Merge pull request #4 from genomic-medicine-sweden/input

structure the samplesheet
genomic-medicine-sweden · Dec 1, 2023 · 7c8e671 · 7c8e671
2 parents 11b263c + 6022888
commit 7c8e671
Show file tree

Hide file tree

Showing 2 changed files with 72 additions and 52 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -5,7 +5,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## v1.0dev - [date]
 
-Initial release of nf-core/metaval, created with the [nf-core](https://nf-co.re/) template.
+Initial release of metaval, created with the [nf-core](https://nf-co.re/) template.
 
 ### `Added`
 

diff --git a/bin/check_samplesheet.py b/bin/check_samplesheet.py
@@ -1,9 +1,5 @@
 #!/usr/bin/env python
 
-
-"""Provide a command line tool to validate and transform tabular samplesheets."""
-
-
 import argparse
 import csv
 import logging
@@ -13,7 +9,6 @@
 
 logger = logging.getLogger()
 
-
 class RowChecker:
     """
     Define a service that can validate and transform each given row.
@@ -24,17 +19,22 @@ class RowChecker:
 
     """
 
-    VALID_FORMATS = (
-        ".fq.gz",
-        ".fastq.gz",
-    )
+    VALID_FORMATS = (".fq.gz", ".fastq.gz")
 
     def __init__(
         self,
         sample_col="sample",
-        first_col="fastq_1",
-        second_col="fastq_2",
-        single_col="single_end",
+        run_accession_col="run_accession",
+        instrument_platform_col="instrument_platform",
+        reads_type_col="reads_type",
+        fastq_1_col="fastq_1",
+        fastq_2_col="fastq_2",
+        fasta_col="fasta",
+        kraken2_report_col="kraken2_report",
+        kraken2_classifiedout_col="kraken2_classifiedout",
+        centrifuge_out_col="centrifuge_out",
+        centrifuge_result_col="centrifuge_result",
+        diamond_col="diamond",
         **kwargs,
     ):
         """
@@ -43,20 +43,43 @@ def __init__(
         Args:
             sample_col (str): The name of the column that contains the sample name
                 (default "sample").
-            first_col (str): The name of the column that contains the first (or only)
-                FASTQ file path (default "fastq_1").
-            second_col (str): The name of the column that contains the second (if any)
-                FASTQ file path (default "fastq_2").
-            single_col (str): The name of the new column that will be inserted and
-                records whether the sample contains single- or paired-end sequencing
-                reads (default "single_end").
+            run_accession_col (str): The name of the column that contains the run accession
+                (default "run_accession").
+            instrument_platform_col (str): The name of the column that contains the instrument platform
+                (default "instrument_platform").
+            reads_type_col (str): The name of the column that contains the reads type, shortread or longread
+                (default "reads_type").
+            fastq_1_col (str): The name of the column that contains the first (or only)
+                FASTQ file path (default "fastq_1") from bowtie2 unmapped read 1 against the human genome.
+            fastq_2_col (str): The name of the column that contains the second (if any)
+                FASTQ file path (default "fastq_2") from bowtie2 unmapped read 2 against the human genome.
+            fasta_col (str): The name of the column that contains the FASTA information
+                (default "fasta") from minimap2 unmapped read against the human genome.
+            kraken2_report_col (str): The name of the column that contains the kraken2 report
+                (default "kraken2_report").
+            kraken2_classifiedout_col (str): The name of the column that contains the kraken2 classifiedout
+                (default "kraken2_classifiedout") with the format extension "kraken2.kraken2.classifiedreads.txt".
+            centrifuge_out_col (str): The name of the column that contains the centrifuge out (kraken2-like report)
+                (default "centrifuge_out").
+            centrifuge_result_col (str): The name of the column that contains the centrifuge result
+                (default "centrifuge_result") with the format extension "centrifuge.results.txt".
+            diamond_col (str): The name of the column that contains the diamond information
+                (default "diamond") with the format extension ".csv".
 
         """
         super().__init__(**kwargs)
         self._sample_col = sample_col
-        self._first_col = first_col
-        self._second_col = second_col
-        self._single_col = single_col
+        self._run_accession_col = run_accession_col
+        self._instrument_platform_col = instrument_platform_col
+        self._reads_type_col = reads_type_col
+        self._fastq_1_col = fastq_1_col
+        self._fastq_2_col = fastq_2_col
+        self._fasta_col = fasta_col
+        self._kraken2_report_col = kraken2_report_col
+        self._kraken2_classifiedout_col = kraken2_classifiedout_col
+        self._centrifuge_out_col = centrifuge_out_col
+        self._centrifuge_result_col = centrifuge_result_col
+        self._diamond_col = diamond_col
         self._seen = set()
         self.modified = []
 
