Merge pull request #4 from ozolotareva/small_example

Small example
ozolotareva · Sep 9, 2024 · b2ae0c3 · b2ae0c3
2 parents 9701f5d + b8a8530
commit b2ae0c3
Show file tree

Hide file tree

Showing 7 changed files with 212 additions and 83 deletions.
diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
@@ -1,5 +1,5 @@
 # action.yml
-name: 'Run tests'
+name: 'Tests'
 
 on:
   # Triggers the workflow on push or pull request events but only for the "main" branch
@@ -25,6 +25,8 @@ jobs:
           entrypoint: bash
           args: >
             -c "
-            cd /github/workspace &&
-            python test/test_run_unpast.py
+            cd /github/workspace
+            && pip install pytest --target /tmp
+            && PYTHONPATH=/tmp python -m pytest test -m 'not slow'
+            && PYTHONPATH=/tmp python -m pytest test --durations=0
             "
diff --git a/README.md b/README.md
@@ -19,7 +19,7 @@ mkdir -p results
 
 # running UnPaSt with default parameters and example data
 command="python run_unpast.py --exprs test/scenario_B500.exprs.tsv.gz --basename results/scenario_B500"
-docker run -u $(id -u):$(id -g) -v $(pwd):/data --entrypoint bash freddsle/unpast -c "cd /data && $command"
+docker run -u $(id -u):$(id -g) -v "$(pwd)":/data --entrypoint bash freddsle/unpast -c "cd /data && $command"
 ```
 
 

diff --git a/test/test_input/synthetic_clear_biclusters.tsv b/test/test_input/synthetic_clear_biclusters.tsv
@@ -1,51 +1,24 @@
-	sample_0	sample_1	sample_2	sample_3	sample_4	sample_5	sample_6	sample_7	sample_8	sample_9	sample_10	sample_11	sample_12	sample_13	sample_14	sample_15	sample_16	sample_17	sample_18	sample_19
-feature_0	1.042	0.072	1.0	    0.03	1.015	0.009	1.019	0.035	1.04	0.054	1.042	0.069	1.02	0.088	1.003	0.067	1.042	0.056	1.014	0.02
-feature_1	0.08	0.097	0.031	0.069	0.088	0.089	0.009	0.004	0.017	0.088	0.01	0.042	0.096	0.053	0.069	0.032	0.069	0.083	0.002	0.075
-feature_2	1.099	0.075	1.028	0.079	1.01	0.045	1.091	0.029	1.029	0.013	1.002	0.068	1.021	0.027	1.049	0.005	1.057	0.015	1.059	0.07
-feature_3	0.01	0.041	0.069	0.041	0.005	0.054	0.066	0.051	0.094	0.059	0.09	0.014	0.014	0.081	0.04	0.017	0.093	0.035	0.075	0.073
-feature_4	1.088	0.062	1.075	0.035	1.027	0.09	1.043	0.096	1.066	0.062	1.011	0.095	1.045	0.058	1.041	0.024	1.09	0.057	1.0	0.062
-feature_5	0.033	0.053	0.089	0.036	0.091	0.062	0.002	0.093	0.069	0.1	0.017	0.014	0.093	0.07	0.007	0.076	0.075	0.092	0.071	0.012
-feature_6	1.002	0.003	1.003	0.025	1.086	0.054	1.055	0.084	1.012	0.028	1.059	0.097	1.056	0.002	1.08	0.023	1.081	0.039	1.086	0.075
-feature_7	0.056	0.014	0.006	0.012	0.004	0.011	0.023	0.071	0.056	0.001	0.007	0.097	0.057	0.02	0.025	0.074	0.02	0.058	0.097	0.085
-feature_8	1.024	0.049	1.062	0.083	1.016	0.002	1.007	0.049	1.061	0.057	1.032	0.099	1.058	0.038	1.055	0.075	1.067	0.026	1.007	0.037
