[WIP] nextflow script generation

bin/PSOT

@@ -2,13 +2,87 @@
 import argparse
 from config import load_config
 
-parser = argparse.ArgumentParser(description='Make bioinformatic observations on aminoacid sequences')
-parser.add_argument('--fasta', '-f', help='A fasta file with aminoacid sequences')
-parser.add_argument('--listanalyses', '-l', action='store_true', help='Show available analysis steps')
+config = load_config()
 
-args = parser.parse_args()
+def main():
+    parser = argparse.ArgumentParser(description='Make bioinformatic observations on aminoacid sequences')
 
-config = load_config()
+    subparsers = parser.add_subparsers()
+    info_parser = subparsers.add_parser('info')
+    info_parser.add_argument('--listanalyses', '-l', action='store_true', help='Show available analysis steps')
+    info_parser.set_defaults(func=info)
+
+    analyze_parser = subparsers.add_parser('analyze')
+
+    analyze_parser.add_argument('--fasta', '-f', required=True, help='A fasta file with aminoacid sequences')
+    analyze_parser.add_argument('--output', '-o', required=True, help='The output directory for the json documents')
+    analyze_parser.add_argument('--profile', '-p', default='fast', help='The profile to use')
+    analyze_parser.add_argument('--config', '-c', help='The config to use')
+    analyze_parser.set_defaults(func=analyze)
+
+    args = parser.parse_args()
+    args.func(args)
+
+
+def info(args):
+    show_analyses()
+
+def analyze(args):
+    print()
+    modules = get_modules_for_profile(args.profile)
+    generate_nextflow_script(modules)
+    
+def get_modules_for_profile(profile):
+    # find profile by name
+    p = [x for x in config['profiles'] if x['name'] == profile][0]
+    # find all modules for profile from modules config section
+    m = [x for x in config['modules'] if x['name'] in p['modules']]
+    import copy
+    modules = copy.deepcopy(m)
+    # generate unique ids for module
+    for module in modules:
+        module['id'] = module['name']
+
+    return modules
+
+def generate_nextflow_script(modules):
+    print('''params.file = "example/proteins.fas"
+Channel.fromPath(param.file).set{fasta}''')
+    target_channels = ["for_"+m['id'] for m in modules]
+    print('fasta.join{'+';'.join(target_channels)+'}')
+
+    from string import Template
+    analysis_template = Template ('''
+process ${id} {
+    input:
+    file fasta from for_${id}
+
+    output:
+    file "$${fasta}.${id}.results" into ${id}_results
+
+    script:
+    """
+    ${analysis_script} --fasta $$fasta --output $${fasta}.${id}.results
+    """
+}
+''')
+    convert_template = Template ('''
+process convert_${id}_to_json {
+    input:
+    file result from ${id}_results
+
+    output:
+    file "$${result}.json" into ${id}_json
+
+    script:
+    """
+    ${converter_script} --result $$result --output $${result}.json
+    """
+}
+''')
+    for m in modules:
+        print(analysis_template.substitute(m))
+        print(convert_template.substitute(m))
 
 def show_analyses():
     print('Profiles:')
@@ -22,8 +96,6 @@ def show_analyses():
     print('Available modules for custom profile:')
     for module in config['modules']:
         print('   {0:<20} - {1}'.format(module['name'], module['info']))
-    exit(0)
-if args.listanalyses: 
-    show_analyses()
 
-        
+if __name__ == "__main__":
+    main()