Merge pull request #268 from jonasscheid/handle-variants

Refactor variant prediction
nf-core · Feb 3, 2025 · 69080c4 · 69080c4
2 parents 048e86d + 5781a91
commit 69080c4
Show file tree

Hide file tree

Showing 56 changed files with 1,214 additions and 755 deletions.
diff --git a/assets/schema_input.json b/assets/schema_input.json
@@ -14,28 +14,7 @@
                 "meta": ["sample"]
             },
             "alleles": {
-                "anyOf": [
-                    {
-                        "type": "string",
-                        "pattern": "^\\S+\\.txt$"
-                    },
-                    {
-                        "type": "string",
-                        "pattern": "^([A-E]{1}[*][0-9]{2}[:][0-9]{2})(;[A-E]{1}[*][0-9]{2}[:][0-9]{2})*$"
-                    },
-                    {
-                        "type": "string",
-                        "pattern": "^((DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2})(;(DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2})*$"
-                    },
-                    {
-                        "type": "string",
-                        "pattern": "^((DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2}[-](DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2})(;((DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2}[-](DR|DP|DQ)[AB][0-9][*][0-9]{2}[:][0-9]{2}))*$"
-                    },
-                    {
-                        "type": "string",
-                        "pattern": "^[H][-][2][-][A-Za-z]{2,3}$"
-                    }
-                ],
+                "type": "string",
                 "errorMessage": "Alleles must be provided as string or file with extension '.txt'. Please check the documentation for more information.",
                 "meta": ["alleles"]
             },

diff --git a/bin/epaa.py b/bin/epaa.py
diff --git a/bin/split_peptides.py b/bin/split_peptides.py
@@ -5,10 +5,9 @@
 import math
 from pathlib import Path
 
-def split_peptides(input_file, output_base, min_size, max_chunks):
+def split_peptides(input_file, min_size, max_chunks):
     """Splits the peptide input file into smaller chunks in a single pass."""
     input_path = Path(input_file)
-    output_base = Path(output_base)
 
     with input_path.open("r") as infile:
         lines = infile.readlines()  # Read all lines into memory
@@ -24,11 +23,11 @@ def split_peptides(input_file, output_base, min_size, max_chunks):
     chunk_size = max(min_size, math.ceil(total_size / num_chunks))
 
     for chunk_idx in range(num_chunks):
-        chunk_file = output_base.with_name(f"{output_base.stem}_chunk_{chunk_idx}.tsv")
         start = chunk_idx * chunk_size
         end = start + chunk_size
-
-        with chunk_file.open("w") as outfile:
+        # Some upstream processes insert dots in filename (e.g. snpsift split -> variant.chr8.tsv), which can cause downstream issues
+        outfile_name = f"{input_path.stem.replace('.','_')}_chunk_{chunk_idx}.tsv"
+        with open(outfile_name, "w") as outfile:
             outfile.write(header)
             outfile.writelines(data_lines[start:end])
 
@@ -38,12 +37,11 @@ def split_peptides(input_file, output_base, min_size, max_chunks):
 def main():
     parser = argparse.ArgumentParser(description="Split a peptide file into smaller chunks.")
     parser.add_argument("-i", "--input", required=True, help="Input file containing peptides.")
-    parser.add_argument("-o", "--output_base", required=True, help="Base filename for output files.")
     parser.add_argument("--min_size", type=int, required=True, help="Minimum peptides per file.")
     parser.add_argument("--max_chunks", type=int, required=True, help="Maximum number of chunks.")
 
     args = parser.parse_args()
-    split_peptides(args.input, args.output_base, args.min_size, args.max_chunks)
+    split_peptides(args.input, args.min_size, args.max_chunks)
 
 if __name__ == "__main__":
     main()
diff --git a/conf/modules.config b/conf/modules.config
@@ -32,23 +32,30 @@ process {
     }
 
     withName: SNPSIFT_SPLIT {
-        publishDir = [
-            path: { "${params.outdir}/split_input/${meta.sample}" },
-            mode: params.publish_dir_mode
-        ]
+        publishDir  = [ enabled: false ]
     }
 
-    withName: EPYTOPE_PEPTIDE_PREDICTION_VAR {
-        // Argument list needs to end with --somatic_mutation
+    withName: EPYTOPE_VARIANT_PREDICTION {
+        ext.prefix = {"${vcf.baseName}"}
         ext.args   = [
-                    genome_reference != 'grch37' & genome_reference != 'grch38' ? "--genome_reference '${genome_reference}'" : '',
-                    genome_reference == 'grch37'                                ? "--genome_reference 'https://grch37.ensembl.org/'" : '',
-                    genome_reference == 'grch38'                                ? "--genome_reference 'https://www.ensembl.org'" : '',
-                    '--somatic_mutation'
+            genome_reference != 'grch37' & genome_reference != 'grch38' ? "--genome_reference '${genome_reference}'" : '',
+            genome_reference == 'grch37'                                ? "--genome_reference 'https://grch37.ensembl.org/'" : '',
+            genome_reference == 'grch38'                                ? "--genome_reference 'https://www.ensembl.org'" : '',
+            params.proteome_reference ? "--proteome_reference ${params.proteome_reference}" : "",
+            params.fasta_output ? "--fasta_output" : "",
         ].join(' ').trim()
         publishDir = [
-            path: { "${params.outdir}/split_predictions/${meta.sample}" },
-            mode: params.publish_dir_mode
+            path: { "${params.outdir}/epytope" },
+            mode: params.publish_dir_mode,
+            pattern: '*.tsv'
+        ]
+    }
+    withName: CAT_FASTA {
+        ext.prefix = {"${meta.sample}"}
+        publishDir = [
+            path: { "${params.outdir}/epytope" },
+            mode: params.publish_dir_mode,
+            pattern: '*.fasta'
         ]
     }
 