-feature_9	0.063	0.021	0.075	0.007	0.026	0.08	0.019	0.064	0.052	0.092	0.026	0.007	0.074	0.077	0.091	0.093	0.001	0.023	0.062	0.095
-feature_10	1.095	0.056	1.092	0.064	1.039	0.049	1.06	0.055	1.093	0.092	1.039	0.096	1.017	0.013	1.014	0.051	1.002	0.095	1.083	0.002
-feature_11	0.018	0.033	0.013	0.081	0.034	0.094	0.058	0.088	0.084	0.091	0.046	0.055	0.08	0.029	0.049	0.06	0.002	0.059	0.043	0.081
-feature_12	1.032	0.089	1.058	0.018	1.079	0.061	1.005	0.042	1.068	0.092	1.0	0.098	1.038	0.097	1.06	0.083	1.057	0.063	1.029	0.059
-feature_13	0.075	0.086	0.076	0.07	0.086	0.032	0.067	0.045	0.038	0.041	0.04	0.032	0.062	0.043	0.097	0.068	0.02	0.043	0.034	0.08
-feature_14	1.088	0.09	1.066	0.027	1.025	0.085	1.053	0.08	1.057	0.073	1.052	0.077	1.057	0.047	1.034	0.007	1.038	0.008	1.098	0.018
-feature_15	0.081	0.087	0.069	0.057	0.016	0.047	0.035	0.023	0.059	0.031	0.092	0.091	0.026	0.011	0.019	0.05	0.073	0.021	0.025	0.085
-feature_16	1.042	0.062	1.023	0.01	1.052	0.048	1.015	0.062	1.054	0.065	1.014	0.075	1.022	0.052	1.079	0.002	1.032	0.087	1.084	0.054
-feature_17	0.087	0.095	0.083	0.085	0.01	0.065	0.07	0.061	0.08	0.003	0.077	0.073	0.026	0.026	0.063	0.035	0.08	0.045	0.078	0.099
-feature_18	1.03	0.014	1.09	0.054	1.097	0.064	1.099	0.055	1.053	0.014	1.036	0.003	1.016	0.075	1.003	0.037	1.086	0.069	1.069	0.019
-feature_19	0.044	0.058	0.099	0.02	0.025	0.026	0.075	0.046	0.006	0.051	0.021	0.08	0.03	0.003	0.059	0.084	0.038	0.075	0.051	0.054
-feature_20	1.096	0.08	1.003	0.071	1.047	0.095	1.022	0.027	1.008	0.043	1.011	0.063	1.08	0.07	1.077	0.034	1.085	0.043	1.082	0.063
-feature_21	0.014	0.008	0.002	0.007	0.046	0.011	0.003	0.075	0.039	0.075	0.045	0.045	0.048	0.047	0.08	0.04	0.09	0.004	0.077	0.013
-feature_22	1.062	0.001	1.054	0.0	1.095	0.091	1.08	0.092	1.015	0.016	1.019	0.062	1.091	0.099	1.071	0.073	1.091	0.04	1.025	0.017
-feature_23	0.012	0.081	0.015	0.026	0.082	0.031	0.098	0.027	0.053	0.031	0.091	0.037	0.043	0.051	0.094	0.003	0.072	0.089	0.003	0.052
-feature_24	1.033	0.086	1.056	0.069	1.045	0.063	1.029	0.001	1.058	0.031	1.052	0.092	1.043	0.025	1.037	0.093	1.094	0.084	1.092	0.023
-feature_25	0.009	0.023	0.031	0.017	0.061	0.041	0.082	0.019	0.07	0.024	0.057	0.035	0.006	0.023	0.066	0.05	0.052	0.017	0.057	0.1
-feature_26	1.082	0.059	1.098	0.09	1.06	0.003	1.009	0.007	1.045	0.038	1.098	0.017	1.097	0.077	1.082	0.063	1.067	0.048	1.001	0.035
-feature_27	0.049	0.073	0.047	0.046	0.014	0.001	0.076	0.032	0.098	0.022	0.034	0.052	0.075	0.046	0.012	0.031	0.05	0.067	0.077	0.013
-feature_28	1.002	0.052	1.081	0.001	1.067	0.069	1.045	0.091	1.064	0.001	1.048	0.086	1.083	0.065	1.067	0.058	1.027	0.056	1.067	0.035
-feature_29	0.086	0.02	0.075	0.029	0.077	0.043	0.081	0.035	0.021	0.077	0.031	0.073	0.074	0.022	0.021	0.02	0.014	0.038	0.003	0.011
-feature_30	1.067	0.08	1.008	0.023	1.021	0.092	1.071	0.055	1.03	0.083	1.044	0.092	1.071	0.048	1.013	0.098	1.016	0.02	1.043	0.04
