Resolve "update check design file"

Closes #13 (closed)

workflow/scripts/check_design.py

@@ -5,6 +5,7 @@
 import argparse
 import logging
 import pandas as pd
+import os
 
 EPILOG = '''
 For more details:
@@ -49,14 +50,15 @@ def get_args():
 
 
 def check_design_headers(design, paired, atac):
-    '''Check if design file conforms to sequencing type.'''
+    '''Check if design file has proper headers.'''
 
     # Default headers
     design_template = [
         'sample_id',
         'experiment_id',
         'replicate',
-        'fastq_read1']
+        'fastq_read1',
+        ]
 
     design_headers = list(design.columns.values)
 
@@ -76,6 +78,46 @@ def check_design_headers(design, paired, atac):
         raise Exception("Missing column headers: %s" % list(missing_headers))
 
 
+def check_samples(design):
+    '''Check if design file has the correct sample name mapping.'''
+
+    logger.info("Running sample check.")
+
+    samples = design.groupby('sample_id') \
+                            .apply(list)
+
+    malformated_samples = []
+    chars = set('-.')
+    for sample in samples.index.values:
+        if(any(char.isspace() for char in sample) | any((char in chars) for char in sample)):
+            malformated_samples.append(sample)
+
+    if len(malformated_samples) > 0:
+        logger.error('Malformed samples from design file: %s', list(malformated_samples))
+        raise Exception("Malformed samples from design file: %s" %
+                        list(malformated_samples))
+
+
+def check_experiments(design):
+    '''Check if design file has the correct experiment name mapping.'''
+
+    logger.info("Running experiment check.")
+
+    experiments = design.groupby('experiment_id') \
+                            .apply(list)
+
+    malformated_experiments = []
+    chars = set('-.')
+    for experiment in experiments.index.values:
+        if(any(char.isspace() for char in experiment) | any((char in chars) for char in experiment)):
+            malformated_experiments.append(experiment)
+
+    if len(malformated_experiments) > 0:
+        logger.error('Malformed experiment from design file: %s', list(malformated_experiments))
+        raise Exception("Malformed experiment from design file: %s" %
+                        list(malformated_experiments))
+
+
 def check_controls(design):
     '''Check if design file has the correct control mapping.'''
 
@@ -90,7 +132,7 @@ def check_controls(design):
 
 
 def check_replicates(design):
-    '''Check if design file has unique replicate numbersfor an experiment.'''
+    '''Check if design file has unique replicate numbers for an experiment.'''
 
     logger.info("Running replicate check.")
 
@@ -111,7 +153,7 @@ def check_replicates(design):
 
 
 def check_files(design, fastq, paired):
-    '''Check if design file has the files found.'''
+    '''Check if design file fastq lists are actually present.'''
 
     logger.info("Running file check.")
 
@@ -139,6 +181,24 @@ def check_files(design, fastq, paired):
     return design
 
 
+def get_length(checked_design_df):
+    '''Adds length of reads to design file.'''
+
+    logger.info("Adding read length to design file.")
+
+    fq_len_list = []
+
+    files = checked_design_df['fastq_read1']
+
+    for f in files:
+        len_fq = os.popen("zcat %s | head -n 2 | tail -n 1 | wc -c " % (f)).read().rstrip()
+        fq_len_list.append(len_fq)
+
+    checked_design_df['fq_length'] = fq_len_list
+
+    return checked_design_df
+
+
 def main():
     args = get_args()
     design = args.design
@@ -161,7 +221,12 @@ def main():
         check_controls(design_df)
 
     check_replicates(design_df)
-    new_design_df = check_files(design_df, fastq_df, paired)
+    check_samples(design_df)
+    check_experiments(design_df)
+    checked_design_df = check_files(design_df, fastq_df, paired)
+
+    # Add length of each read to design file
+    new_design_df = get_length(checked_design_df)
 
     # Write out new design file
     new_design_df.to_csv('design.tsv', header=True, sep='\t', index=False)