_index.tar.gz
+      read_one_fastqs_gz:
+        - - fastqs/test_R1.fq.gz
+      read_two_fastqs_gz:
+        - - fastqs/test_R2.fq.gz
+    assertions:
+      outputs:
+        quant_results_tar_gz:
+          - Name: test.tar.gz
+  - name: quantifies_single_end_reads
+    inputs:
+      index_tar_gz:
+        - salmon/salmon_index.tar.gz
+      read_one_fastqs_gz:
+        - - fastqs/test_R1.fq.gz
+    assertions:
+      outputs:
+        quant_results_tar_gz:
+          - Name: test.tar.gz