@@ -73,7 +96,7 @@ def validate_and_transform(self, row):
         self._validate_first(row)
         self._validate_second(row)
         self._validate_pair(row)
-        self._seen.add((row[self._sample_col], row[self._first_col]))
+        self._seen.add((row[self._sample_col], row[self._fastq_1_col]))
         self.modified.append(row)
 
     def _validate_sample(self, row):
@@ -85,25 +108,22 @@ def _validate_sample(self, row):
 
     def _validate_first(self, row):
         """Assert that the first FASTQ entry is non-empty and has the right format."""
-        if len(row[self._first_col]) <= 0:
+        if len(row[self._fastq_1_col]) <= 0:
             raise AssertionError("At least the first FASTQ file is required.")
-        self._validate_fastq_format(row[self._first_col])
+        self._validate_fastq_format(row[self._fastq_1_col])
 
     def _validate_second(self, row):
         """Assert that the second FASTQ entry has the right format if it exists."""
-        if len(row[self._second_col]) > 0:
-            self._validate_fastq_format(row[self._second_col])
+        if len(row[self._fastq_2_col]) > 0:
+            self._validate_fastq_format(row[self._fastq_2_col])
 
     def _validate_pair(self, row):
         """Assert that read pairs have the same file extension. Report pair status."""
-        if row[self._first_col] and row[self._second_col]:
-            row[self._single_col] = False
-            first_col_suffix = Path(row[self._first_col]).suffixes[-2:]
-            second_col_suffix = Path(row[self._second_col]).suffixes[-2:]
+        if row[self._fastq_1_col] and row[self._fastq_2_col]:
+            first_col_suffix = Path(row[self._fastq_1_col]).suffixes[-1]
+            second_col_suffix = Path(row[self._fastq_2_col]).suffixes[-1]
             if first_col_suffix != second_col_suffix:
                 raise AssertionError("FASTQ pairs must have the same file extensions.")
-        else:
-            row[self._single_col] = True
 
     def _validate_fastq_format(self, filename):
         """Assert that a given filename has one of the expected FASTQ extensions."""
@@ -118,7 +138,7 @@ def validate_unique_samples(self):
         Assert that the combination of sample name and FASTQ filename is unique.
 
         In addition to the validation, also rename all samples to have a suffix of _T{n}, where n is the
-        number of times the same sample exist, but with different FASTQ files, e.g., multiple runs per experiment.
+        number of times the same sample exists, but with different FASTQ files, e.g., multiple runs per experiment.
 
         """
         if len(self._seen) != len(self.modified):
@@ -164,7 +184,7 @@ def sniff_format(handle):
 
 def check_samplesheet(file_in, file_out):
     """
-    Check that the tabular samplesheet has the structure expected by nf-core pipelines.
+    Check that the tabular samplesheet has the structure expected by metaval.
 
     Validate the general shape of the table, expected columns, and each row. Also add
     an additional column which records whether one or two FASTQ reads were found.
@@ -174,26 +194,26 @@ def check_samplesheet(file_in, file_out):
             CSV, TSV, or any other format automatically recognized by ``csv.Sniffer``.
         file_out (pathlib.Path): Where the validated and transformed samplesheet should
             be created; always in CSV format.
-
-    Example:
-        This function checks that the samplesheet follows the following structure,
-        see also the `viral recon samplesheet`_::
-
-            sample,fastq_1,fastq_2
-            SAMPLE_PE,SAMPLE_PE_RUN1_1.fastq.gz,SAMPLE_PE_RUN1_2.fastq.gz
-            SAMPLE_PE,SAMPLE_PE_RUN2_1.fastq.gz,SAMPLE_PE_RUN2_2.fastq.gz
-            SAMPLE_SE,SAMPLE_SE_RUN1_1.fastq.gz,
-
-    .. _viral recon samplesheet:
-        https://raw.githubusercontent.com/nf-core/test-datasets/viralrecon/samplesheet/samplesheet_test_illumina_amplicon.csv
-
     """
-    required_columns = {"sample", "fastq_1", "fastq_2"}
+    required_columns = [
+        "sample",
+        "run_accession",
+        "instrument_platform",
+        "reads_type",
+        "fastq_1",
+        "fastq_2",
+        "fasta",
+        "kraken2_report",
+        "kraken2_classifiedout",
+        "centrifuge_out",
+        "centrifuge_result",
+        "diamond",
+    ]
     # See https://docs.python.org/3.9/library/csv.html#id3 to read up on `newline=""`.
     with file_in.open(newline="") as in_handle:
         reader = csv.DictReader(in_handle, dialect=sniff_format(in_handle))
         # Validate the existence of the expected header columns.
-        if not required_columns.issubset(reader.fieldnames):
+        if not required_columns == reader.fieldnames:
             req_cols = ", ".join(required_columns)
             logger.critical(f"The sample sheet **must** contain these column headers: {req_cols}.")
             sys.exit(1)
@@ -207,7 +227,7 @@ def check_samplesheet(file_in, file_out):
                 sys.exit(1)
         checker.validate_unique_samples()
     header = list(reader.fieldnames)
-    header.insert(1, "single_end")
+ #   header.insert(1, "single_end")
     # See https://docs.python.org/3.9/library/csv.html#id3 to read up on `newline=""`.
     with file_out.open(mode="w", newline="") as out_handle:
         writer = csv.DictWriter(out_handle, header, delimiter=",")