-feature_31	0.015	0.073	0.019	0.064	0.075	0.021	0.06	0.075	0.064	0.06	0.03	0.073	0.095	0.043	0.078	0.006	0.084	0.019	0.04	0.03
-feature_32	1.008	0.09	1.037	0.053	1.049	0.013	1.021	0.008	1.051	0.026	1.036	0.011	1.079	0.011	1.099	0.018	1.057	0.004	1.079	0.019
-feature_33	0.053	0.074	0.015	0.055	0.022	0.076	0.072	0.018	0.086	0.002	0.086	0.056	0.04	0.076	0.072	0.099	0.028	0.0	0.093	0.086
-feature_34	1.073	0.052	1.071	0.078	1.037	0.077	1.075	0.061	1.04	0.07	1.0	0.077	1.09	0.024	1.012	0.022	1.03	0.088	1.054	0.029
-feature_35	0.014	0.029	0.061	0.032	0.046	0.044	0.083	0.043	0.035	0.067	0.022	0.047	0.031	0.063	0.088	0.045	0.078	0.046	0.066	0.013
-feature_36	1.043	0.091	1.061	0.077	1.05	0.05	1.084	0.007	1.057	0.094	1.052	0.019	1.085	0.025	1.07	0.054	1.095	0.062	1.084	0.001
-feature_37	0.099	0.008	0.032	0.095	0.001	0.082	0.086	0.044	0.026	0.08	0.048	0.013	0.093	0.09	0.049	0.086	0.042	0.068	0.04	0.051
-feature_38	1.019	0.096	1.029	0.01	1.014	0.001	1.072	0.056	1.079	0.051	1.079	0.07	1.078	0.041	1.065	0.018	1.032	0.017	1.041	0.024
-feature_39	0.041	0.098	0.032	0.098	0.064	0.038	0.086	0.062	0.025	0.079	0.043	0.036	0.033	0.07	0.027	0.081	0.03	0.054	0.049	0.086
-feature_40	1.089	0.018	1.059	0.09	1.045	0.092	1.028	0.061	1.068	0.023	1.001	0.042	1.094	0.034	1.078	0.017	1.034	0.014	1.072	0.07
-feature_41	0.069	0.025	0.069	0.023	0.042	0.037	0.036	0.006	0.063	0.071	0.061	0.065	0.017	0.015	0.051	0.088	0.018	0.046	0.043	0.05
-feature_42	1.016	0.034	1.026	0.084	1.08	0.043	1.061	0.015	1.051	0.03	1.086	0.067	1.063	0.012	1.047	0.099	1.095	0.065	1.015	0.064
-feature_43	0.057	0.047	0.043	0.06	0.085	0.075	0.058	0.092	0.006	0.099	0.005	0.02	0.042	0.011	0.062	0.005	0.028	0.006	0.07	0.067
-feature_44	1.038	0.019	1.075	0.034	1.08	0.049	1.053	0.003	1.064	0.035	1.023	0.043	1.038	0.047	1.098	0.036	1.077	0.055	1.089	0.035
-feature_45	0.025	0.091	0.004	0.095	0.056	0.038	0.1	0.006	0.052	0.003	0.057	0.018	0.063	0.098	0.087	0.045	0.071	0.078	0.049	0.053
-feature_46	1.015	0.037	1.014	0.073	1.048	0.045	1.089	0.053	1.041	0.027	1.007	0.042	1.003	0.029	1.05	0.097	1.011	0.067	1.05	0.078
-feature_47	0.014	0.008	0.04	0.08	0.019	0.077	0.029	0.022	0.002	0.04	0.038	0.066	0.007	0.015	0.002	0.011	0.065	0.04	0.032	0.056
-feature_48	1.099	0.083	1.07	0.092	1.004	0.007	1.047	0.035	1.094	0.049	1.054	0.09	1.045	0.088	1.025	0.027	1.033	0.055	1.022	0.067
-feature_49	0.014	0.009	0.087	0.024	0.039	0.057	0.053	0.008	0.087	0.095	0.081	0.028	0.053	0.034	0.055	0.097	0.031	0.067	0.033	0.077
+	sample_0	sample_1	sample_2	sample_3	sample_4	sample_5	sample_6	sample_7	sample_8	sample_9	sample_10	sample_11	sample_12	sample_13	sample_14	sample_15	sample_16	sample_17	sample_18	sample_19	sample_20	sample_21	sample_22
+feature_0	0.042	0.072	0.0	0.03	0.015	0.009	0.019	0.035	0.04	0.054	0.042	0.069	0.02	0.088	0.003	0.067	0.042	0.056	0.014	0.02	0.08	0.097	0.031
