diff --git a/workflow/scripts/pool_and_psuedoreplicate.py b/workflow/scripts/pool_and_psuedoreplicate.py
index 8f313010ffa7778f5b4191f9da91c06d5ab74640..6c37eed940ca89f71ebd3c49a2dae042f94d5b05 100644
--- a/workflow/scripts/pool_and_psuedoreplicate.py
+++ b/workflow/scripts/pool_and_psuedoreplicate.py
@@ -204,77 +204,32 @@ def generate_design(paired, cutoff_ratio, design_df, cwd, no_reps, no_unique_con
pool_control_tmp = bedpe_to_tagalign(pool_control, "pool_control")
pool_control = pool_control_tmp
- if not replicated:
-
- # Duplicate rows and update for pool and psuedoreplicates and update tagAlign with single end data
- experiment_id = design_df.at[0, 'experiment_id']
- replicate = design_df.at[0, 'replicate']
- design_new_df = design_df.loc[np.repeat(design_df.index, 4)].reset_index()
- pool_experiment = design_df.tag_align.unique()
-
- # Update tagAlign with single end data
- if paired:
- design_new_df['tag_align'] = design_new_df['se_tag_align']
- design_new_df.drop(labels='se_tag_align', axis=1, inplace=True)
-
- design_new_df['replicate'] = design_new_df['replicate'].astype(str)
- design_new_df.at[1, 'sample_id'] = experiment_id + '_pr1'
- design_new_df.at[1, 'replicate'] = '1_pr'
- design_new_df.at[1, 'xcor'] = 'Calculate'
- design_new_df.at[2, 'sample_id'] = experiment_id + '_pr2'
- design_new_df.at[2, 'replicate'] = '2_pr'
- design_new_df.at[2, 'xcor'] = 'Calculate'
- design_new_df.at[3, 'sample_id'] = experiment_id + '_pooled'
- design_new_df.at[3, 'replicate'] = 'pooled'
- design_new_df.at[3, 'xcor'] = 'Calculate'
- design_new_df.at[3, 'tag_align'] = design_new_df.at[0, 'tag_align']
-
- # Make 2 self psuedoreplicates
- self_pseudoreplicates_dict = {}
- for rep, tag_file in zip(design_df['replicate'], design_df['tag_align']):
- replicate_prefix = experiment_id + '_' + str(rep)
- self_pseudoreplicates_dict[rep] = self_psuedoreplication(tag_file, replicate_prefix, paired)
-
-
- # Update design to include new self pseudo replicates
- self_pseudoreplicates_df = pd.DataFrame.from_dict(self_pseudoreplicates_dict)
- pool_pseudoreplicates_dict = {}
- for rep, pseudorep_file in self_pseudoreplicates_df.iterrows():
- path_to_file = cwd + '/' + pseudorep_file
- replicate = rep + 1
- design_new_df.loc[replicate, 'tag_align'] = path_to_file
- pool_pseudoreplicates_dict[replicate] = path_to_file
-
- # Drop index column
- design_new_df.drop(labels='index', axis=1, inplace=True)
-
- else:
- # Make pool of replicates
- replicate_files = design_df.tag_align.unique()
- experiment_id = design_df.at[0, 'experiment_id']
- pool_experiment = pool(replicate_files, experiment_id + "_pooled", paired)
-
- # Make self psuedoreplicates equivalent to number of replicates
- pseudoreplicates_dict = {}
- for rep, tag_file in zip(design_df['replicate'], design_df['tag_align']):
- replicate_prefix = experiment_id + '_' + str(rep)
- pr_dict = self_psuedoreplication(tag_file, replicate_prefix, paired)
- pseudoreplicates_dict[rep] = pr_dict
-
- # Merge self psuedoreplication for each true replicate
- pseudoreplicates_df = pd.DataFrame.from_dict(pseudoreplicates_dict)
- pool_pseudoreplicates_dict = {}
- for index, row in pseudoreplicates_df.iterrows():
- replicate_id = index + 1
- pr_filename = experiment_id + ".pr" + str(replicate_id) + '.tagAlign.gz'
- pool_replicate = pool(row, pr_filename, False)
- pool_pseudoreplicates_dict[replicate_id] = pool_replicate
-
- design_new_df = design_df
- # Update tagAlign with single end data
- if paired:
- design_new_df['tag_align'] = design_new_df['se_tag_align']
- design_new_df.drop(labels='se_tag_align', axis=1, inplace=True)
+ # Duplicate rows and update for pool and psuedoreplicates and update tagAlign with single end data
+ experiment_id = design_df.at[0, 'experiment_id']
+ replicate_files = design_df.tag_align.unique()
+ pool_experiment = pool(replicate_files, experiment_id + "_pooled", paired)
+
+ # Make 2 self psuedoreplicates
+ pseudoreplicates_dict = {}
+ for rep, tag_file in zip(design_df['replicate'], design_df['tag_align']):
+ replicate_prefix = experiment_id + '_' + str(rep)
+ pr_dict = self_psuedoreplication(tag_file, replicate_prefix, paired)
+ pseudoreplicates_dict[rep] = pr_dict
+
+ # Update design to include new self pseudo replicates
+ pseudoreplicates_df = pd.DataFrame.from_dict(pseudoreplicates_dict)
+ pool_pseudoreplicates_dict = {}
+ for index, row in pseudoreplicates_df.iterrows():
+ replicate_id = index + 1
+ pr_filename = experiment_id + ".pr" + str(replicate_id) + '.tagAlign.gz'
+ pool_replicate = pool(row, pr_filename, False)
+ pool_pseudoreplicates_dict[replicate_id] = pool_replicate
+
+ design_new_df = design_df #.loc[np.repeat(design_df.index, 4)].reset_index()
+ # Update tagAlign with single end data
+ if paired:
+ design_new_df['tag_align'] = design_new_df['se_tag_align']
+ design_new_df.drop(labels='se_tag_align', axis=1, inplace=True)
# If paired change to single End
if paired:
diff --git a/workflow/tests/test_pool_and_psuedoreplicate.py b/workflow/tests/test_pool_and_psuedoreplicate.py
index c780122ed95c4bb720852976336054c538f11e83..f251e888f906de1c7e360502cadf45c8c9b85a3c 100644
--- a/workflow/tests/test_pool_and_psuedoreplicate.py
+++ b/workflow/tests/test_pool_and_psuedoreplicate.py
@@ -70,7 +70,6 @@ def test_single_rep(design_experiment_2):
shutil.copy(test_design_path + 'A_1.tagAlign.gz', cwd)
shutil.copy(test_design_path + 'B_1.tagAlign.gz', cwd)
single_rep = pool_and_psuedoreplicate.generate_design('false', 1.2, design_experiment_2, cwd, 1, 1)
- print(single_rep)
assert single_rep.shape[0] == 4
@pytest.mark.singleend