Add feature: dbxref lookup (closes #23)

Sem-Ver: feature

psot/main.py

@@ -24,6 +24,7 @@ def main():
     analyze_parser.add_argument('--profile', '-p', default='fast', help='The profile to use')
     analyze_parser.add_argument('--live', '-l', action='store_true', help='Report results as they are computed, not only at the end of the computation. The live results will be available in the $output/live.')
     analyze_parser.add_argument('--config', '-c', help='The config to use')
+    analyze_parser.add_argument('--fetch_informations', '-i', action='store_true', help='Fetch informations')
     analyze_parser.add_argument('--debug', '-d', action='store_true', help='Debug mode, computation directory will not be removed after computation')
     analyze_parser.add_argument('--execution_dir', '-e', help='Use the specified execution directory and do not delete it after the computation')
     analyze_parser.add_argument('--use_cluster', '-C', action='store_true', help='Use compute cluster for execution')
@@ -59,6 +60,7 @@ def generate_execution(config, args):
     execution = {}
     execution['debug'] = args.debug
     execution['use_cluster'] = args.use_cluster
+    execution['fetch_informations'] = args.fetch_informations 
     execution['mode'] = 'live' if args.live else 'complete'
     execution['bin_path'] = config['app']['bin_path']
     execution['script_path'] = config['app']['script_path']

psot/nextflow.py

@@ -62,6 +62,20 @@ process convert_${id}_to_json {
     """
 }
 ''')
+convert_info_template = Template ('''
+process convert_${id}_to_json {
+    input:
+    file result from ${id}_results
+
+    output:
+    file "$${result}.json" into ${id}_json_info
+
+    script:
+    """
+    ${converter_script} --result $$result --output $${result}.json ${converter_params}
+    """
+}
+''')
 convert_template = Template ('''
 process convert_${id}_to_json {
     input:
@@ -76,18 +90,33 @@ process convert_${id}_to_json {
     """
 }
 ''')
-fetch_template = Template('''
-process fetch_dbxrefs {
+retrieve_informations_template = Template('''
+process retrieve_informations_for_${id} {
 
     input:
-    file result from ${id}_fetch_json
+    file result from ${id}_json_info
 
     output:
-    file "$${result}_fetched.json" into ${id}_json
+    file "$${result.simpleName}_info.json" into ${id}_json
 
     script:
     """
-    resolve_dbxrefs.py --input $$result --retreiver ../dbxref/scripts --output $${result}_fetched.json
+    resolve_dbxrefs.py --input $$result --output $${result.simpleName}_info.json
+    """
+}
+''')
+retrieve_informations_live_template = Template('''
+process retrieve_informations_for_${id} {
+
+    input:
+    file result from ${id}_json_info
+
+    output:
+    file "$${result.simpleName}_info.json" into ${id}_json, ${id}_json_live
+
+    script:
+    """
+    resolve_dbxrefs.py --input $$result --output $${result.simpleName}_info.json
     """
 }
 ''')
@@ -168,11 +197,20 @@ Channel.fromPath(params.fasta).set{fasta}''')
             config['clusterOptions'] = ''
 
         fragments.append(analysis_template.substitute(config))
-        if execution['mode'] == 'live':
+        if execution['mode'] == 'live' and not execution['fetch_informations']:
             fragments.append(convert_live_template.substitute(flatten(m)))
             copy = deepcopy(m)
             copy['output'] = execution['output']
             fragments.append(live_results_template.substitute(flatten(copy)))
+        elif execution['mode'] == 'live' and execution['fetch_informations']:
+            fragments.append(convert_info_template.substitute(flatten(m)))
+            fragments.append(retrieve_informations_live_template.substitute(flatten(m)))
+            copy = deepcopy(m)
+            copy['output'] = execution['output']
+            fragments.append(live_results_template.substitute(flatten(copy)))
+        elif execution['mode'] == 'complete' and execution['fetch_informations']:
+          fragments.append(convert_info_template.substitute(flatten(m)))
+          fragments.append(retrieve_informations_template.substitute(flatten(m)))
         else:
             fragments.append(convert_template.substitute(flatten(m)))
 

tests/test_fetch_informations.py

+import unittest
+import pkg_resources
+from tempfile import mkdtemp
+from shutil import rmtree
+from subprocess import run
+import os.path
+import json
+
+class TestFetchInformations(unittest.TestCase):
+
+    def test_that_psot_runs_with_fetch_informations(self):
+      self.run_psot(['-i'])
+
+    def test_that_psot_runs_with_fetch_informations_and_live_results(self):
+      self.run_psot(['-i', '-l'])
+
+    def run_psot(self, parameters):
+        testdata = pkg_resources.resource_filename('tests', 'data/single.fas')
+        result_directory = mkdtemp()
+        execution = ["psot", 'analyze', '-f', testdata, '-o', result_directory, '-p', 'complete']
+        execution.extend(parameters)
+        proc = run(execution)
+        # first check if the result file exists
+        expected_file = result_directory +'/seq_1.cds_10.json'
+        self.assertTrue(os.path.exists(expected_file))
+        # now check if it is a valid json and contains the id and a non-empty list of results
+        with open(expected_file, 'r') as f:
+          result = json.load(f)
+          self.assertTrue('id' in result)
+          self.assertTrue('computations' in result)
+          self.assertGreater(len(result['computations']), 0)
+        rmtree(result_directory)

tests/test_psot.py

@@ -28,7 +28,5 @@ class TestPsot(unittest.TestCase):
               self.assertTrue('id' in result)
               self.assertTrue('computations' in result)
               self.assertGreater(len(result['computations']), 0)
-
-              
             rmtree(result_directory)