+feature_1	0.069	1.088	0.089	1.009	0.004	1.017	0.088	1.01	0.042	1.096	0.053	1.069	0.032	1.069	0.083	1.002	0.075	1.099	0.075	1.028	0.079	1.01	0.045
+feature_2	0.091	0.029	0.029	0.013	0.002	0.068	0.021	0.027	0.049	0.005	0.057	0.015	0.059	0.07	0.01	0.041	0.069	0.041	0.005	0.054	0.066	0.051	0.094
+feature_3	0.059	1.09	0.014	1.014	0.081	1.04	0.017	1.093	0.035	1.075	0.073	1.088	0.062	1.075	0.035	1.027	0.09	1.043	0.096	1.066	0.062	1.011	0.095
+feature_4	0.045	0.058	0.041	0.024	0.09	0.057	0.0	0.062	0.033	0.053	0.089	0.036	0.091	0.062	0.002	0.093	0.069	0.1	0.017	0.014	0.093	0.07	0.007
+feature_5	0.076	1.075	0.092	1.071	0.012	1.002	0.003	1.003	0.025	1.086	0.054	1.055	0.084	1.012	0.028	1.059	0.097	1.056	0.002	1.08	0.023	1.081	0.039
+feature_6	0.086	0.075	0.056	0.014	0.006	0.012	0.004	0.011	0.023	0.071	0.056	0.001	0.007	0.097	0.057	0.02	0.025	0.074	0.02	0.058	0.097	0.085	0.024
+feature_7	0.049	1.062	0.083	1.016	0.002	1.007	0.049	1.061	0.057	1.032	0.099	1.058	0.038	1.055	0.075	1.067	0.026	1.007	0.037	1.063	0.021	1.075	0.007
+feature_8	0.026	0.08	0.019	0.064	0.052	0.092	0.026	0.007	0.074	0.077	0.091	0.093	0.001	0.023	0.062	0.095	0.095	0.056	0.092	0.064	0.039	0.049	0.06
+feature_9	0.055	1.093	0.092	1.039	0.096	1.017	0.013	1.014	0.051	1.002	0.095	1.083	0.002	1.018	0.033	1.013	0.081	1.034	0.094	1.058	0.088	1.084	0.091
+feature_10	0.046	0.055	0.08	0.029	0.049	0.06	0.002	0.059	0.043	0.081	0.032	0.089	0.058	0.018	0.079	0.061	0.005	0.042	0.068	0.092	0.0	0.098	0.038
+feature_11	0.097	1.06	0.083	1.057	0.063	1.029	0.059	1.075	0.086	1.076	0.07	1.086	0.032	1.067	0.045	1.038	0.041	1.04	0.032	1.062	0.043	1.097	0.068
+feature_12	0.02	0.043	0.034	0.08	0.088	0.09	0.066	0.027	0.025	0.085	0.053	0.08	0.057	0.073	0.052	0.077	0.057	0.047	0.034	0.007	0.038	0.008	0.098
+feature_13	0.018	1.081	0.087	1.069	0.057	1.016	0.047	1.035	0.023	1.059	0.031	1.092	0.091	1.026	0.011	1.019	0.05	1.073	0.021	1.025	0.085	1.042	0.062
+feature_14	0.023	0.01	0.052	0.048	0.015	0.062	0.054	0.065	0.014	0.075	0.022	0.052	0.079	0.002	0.032	0.087	0.084	0.054	0.087	0.095	0.083	0.085	0.01
+feature_15	0.065	1.07	0.061	1.08	0.003	1.077	0.073	1.026	0.026	1.063	0.035	1.08	0.045	1.078	0.099	1.03	0.014	1.09	0.054	1.097	0.064	1.099	0.055
+feature_16	0.053	0.014	0.036	0.003	0.016	0.075	0.003	0.037	0.086	0.069	0.069	0.019	0.044	0.058	0.099	0.02	0.025	0.026	0.075	0.046	0.006	0.051	0.021
+feature_17	0.08	1.03	0.003	1.059	0.084	1.038	0.075	1.051	0.054	1.096	0.08	1.003	0.071	1.047	0.095	1.022	0.027	1.008	0.043	1.011	0.063	1.08	0.07
+feature_18	0.077	0.034	0.085	0.043	0.082	0.063	0.014	0.008	0.002	0.007	0.046	0.011	0.003	0.075	0.039	0.075	0.045	0.045	0.048	0.047	0.08	0.04	0.09
+feature_19	0.004	1.077	0.013	1.062	0.001	1.054	0.0	1.095	0.091	1.08	0.092	1.015	0.016	1.019	0.062	1.091	0.099	1.071	0.073	1.091	0.04	1.025	0.017
