Skip to content

Commit 92b461a

Browse files
author
Josh Loecker
authored
Consolidate usage of command line arguments (#179)
* Create a new `arguments.py` file to hold standardized arguments across COMO * Use the newly-created `arguments.py` standard * Add arguments from `create_context_speciifc_model.py` * Refactor code to incorporate standardized arguments across COMO * Include `disease_analysis.py` arguments * Refactor code to incorporate standardized arguments across COMO * Add arguments from `knock_out_simulation.py` * Refactor code to incorporate standardized arguments across COMO * Added `merge_xomics.py` arguments * Use standardized `arguments.py` * Added proteomics arguments * Use standardized `arguments.py` * Added `rnaseq_gen.py` arguments * Use standardized `arguments.py` file * Added `rnaseq_preproces.py` arguments * Use new `arguments.py` standard * Fixed imports based on unit test results * Fix fast bioservices usage * Use multi_bioservices for this branch for now becuase we have not merged the related pull request yet * Fix usage of multi/fast bioservices * Fix multi_bioservices usage
1 parent b3edc3c commit 92b461a

12 files changed

Lines changed: 1685 additions & 1439 deletions

‎main/src/GSEpipelineFast.py‎

Lines changed: 104 additions & 87 deletions
Large diffs are not rendered by default.

‎main/src/arguments.py‎

Lines changed: 565 additions & 0 deletions
Large diffs are not rendered by default.

‎main/src/cluster_rnaseq.py‎

Lines changed: 59 additions & 141 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,31 @@
11
# !/usr/bin/python3
22

3-
import os
4-
import sys
53
# from rpy2.robjects.packages import importr
64
# from rpy2.robjects import pandas2ri
75
import argparse
8-
import numpy as np
6+
import os
7+
import sys
98
from pathlib import Path
109

10+
import numpy as np
1111
import rpy2_api
12-
from project import Configs
12+
from arguments import (
13+
batch_ratio_arg,
14+
cluster_algorithm_arg,
15+
context_names_arg,
16+
filtering_technique_arg,
17+
label_arg,
18+
min_count_arg,
19+
min_dist_arg,
20+
num_neighbors_batch_arg,
21+
num_neighbors_context_arg,
22+
num_neighbors_replicate_arg,
23+
quantile_arg,
24+
random_seed_arg,
25+
replicate_ratio_arg,
26+
)
1327
from como_utilities import stringlist_to_list
28+
from project import Configs
1429

