feat: Adding somatic neoepitope preparation workflow

docs/index.rst

@@ -69,6 +69,7 @@ Project Info
     step/ngs_data_qc
     step/ngs_mapping
     step/somatic_gene_fusion_calling
+    step/somatic_neoepitope_prediction
     step/somatic_purity_ploidy_estimate
     step/somatic_targeted_seq_cnv_calling
     step/somatic_variant_annotation

snappy_pipeline/apps/snappy_snake.py

@@ -35,6 +35,7 @@
     somatic_gene_fusion_calling,
     somatic_hla_loh_calling,
     somatic_msi_calling,
+    somatic_neoepitope_prediction,
     somatic_purity_ploidy_estimate,
     somatic_targeted_seq_cnv_calling,
     somatic_variant_annotation,
@@ -86,6 +87,7 @@
     "somatic_gene_fusion_calling": somatic_gene_fusion_calling,
     "somatic_hla_loh_calling": somatic_hla_loh_calling,
     "somatic_msi_calling": somatic_msi_calling,
+    "somatic_neoepitope_prediction": somatic_neoepitope_prediction,
     "somatic_purity_ploidy_estimate": somatic_purity_ploidy_estimate,
     "somatic_targeted_seq_cnv_calling": somatic_targeted_seq_cnv_calling,
     "somatic_variant_annotation": somatic_variant_annotation,

snappy_pipeline/models/annotation.py

@@ -35,4 +35,7 @@ class Vep(SnappyModel):
     ]
     num_threads: int = 8
     buffer_size: int = 1000
+    plugins: list[str] = []
+    """To use this option in VEP, you should download the plugin repository from the link https://github.com/Ensembl/VEP_plugins"""
+    plugins_dir: str = ""
     output_options: list[str] = ["everything"]

snappy_pipeline/workflow_model.py

@@ -36,6 +36,9 @@
 from snappy_pipeline.workflows.somatic_variant_filtration.model import SomaticVariantFiltration
 from snappy_pipeline.workflows.somatic_variant_signatures.model import SomaticVariantSignatures
 from snappy_pipeline.workflows.somatic_wgs_cnv_calling.model import SomaticWgsCnvCalling
+from snappy_pipeline.workflows.somatic_neoepitope_prediction.model import (
+    SomaticNeoepitopePrediction,
+)
 from snappy_pipeline.workflows.somatic_wgs_sv_calling.model import SomaticWgsSvCalling
 from snappy_pipeline.workflows.sv_calling_targeted.model import SvCallingTargeted
 from snappy_pipeline.workflows.sv_calling_wgs.model import SvCallingWgs
@@ -113,6 +116,7 @@ class StepConfig(TypedDict, total=False):
     somatic_gene_fusion_calling: SomaticGeneFusionCalling
     somatic_hla_loh_calling: SomaticHlaLohCalling
     somatic_msi_calling: SomaticMsiCalling
+    somatic_neoepitope_prediction: SomaticNeoepitopePrediction
     somatic_purity_ploidy_estimate: SomaticPurityPloidyEstimate
     somatic_targeted_seq_cnv_calling: SomaticTargetedSeqCnvCalling
     somatic_variant_annotation: SomaticVariantAnnotation

snappy_pipeline/workflows/hla_typing/Snakefile

@@ -67,51 +67,53 @@ rule hla_typing_link_out_run:
 
 # OptiType --------------------------------------------------------------------
 
-
-rule hla_typing_optitype_run:
-    input:
-        **wf.get_input_files("optitype", "run"),
-    output:
-        **wf.get_output_files("optitype", "run"),
-    threads: wf.get_resource("optitype", "run", "threads")
-    resources:
-        time=wf.get_resource("optitype", "run", "time"),
-        memory=wf.get_resource("optitype", "run", "memory"),
-        partition=wf.get_resource("optitype", "run", "partition"),
-        tmpdir=wf.get_resource("optitype", "run", "tmpdir"),
-    log:
-        **wf.get_log_file("optitype", "run"),
-    params:
-        args=wf.substep_dispatch("optitype", "get_args", "run"),
-    wrapper:
-        wf.wrapper_path("optitype")
+if "optitype" in wf.w_config.step_config["hla_typing"].tools:
+
+    rule hla_typing_optitype_run:
+        input:
+            **wf.get_input_files("optitype", "run"),
+        output:
+            **wf.get_output_files("optitype", "run"),
+        threads: wf.get_resource("optitype", "run", "threads")
+        resources:
+            time=wf.get_resource("optitype", "run", "time"),
+            memory=wf.get_resource("optitype", "run", "memory"),
+            partition=wf.get_resource("optitype", "run", "partition"),
+            tmpdir=wf.get_resource("optitype", "run", "tmpdir"),
+        log:
+            **wf.get_log_file("optitype", "run"),
+        params:
+            args=wf.substep_dispatch("optitype", "get_args", "run"),
+        wrapper:
+            wf.wrapper_path("optitype")
 