+feature_20	0.012	0.081	0.015	0.026	0.082	0.031	0.098	0.027	0.053	0.031	0.091	0.037	0.043	0.051	0.094	0.003	0.072	0.089	0.003	0.052	0.033	0.086	0.056
+feature_21	0.069	1.045	0.063	1.029	0.001	1.058	0.031	1.052	0.092	1.043	0.025	1.037	0.093	1.094	0.084	1.092	0.023	1.009	0.023	1.031	0.017	1.061	0.041
+feature_22	0.082	0.019	0.07	0.024	0.057	0.035	0.006	0.023	0.066	0.05	0.052	0.017	0.057	0.1	0.082	0.059	0.098	0.09	0.06	0.003	0.009	0.007	0.045
diff --git a/test/test_input/synthetic_small_example.tsv b/test/test_input/synthetic_small_example.tsv
@@ -0,0 +1,6 @@
+	sample_0	sample_1	sample_2	sample_3	sample_4
+feature_0	1.055	1.071	0.029	0.051	0.089
+feature_1	1.09	1.013	0.021	0.005	0.044
+feature_2	0.003	0.046	0.065	0.028	0.068
+feature_3	0.059	0.002	0.056	0.026	0.042
+feature_4	0.028	0.069	0.044	0.016	0.054
diff --git a/test/test_run_unpast.py b/test/test_run_unpast.py
@@ -1,73 +1,106 @@
+"""Tests for run_unpast, and hence all the core code. Usage: python -m pytest test/test_run_unpast.py"""
 import os
 import sys
 import pandas as pd
+import pytest
 
 TEST_DIR = os.path.dirname(__file__)
 RESULTS_DIR = os.path.join(TEST_DIR, "results")
 if not os.access(RESULTS_DIR, os.W_OK):
-    # repo dir is currently read-only during github-actions testing
-    RESULTS_DIR = '/tmp/unpast/results'
+    # repo dir is currently read-only during the tesging stage in github-action
+    RESULTS_DIR = "/tmp/unpast/results"
 REFERENCE_OUTPUT_DIR = os.path.join(TEST_DIR, "test_reference_output")
 
-sys.path.append(os.path.join(TEST_DIR, ".."))
+# sys.path.append(os.path.join(TEST_DIR, ".."))
 from run_unpast import run
 
+
 ### Helper functions ###
 
-def run_unpast_on_file(filename, basename):
+
+def run_unpast_on_file(filename, basename, *args, **kwargs):
     run(
         os.path.join(TEST_DIR, filename),
         out_dir=RESULTS_DIR,
         basename=basename,
+        *args,
+        **kwargs,
     )
     return parse_answer(RESULTS_DIR, basename)
 
 
 def parse_answer(answer_dir, basename):
     files = os.listdir(answer_dir)
-    answer_files = [f for f in files if f.startswith(basename) and f.endswith("biclusters.tsv")]
+    answer_files = [
+        f for f in files if f.startswith(basename) and f.endswith("biclusters.tsv")
+    ]
     assert len(answer_files) == 1, f"There are {len(answer_files)} files instead of 1"
-    return pd.read_csv(os.path.join(answer_dir, answer_files[0]), sep="\t", comment='#')
+    return pd.read_csv(os.path.join(answer_dir, answer_files[0]), sep="\t", comment="#")
+
+
+def parse_to_features_samples_ids(answer):
+    def to_set_of_nums(s):
+        return set(map(int, s.strip().split()))
+
+    return (
+        to_set_of_nums(answer["gene_indexes"]),
+        to_set_of_nums(answer["sample_indexes"]),
+    )
+
 
 ### Tests ###
 
+
+@pytest.mark.slow
 def test_smoke():
     """Smoke test - check that the program runs on some input without failure."""