1530
# enable r to py conversion
1631
# pandas2ri.activate()
@@ -36,133 +51,30 @@ def main() -> None:
3651
"""
3752
Cluster RNA-seq Data
3853
"""
39-
54+
4055
parser = argparse.ArgumentParser(
4156
prog="cluster_rnaseq.py",
4257
description="Cluster RNA-seq Data using Multiple Correspondence Analysis or UMAP. Clusters at the replicate, "
43-
"batch/study, and context levels.",
58+
"batch/study, and context levels.",
4459
epilog="For additional help, please post questions/issues in the MADRID GitHub repo at "
45-
"https://github.com/HelikarLab/MADRID or email babessell@gmail.com",
46-
)
47-
parser.add_argument(
48-
"-n",
49-
"--context-names",
50-
type=str,
51-
required=True,
52-
dest="context_names",
53-
help="""Tissue/cell name of models to generate. If making multiple models in a batch, then
54-
use the format: \"['context1', 'context2', ... etc]\". Note the outer double-quotes and the
55-
inner single-quotes are required to be interpreted. This a string, not a python list""",
56-
)
57-
parser.add_argument(
58-
"-t",
59-
"--filt-technique",
60-
type=str,
61-
required=True,
62-
dest="technique",
63-
help="'zfpkm', 'quantile', or 'cpm'",
64-
)
65-
parser.add_argument(
66-
"-a",
67-
"--cluster-algorithm",
68-
type=str,
69-
required=False,
70-
default="umap",
71-
dest="clust_algo",
72-
help="""Clustering algorithm to use. 'mca' or 'umap'.""",
73-
)
74-
parser.add_argument(
75-
"-l",
76-
"--label",
77-
type=str,
78-
required=False,
79-
default=True,
80-
dest="label",
81-
help="""True to label replicate/batch/context names on the plots. May be ugly for large sets""",
82-
)
83-
parser.add_argument(
84-
"-d",
85-
"--min-dist",
86-
type=float,
87-
required=False,
88-
default=0.01,
89-
dest="min_dist",
90-
help="""Minimum distance for UMAP clustering. Must be between 0 and 1""",
91-
)
92-
parser.add_argument(
93-
"-r",
94-
"--replicate-ratio",
95-
type=str,
96-
required=False,
97-
default=0.9,
98-
dest="rep_ratio",
99-
help="""Ratio of genes active in replicates for a batch/study to be active""",
100-
)
101-
parser.add_argument(
102-
"-b",
103-
"--batch-ratio",
104-
type=str or float,
105-
required=False,
106-
default=0.9,
107-
dest="batch_ratio",
108-
help="""Ratio of genes active in a batch/study to be active in the context""",
109-
)
110-
parser.add_argument(
111-
"-nr",
112-
"--n-neighbors-rep",
113-
type=str or float,
114-
required=False,
115-
default="default",
116-
dest="n_neigh_rep",
117-
help="""N nearest neighbors for replicate clustering, 'default' is total number of replicates""",
118-
)
119-
parser.add_argument(
120-
"-nb",
121-
"--n-neighbors-batch",
122-
type=str or float,
123-
required=False,
124-
default="default",
125-
dest="n_neigh_batch",
126-
help="""N nearest neighbors for batch clustering, 'default' is total number of batches""",
127-
)
128-
parser.add_argument(
129-
"-nc",
130-
"--n-neighbors-context",
131-
type=str or float,
132-
required=False,
133-
default="default",
134-
dest="n_neigh_cont",
135-
help="""N nearest neighbors for context clustering, 'default' is total number of contexts""",
136-
)
137-
parser.add_argument(
138-
"-c",
139-
"--min-count",
140-
type=str or int,
141-
required=False,
142-
default="default",
143-
dest="min_count",
144-
help="""Ratio of active genes in a batch/study to be active in the context""",
145-
)
146-
parser.add_argument(
147-
"-q",
148-
"--quantile",
149-
type=str or int,
150-
required=False,
151-
default=0.5,
152-
dest="quantile",
153-
help="""Ratio of active genes in a batch/study to be active in the context""",
154-
)
155-
parser.add_argument(
156-
"-s",
157-
"--seed",
158-
type=int,
159-
required=False,
160-
default=-1,
161-
dest="seed",
162-
help="""Random seed for clustering algorithm initialization""",
60+
"https://github.com/HelikarLab/MADRID or email babessell@gmail.com",
16361
)
62+
63+
parser.add_argument(**context_names_arg)
64+
parser.add_argument(**filtering_technique_arg)
65+
parser.add_argument(**cluster_algorithm_arg)
66+
parser.add_argument(**label_arg)
67+
parser.add_argument(**min_dist_arg)
68+
parser.add_argument(**replicate_ratio_arg)
69+
parser.add_argument(**batch_ratio_arg)
70+
parser.add_argument(**num_neighbors_replicate_arg)
71+
parser.add_argument(**num_neighbors_batch_arg)
72+
parser.add_argument(**num_neighbors_context_arg)
73+
