biomed-multi-omic/pyproject.toml at main · BiomedSciAI/biomed-multi-omic · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
# configuration approach followed:
# - whenever possible, prefer pyproject.toml
# - for configurations insufficiently supported by pyproject.toml, use setup.cfg instead
# - setup.py discouraged; minimal stub included only for compatibility with legacy tools

[build-system]
requires = ["setuptools>=45", "setuptools_scm[toml]>=6.2"]
build-backend = "setuptools.build_meta"

[project]
name = "bmfm-targets"
description = "Biomedical foundational models for target discovery"
authors = [
  { name = "Michael M. Danziger", email = "michael.danziger@ibm.com" },
  { name = "Viatcheslav Gurev", email = "vgurev@us.ibm.com" },
  { name = "Bharath Dandala", email = "bdand@us.ibm.com" },
  { name = "Pablo Meyer", email = "pmeyerr@us.ibm.com" },
]
version = "0.20.0"
readme = "README.md"
# due to how PEP 440 defines version matching, prefer [incl, excl) definitions like below:
requires-python = ">=3.10"
dependencies = [
  "scanpy",
  "numpy",
  "scipy",
  "torch",
  "rdata",
  "pytorch-lightning>=2.0.0",
  "transformers>=4.54.0,<5.0.0",
  "omegaconf",
  "hydra-core",
  "clearml>1.13,<2",
  "rich",
  "torchmetrics==1.1.0",
  "tensorboardX",
  "pandas>=2,<3",
  "einops",
  "cellxgene_census[experimental]>1.13.0", #experimental required to use built-in iterable dataset
  "tiledbsoma",
  "pysam",
  "litdata",
  "captum",
  "pydantic",
  "rnanorm",
  "hic-straw",
  "focal-loss-torch @ git+https://github.com/mathiaszinnen/focal_loss_torch.git",
  "torchdata<0.10",
  "peft",
  "huggingface_hub",
  "numexpr",
  "pybigwig",
  "igraph",
]

[project.optional-dependencies]
test = ["pytest", "pytest-cov"]
bulk_rna = ["GEOparse", "biopython"]
benchmarking = [
  "scib",
  "scib-metrics",
  "cz-benchmarks @ git+https://github.com/sivanravidos/cz-benchmarks.git@main"
]
baselines=[
  "pyliger",
  "scanorama",
  "harmony-pytorch",
  "cz-benchmarks",
]
notebook = ["cellxgene-ontology-guide", "jupyter", "swatch"]

[project.urls]
repository = "https://github.com/BiomedSciAI/biomed-multi-omic/"

[project.scripts]
bmfm-targets-run = "bmfm_targets.tasks.scbert.scbert_main:main"

[tool.setuptools.packages]
find = {}

[tool.setuptools.package-data]
bmfm_targets = ["py.typed"]
"bmfm_targets.tests.resources.pretrain.panglao" = [
  "metadata/*.csv",
  "metadata/*.txt",
  "rdata/*.RData",
]


[tool.black]
line-length = 88
skip-string-normalization = false
target-version = ['py310']

[tool.mypy]
check_untyped_defs = true
ignore_missing_imports = true

[[tool.mypy.overrides]]
module = "pytest.*"

[tool.ruff]
target-version = "py310"
extend-include = ["*.ipynb"]

# Activate all the rules that are pyupgrade-related
select = [
  "UP",   # pyupgrade
  "D",    # pydocstyle
  "PT",   # pytest style checking
  "C4",   # comprehensions style checking
  "PD",   # pandas style checking
  "F",    # pyflakes: is-literal
  "W605", # pycodestyle: invalid-escape-sequence
  "I",    # isort
]
# On top of the Google convention, disable `D417`, which requires
# documentation for every function parameter.
ignore = [
  "D100",  # pydocstyle: Missing module docstring
  "D101",  # pydocstyle: Missing module-level docstring
  "D102",  # pydocstyle: Missing docstring in public module
  "D103",  # pydocstyle: Missing class docstring
  "D105",  # pydocstyle: Missing docstring in magic method
  "D107",  # pydocstyle: Missing parameter descriptions in the docstring
  "D203",  # pydocstyle: 1 blank line required before class docstring
  "D205",  # pydocstyle: 1 blank line required between summary line and description
  "D212",  # pydocstyle: Multi-line docstring summary should start at the first line
  "D401",  # pydocstyle: First line should be in imperative mood
  "D417",  # pydocstyle: Missing argument descriptions in the docstring
  "F841",  # flake8: unused variable
  "PD011", # pandas do not use .values (false positives causing bugs in torch code)
  "PD015", # Use .merge method instead of pd.merge function. They have equivalent functionality.
  "PT011", #TODO remove
]
[tool.ruff.per-file-ignores]
"__init__.py" = ["I001"]

[tool.coverage.report]

exclude_lines = ["pragma: no cover", "abc.abstractmethod", "@abstract"]

[tool.coverage.run]
omit = [
  "bmfm_targets/tests/*",
  "bmfm_targets/datasets/bulk_rna/bulk_rna_extraction_utils.py",
  "bmfm_targets/datasets/til/*_util.py",
  "bmfm_targets/datasets/scp1884/*_util.py",
  "bmfm_targets/datasets/scibd300k/scIBD300k_util.py",
  "bmfm_targets/datasets/SNPdb/hic_converter.py",
  "bmfm_targets/datasets/bulk_rna/ua_utils.py",
  "bmfm_targets/datasets/SNPdb/tabix_converter.py",
  "bmfm_targets/datasets/scibd/scIBD_util.py",
  "bmfm_targets/tokenization/create.py",
  "bmfm_targets/datasets/cellxgene/cellxgene_calculate_medians.py",
  "bmfm_targets/datasets/bulk_rna/bulk_rna_download.py",
  "bmfm_targets/datasets/*/get_and_process*",
  "bmfm_targets/datasets/perturbation/vcc/vcc_util.py",
  "bmfm_targets/datasets/perturbation/vcc/perturbation_util.py",
  "bmfm_targets/datasets/sciplex3/cell_drug_response_utils.py",
  "bmfm_targets/datasets/SNPdb/parquet2litdata_token_label.py"
]