diff --git a/conf/test.config b/conf/test.config
@@ -25,5 +25,8 @@ params {
 
     // Input data
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_variants.csv'
-
+    tools = 'mhcflurry'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
 }
diff --git a/conf/test_full.config b/conf/test_full.config
@@ -16,6 +16,11 @@ params {
 
     // Input data for full size test
     input = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_full_test.csv'
-    tools = 'syfpeithi,mhcflurry,mhcnuggets,mhcnuggetsii'
+    tools = 'mhcflurry,mhcnuggets,mhcnuggetsii'
     //TODO: Add netmhcpan and netmhciipan
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
+    min_peptide_length_classII = 14
+    max_peptide_length_classII = 15
 }
diff --git a/conf/test_grch38.config b/conf/test_grch38.config
@@ -21,4 +21,9 @@ params {
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_variants.csv'
     genome_reference = 'grch38'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
+    min_peptide_length_classII = 14
+    max_peptide_length_classII = 15
 }
diff --git a/conf/test_mhcflurry.config b/conf/test_mhcflurry.config
@@ -21,4 +21,7 @@ params {
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_variants.csv'
     tools = 'mhcflurry'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
 }
diff --git a/conf/test_mhcnuggets.config b/conf/test_mhcnuggets.config
@@ -21,4 +21,9 @@ params {
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_variants_class1_and_2.csv'
     tools = 'mhcnuggets,mhcnuggetsii'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
+    min_peptide_length_classII = 14
+    max_peptide_length_classII = 15
 }
diff --git a/conf/test_netmhciipan.config b/conf/test_netmhciipan.config
@@ -22,4 +22,7 @@ params {
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_peptides_class2.csv'
     tools = 'netmhciipan'
     netmhciipan_path = './non-free/netmhciipan.tar.gz'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classII = 14
+    max_peptide_length_classII = 15
 }
diff --git a/conf/test_netmhcpan.config b/conf/test_netmhcpan.config
@@ -22,4 +22,7 @@ params {
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_peptides.csv'
     tools = 'netmhcpan'
     netmhcpan_path = './non-free/netmhcpan.tar.gz'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
 }
diff --git a/conf/test_peptides.config b/conf/test_peptides.config
@@ -21,4 +21,7 @@ params {
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_peptides.csv'
     tools = 'mhcflurry,mhcnuggets'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
 }
diff --git a/conf/test_peptides_h2.config b/conf/test_peptides_h2.config
@@ -20,5 +20,9 @@ params {
     config_profile_description = 'Peptide list based test profile for mouse alleles by MHCnuggets'
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_peptides_mouse.csv'
-    tools = 'mhcnuggets-class-1'
+    // TODO: test with multiple tools
+    tools = 'mhcflurry,mhcnuggets'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
 }
diff --git a/conf/test_proteins.config b/conf/test_proteins.config
@@ -20,4 +20,10 @@ params {
     config_profile_description = 'Protein input test profile'
 
     input  = params.pipelines_testdata_base_path + 'epitopeprediction/testdata/sample_sheets/sample_sheet_proteins.csv'
+    tools = 'mhcflurry'
+    // Reduce number of possible peptide lengths to speed up test
+    min_peptide_length_classI = 9
+    max_peptide_length_classI = 10
+    min_peptide_length_classII = 14
+    max_peptide_length_classII = 15
 }
diff --git a/modules.json b/modules.json
@@ -5,20 +5,30 @@
         "https://github.com/nf-core/modules.git": {
             "modules": {
                 "nf-core": {
+                    "cat/cat": {
+                        "branch": "master",
+                        "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",
+                        "installed_by": ["modules"]
+                    },
                     "csvtk/concat": {
                         "branch": "master",
                         "git_sha": "aa5c23023134cf2d8b75a95d53557890e40261b9",
                         "installed_by": ["modules"]
                     },
                     "gunzip": {
                         "branch": "master",
-                        "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",
+                        "git_sha": "ce35ce92566b3328b405253543b9b2b4d4e5f4f7",
                         "installed_by": ["modules"]
                     },
                     "multiqc": {
                         "branch": "master",
                         "git_sha": "cf17ca47590cc578dfb47db1c2a44ef86f89976d",
                         "installed_by": ["modules"]
+                    },
+                    "snpsift/split": {
+                        "branch": "master",
+                        "git_sha": "666652151335353eef2fcd58880bcef5bc2928e1",
+                        "installed_by": ["modules"]
                     }
                 }
             },

diff --git a/modules/local/cat_files.nf b/modules/local/cat_files.nf