parser.add_argument(**min_count_arg)
74+
parser.add_argument(**quantile_arg)
75+
parser.add_argument(**random_seed_arg)
16476
args = parser.parse_args()
165-
77+
16678
wd = os.path.join(configs.data_dir, "results")
16779
context_names = stringlist_to_list(args.context_names)
16880
technique = args.technique.lower()
@@ -176,57 +88,63 @@ def main() -> None:
17688
n_neigh_rep = args.n_neigh_rep
17789
n_neigh_batch = args.n_neigh_batch
17890
n_neigh_cont = args.n_neigh_cont
179-
91+
18092
# Set a random seed if none provided
18193
if int(args.seed) == -1:
18294
seed = np.random.randint(0, 100000)
18395
else:
18496
seed = args.seed
185-
97+
18698
if isinstance(min_count, str) and min_count.lower() == "default":
18799
try:
188100
min_count = int(min_count)
189101
except ValueError:
190102
raise ValueError("--min-count must be either 'default' or an integer > 0")
191103
if not isinstance(min_count, str) and min_count < 0:
192104
raise ValueError("--min-count must be either 'default' or an integer > 0")
193-
105+
194106
if isinstance(quantile, str) and not quantile.lower() == "default":
195107
try:
196108
quantile = int(quantile)
197109
except ValueError:
198-
raise ValueError("--quantile must be either 'default' or an integer between 0 and 100")
110+
raise ValueError(
111+
"--quantile must be either 'default' or an integer between 0 and 100"
112+
)
199113
if not isinstance(quantile, str) and 0 > quantile > 100:
200-
raise ValueError("--quantile must be either 'default' or an integer between 0 and 100")
201-
114+
raise ValueError(
115+
"--quantile must be either 'default' or an integer between 0 and 100"
116+
)
117+
202118
if isinstance(rep_ratio, str) and not rep_ratio.lower() == "default":
203119
try:
204120
rep_ratio = float(rep_ratio)
205121
except ValueError:
206122
raise ValueError("--rep-ratio must be 'default' or a float between 0 and 1")
207123
if not isinstance(rep_ratio, str) and 0 > rep_ratio > 1.0:
208124
raise ValueError("--rep-ratio must be 'default' or a float between 0 and 1")
209-
125+
210126
if isinstance(batch_ratio, str) and not batch_ratio.lower() == "default":
211127
try:
212128
batch_ratio = float(batch_ratio)
213129
except ValueError:
214-
raise ValueError("--batch-ratio must be 'default' or a float between 0 and 1")
130+
raise ValueError(
131+
"--batch-ratio must be 'default' or a float between 0 and 1"
132+
)
215133
if not isinstance(batch_ratio, str) and 0 > batch_ratio > 1.0:
216134
raise ValueError("--batch-ratio must be 'default' or a float between 0 and 1")
217-
135+
218136
if technique.lower() not in ["quantile", "tpm", "cpm", "zfpkm"]:
219137
raise ValueError("--technique must be either 'quantile', 'tpm', 'cpm', 'zfpkm'")
220-
138+
221139
if technique.lower() == "tpm":
222140
technique = "quantile"
223-
141+
224142
if clust_algo.lower() not in ["mca", "umap"]:
225143
raise ValueError("--clust_algo must be either 'mca', 'umap'")
226-
144+
227145
if not isinstance(min_dist, str) and 0 > min_dist > 1.0:
228146
raise ValueError("--min_dist must be a float between 0 and 1")
229-
147+
230148
if isinstance(n_neigh_rep, str) and not n_neigh_rep.lower() == "default":
231149
try:
232150
n_neigh_rep = int(n_neigh_rep)
@@ -240,7 +158,7 @@ def main() -> None:
240158
f"--n_neigh_rep must be either 'default' or an integer greater than 1 and less than or equal to "
241159
f"the total number of replicates being clustered across all contexts."
242160
)
243-
161+
244162
if isinstance(n_neigh_batch, str) and not n_neigh_batch.lower() == "default":
245163
try:
246164
n_neigh_batch = int(n_neigh_batch)
@@ -254,7 +172,7 @@ def main() -> None:
254172
f"--n_neigh_batch must be either 'default' or an integer greater than 1 and less than or equal to "
255173
f"the total number of batches being clustered across all contexts."
256174
)
257-
175+
258176
if isinstance(n_neigh_cont, str) and not n_neigh_cont.lower() == "default":
259177
try:
260178
n_neigh_cont = int(n_neigh_cont)
@@ -268,7 +186,7 @@ def main() -> None:
268186
f"--n_neigh_context must be either 'default' or an integer greater than 1 and less than or equal to "
269187
f"the total number of contexts being clustered."
270188
)
271-
189+
272190
cluster_samples = rpy2_api.Rpy2(
273191
r_file_path=r_file_path,
274192
wd=wd,

0 commit comments

Comments
 (0)