 
 # arcasHLA --------------------------------------------------------------------
 
 
 # NB: reference is updated in the installed package
-rule hla_typing_arcashla_prepare_reference:
-    output:
-        touch("work/arcashla.prepare_reference/out/.done"),
-    log:
-        "work/arcashla.prepare_reference/log/arcashla.prepare_reference.log",
-    wrapper:
-        wf.wrapper_path("arcashla/prepare_reference")
-
-
-rule hla_typing_arcashla_run:
-    input:
-        unpack(wf.get_input_files("arcashla", "run")),
-    output:
-        **wf.get_output_files("arcashla", "run"),
-    threads: wf.get_resource("arcashla", "run", "threads")
-    resources:
-        time=wf.get_resource("arcashla", "run", "time"),
-        memory=wf.get_resource("arcashla", "run", "memory"),
-        partition=wf.get_resource("arcashla", "run", "partition"),
-        tmpdir=wf.get_resource("arcashla", "run", "tmpdir"),
-    log:
-        wf.get_log_file("arcashla", "run"),
-    wrapper:
-        wf.wrapper_path("arcashla/run")
+if "arcashla" in wf.w_config.step_config["hla_typing"].tools:
+
+    rule hla_typing_arcashla_prepare_reference:
+        output:
+            touch("work/arcashla.prepare_reference/out/.done"),
+        log:
+            "work/arcashla.prepare_reference/log/arcashla.prepare_reference.log",
+        wrapper:
+            wf.wrapper_path("arcashla/prepare_reference")
+
+    rule hla_typing_arcashla_run:
+        input:
+            unpack(wf.get_input_files("arcashla", "run")),
+        output:
+            **wf.get_output_files("arcashla", "run"),
+        threads: wf.get_resource("arcashla", "run", "threads")
+        resources:
+            time=wf.get_resource("arcashla", "run", "time"),
+            memory=wf.get_resource("arcashla", "run", "memory"),
+            partition=wf.get_resource("arcashla", "run", "partition"),
+            tmpdir=wf.get_resource("arcashla", "run", "tmpdir"),
+        log:
+            wf.get_log_file("arcashla", "run"),
+        wrapper:
+            wf.wrapper_path("arcashla/run")

snappy_pipeline/workflows/somatic_neoepitope_prediction/Snakefile

@@ -0,0 +1,99 @@
+# -*- coding: utf-8 -*-
+"""CUBI Pipeline somatic neoepitope prediction step Snakefile"""
+
+import os
+
+from snappy_pipeline import expand_ref
+from snappy_pipeline.workflows.somatic_neoepitope_prediction import (
+    SomaticNeoepitopePredictionWorkflow,
+)
+
+__author__ = "Pham Gia Cuong"
+
+
+# Configuration ===============================================================
+
+
+configfile: "config.yaml"
+
+
+# Expand "$ref" JSON pointers in configuration (also works for YAML)
+config, lookup_paths, config_paths = expand_ref("config.yaml", config)
+
+# WorkflowImpl Object Setup ===================================================
+wf = SomaticNeoepitopePredictionWorkflow(workflow, config, lookup_paths, config_paths, os.getcwd())
+
+
+localrules:
+    # Linking files from work/ to output/ should be done locally
+    somatic_neoepitope_preparation_link_out_run,
+
+
+rule all:
+    input:
+        wf.get_result_files(),
+
+
+# Generic linking out ---------------------------------------------------------
+
+
+rule somatic_neoepitope_preparation_link_out_run:
+    input:
+        wf.get_input_files("link_out", "run"),
+    output:
+        wf.get_output_files("link_out", "run"),
+    run:
+        shell(wf.get_shell_cmd("link_out", "run", wildcards))
+
+
+rule somatic_neoepitope_prediction_pvactools_install:
+    output:
+        **wf.get_output_files("pvacseq", "install"),
+    params:
+        container="docker://griffithlab/pvactools",
+    resources:
+        time=wf.get_resource("pvacseq", "install", "time"),
+        memory=wf.get_resource("pvacseq", "install", "memory"),
+        partition=wf.get_resource("pvacseq", "install", "partition"),
+    log:
+        **wf.get_log_file("pvacseq", "install"),
+    wrapper:
+        wf.wrapper_path("singularity")
+
+
+rule somatic_neoepitope_preparation:
+    input:
+        unpack(wf.get_input_files("pvacseq", "prepare")),
+    output:
+        **wf.get_output_files("pvacseq", "prepare"),
+    log:
+        **wf.get_log_file("pvacseq", "prepare"),
+    threads: wf.get_resource("pvacseq", "prepare", "threads")
+    resources:
+        time=wf.get_resource("pvacseq", "prepare", "time"),
+        memory=wf.get_resource("pvacseq", "prepare", "memory"),
+        partition=wf.get_resource("pvacseq", "prepare", "partition"),
+        tmpdir=wf.get_resource("pvacseq", "prepare", "tmpdir"),
+    params:
+        **{"args": wf.get_params("pvacseq", "prepare")},
+    wrapper:
+        wf.wrapper_path("pvactools/combining")
+
+
+rule somatic_neoepitope_prediction:
+    input:
+        unpack(wf.get_input_files("pvacseq", "predict")),
+    output:
+        **wf.get_output_files("pvacseq", "predict"),
+    log:
+        **wf.get_log_file("pvacseq", "predict"),
+    threads: wf.get_resource("pvacseq", "predict", "threads")
+    resources:
+        time=wf.get_resource("pvacseq", "predict", "time"),
+        memory=wf.get_resource("pvacseq", "predict", "memory"),
+        partition=wf.get_resource("pvacseq", "predict", "partition"),
+        tmpdir=wf.get_resource("pvacseq", "predict", "tmpdir"),
+    params:
+        **{"args": wf.get_params("pvacseq", "predict")},
+    wrapper:
+        wf.wrapper_path("pvactools/pvacseq")