     run_unpast_on_file(
-        filename = "test_input/synthetic_clear_biclusters.tsv",
-        basename = "test_smoke",
+        filename="test_input/synthetic_clear_biclusters.tsv",
+        basename="test_smoke",
     )
 
 
+@pytest.mark.slow
+def test_simple():
+    """Check that clear biclusters are found."""
+    res = run_unpast_on_file(
+        filename="test_input/synthetic_small_example.tsv",
+        basename="test_simple",
+        min_n_samples=2,
+        clust_method="Louvain",
+    )
+    assert len(res) == 1, "Too many clusters found"
+    features, samples = parse_to_features_samples_ids(res.iloc[0])
+    assert features == {0, 1}
+    assert samples == {0, 1}
+
+
+@pytest.mark.slow
 def test_clear_biclusters():
     """Check that clear biclusters are found."""
     res = run_unpast_on_file(
-        filename = "test_input/synthetic_clear_biclusters.tsv",
-        basename = "test_clear_biclusters",
+        filename="test_input/synthetic_clear_biclusters.tsv",
+        basename="test_clear_biclusters",
     )
-    raise NotImplementedError("TODO: Implement this test")
+
+    found_correct_bicluster = False
+    for _, row in res.iterrows():
+        features, samples = parse_to_features_samples_ids(row)
+        if features == set(range(1, 22, 2)) and samples == set(range(1, 22, 2)):
+            found_correct_bicluster = True
+
+    assert found_correct_bicluster
 
 
+@pytest.mark.slow
 def test_reproducible():
     """Check that the same data is found on a complicated input with no clear answer."""
     res = run_unpast_on_file(
-        filename = "test_input/synthetic_noise.tsv",
-        basename = "test_reproducible",
+        filename="test_input/synthetic_noise.tsv",
+        basename="test_reproducible",
     )
     reference = parse_answer(
-        answer_dir = REFERENCE_OUTPUT_DIR,
-        basename = "test_reproducible",
+        answer_dir=REFERENCE_OUTPUT_DIR,
+        basename="test_reproducible",
     )
     assert res.equals(reference), "The results are not reproducible"
-
-
-if __name__ == "__main__":
-    # run all the tests in this file
-    test_smoke()
-    test_reproducible()
-
-    # deselected test
-    # test_clear_biclusters()
-
-    # TODO: use pytest instead
-    # pytest.main([f"{os.path.abspath(__file__)}"])
diff --git a/test/utils/test_method.py b/test/utils/test_method.py
@@ -0,0 +1,98 @@
+import pandas as pd
+import numpy as np
+from utils.method import zscore, prepare_input_matrix, get_trend
+
+
+def test_get_trend_single_point():
+    sizes = [10]
+    thresholds = [2.5]
+    min_snr = get_trend(sizes, thresholds, plot=False, verbose=False)
+    assert min_snr(10) == 2.5
+
+
+def test_get_trend_multiple_points():
+    sizes = [10, 20, 30, 40, 50]
+    thresholds = [2.5, 3.0, 3.5, 4.0, 4.5]
+    min_snr = get_trend(sizes, thresholds, plot=False, verbose=False)
+    assert np.allclose(min_snr(sizes), thresholds)
+
+
+def test_get_trend_noisy():
+    sizes = [1] * 100 + [2] * 100
+    thresholds = np.linspace(0, 1, 200)
+    min_snr = get_trend(sizes, thresholds, plot=False, verbose=False)
+    # 0.1 tolerance, as set of used points may be not very big
+    assert np.allclose(min_snr([1, 1.5, 2]), [0.25, 0.5, 0.75], atol=0.1)
+
+
+# def test_zscore():
+#     # Test case 1: Basic functionality
+#     df = pd.DataFrame({
+#         'A': [1, 2, 3],
+#         'B': [4, 5, 6],
+#         'C': [7, 8, 9]
+#     })
+#     result = zscore(df)
+#     assert result.shape == df.shape
+#     assert np.allclose(result.mean(), 0, atol=1e-7)
+#     assert np.allclose(result.std(), 1, atol=1e-7)
+
+#     # Test case 2: Handling zero variance
+#     df_zero_var = pd.DataFrame({
+#         'A': [1, 1, 1],
+#         'B': [2, 3, 4],
+#         'C': [5, 6, 7]
+#     })
+#     result_zero_var = zscore(df_zero_var)
+#     assert result_zero_var.shape == df_zero_var.shape
+#     assert np.allclose(result_zero_var.loc['A'], 0)
+
+# def test_prepare_input_matrix():
+#     # Test case 1: Basic functionality
+#     df = pd.DataFrame({
+#         'A': [1, 2, 3, 4, 5],
+#         'B': [2, 3, 4, 5, 6],
+#         'C': [3, 4, 5, 6, 7]
+#     })
+#     result = prepare_input_matrix(df)
+#     assert result.shape == df.shape
+#     assert np.allclose(result.mean(), 0, atol=1e-7)
+#     assert np.allclose(result.std(), 1, atol=1e-7)
+
+#     # Test case 2: Handling zero variance
+#     df_zero_var = pd.DataFrame({
+#         'A': [1, 1, 1, 1, 1],
+#         'B': [2, 3, 4, 5, 6],
+#         'C': [3, 4, 5, 6, 7]
+#     })
+#     result_zero_var = prepare_input_matrix(df_zero_var)
+#     assert result_zero_var.shape == (2, 5)  # One row should be dropped
+
+#     # Test case 3: Handling missing values
+#     df_missing = pd.DataFrame({
+#         'A': [1, 2, np.nan, 4, 5],
+#         'B': [2, 3, 4, np.nan, 6],
+#         'C': [3, 4, 5, 6, 7]
+#     })
+#     result_missing = prepare_input_matrix(df_missing, min_n_samples=4)
+#     assert result_missing.shape == (2, 5)  # One row should be dropped
+
+#     # Test case 4: Ceiling
+#     df_ceiling = pd.DataFrame({
+#         'A': [1, 2, 3, 4, 5],
+#         'B': [2, 3, 4, 5, 6],
+#         'C': [3, 4, 5, 6, 7]
+#     })
+#     result_ceiling = prepare_input_matrix(df_ceiling, ceiling=2)
+#     assert result_ceiling.max().max() <= 2
+#     assert result_ceiling.min().min() >= -2
+
+#     # Test case 5: Non-standardized input
+#     df_non_std = pd.DataFrame({
+#         'A': [10, 20, 30, 40, 50],
+#         'B': [20, 30, 40, 50, 60],
+#         'C': [30, 40, 50, 60, 70]
+#     })
+#     result_non_std = prepare_input_matrix(df_non_std)
+#     assert np.allclose(result_non_std.mean(), 0, atol=1e-7)
+#     assert np.allclose(result_non_std.std(), 1, atol=1e-7)
diff --git a/utils/method.py b/utils/method.py
@@ -219,7 +219,24 @@ def generate_null_dist(
 
 
 def get_trend(sizes, thresholds, plot=True, verbose=True):
-    # smoothens the trend and retunrs a function min_SNR(size; p-val. cutoff)
+    """
+    Smoothens the trend and retunrs a function min_SNR(size; p-val. cutoff)
+        Given a set of points (x_i, y_i), 
+        returns a function f(x) that interpolates the data with LOWESS+linear interpolation
+
+    Args:
+        sizes: values of x_i
+        thresholds: values of y_i
+        plot: if True, plots the trend
+        verbose: if True, prints the LOWESS frac
+    
+    Returns:
+        get_min_snr: a function that returns the minimal SNR for a given size
+    """
+    assert len(sizes) >= 0
+    if len(sizes) == 1: 
+        return lambda x: thresholds[0]
+
     lowess = sm.nonparametric.lowess
     frac = max(1, min(math.floor(int(0.1 * len(sizes))), 15) / len(sizes))
     # if verbose: