pepdata 1.2.4__tar.gz → 1.3.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- pepdata-1.3.0/PKG-INFO +102 -0
- pepdata-1.3.0/README.md +75 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/alleles.py +1 -1
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/mhc.py +1 -3
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/peptide_vectorizer.py +34 -3
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/pmbec.py +1 -1
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/version.py +1 -1
- pepdata-1.3.0/pepdata.egg-info/PKG-INFO +102 -0
- pepdata-1.3.0/pepdata.egg-info/requires.txt +4 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pyproject.toml +3 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/requirements.txt +1 -5
- pepdata-1.3.0/tests/test_blosum.py +68 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/tests/test_iedb_alleles.py +20 -1
- {pepdata-1.2.4 → pepdata-1.3.0}/tests/test_iedb_mhc.py +19 -0
- pepdata-1.3.0/tests/test_ngram.py +119 -0
- pepdata-1.3.0/tests/test_pmbec.py +31 -0
- pepdata-1.2.4/PKG-INFO +0 -86
- pepdata-1.2.4/README.md +0 -55
- pepdata-1.2.4/pepdata.egg-info/PKG-INFO +0 -86
- pepdata-1.2.4/pepdata.egg-info/requires.txt +0 -8
- pepdata-1.2.4/tests/test_blosum.py +0 -17
- pepdata-1.2.4/tests/test_ngram.py +0 -61
- pepdata-1.2.4/tests/test_pmbec.py +0 -7
- {pepdata-1.2.4 → pepdata-1.3.0}/LICENSE.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/MANIFEST.in +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/__init__.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/amino_acid.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/amino_acid_alphabet.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/amino_acid_properties.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/blosum.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/chou_fasman.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/common.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/__init__.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/columns.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/common.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/memoize.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/iedb/tcell.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/BLOSUM30 +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/BLOSUM50 +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/BLOSUM62 +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/__init__.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/amino_acid_properties.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/helix_vs_coil.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/helix_vs_strand.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/pmbec.mat +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/matrices/strand_vs_coil.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/reduced_alphabet.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/residue_contact_energies.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata/static_data.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata.egg-info/SOURCES.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata.egg-info/dependency_links.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/pepdata.egg-info/top_level.txt +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/scripts/check_docs_examples.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/scripts/check_docs_links.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/setup.cfg +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/tests/conftest.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/tests/test_amino_acids.py +0 -0
- {pepdata-1.2.4 → pepdata-1.3.0}/tests/test_iedb_tcell.py +0 -0
pepdata-1.3.0/PKG-INFO
ADDED
|
@@ -0,0 +1,102 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: pepdata
|
|
3
|
+
Version: 1.3.0
|
|
4
|
+
Summary: Immunological peptide datasets and amino acid properties
|
|
5
|
+
Author-email: Alex Rubinsteyn <alex.rubinsteyn@mssm.edu>
|
|
6
|
+
Project-URL: Homepage, https://github.com/openvax/pepdata
|
|
7
|
+
Project-URL: Bug Tracker, https://github.com/openvax/pepdata/issues
|
|
8
|
+
Classifier: Development Status :: 4 - Beta
|
|
9
|
+
Classifier: Environment :: Console
|
|
10
|
+
Classifier: Operating System :: OS Independent
|
|
11
|
+
Classifier: Intended Audience :: Science/Research
|
|
12
|
+
Classifier: License :: OSI Approved :: Apache Software License
|
|
13
|
+
Classifier: Programming Language :: Python
|
|
14
|
+
Classifier: Programming Language :: Python :: 3.11
|
|
15
|
+
Classifier: Programming Language :: Python :: 3.12
|
|
16
|
+
Classifier: Programming Language :: Python :: 3.13
|
|
17
|
+
Classifier: Programming Language :: Python :: 3.14
|
|
18
|
+
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
|
|
19
|
+
Requires-Python: >=3.11
|
|
20
|
+
Description-Content-Type: text/markdown
|
|
21
|
+
License-File: LICENSE.txt
|
|
22
|
+
Requires-Dist: numpy<3.0.0,>=2.0.0
|
|
23
|
+
Requires-Dist: pandas<4,>=3
|
|
24
|
+
Requires-Dist: scikit-learn>=0.14.1
|
|
25
|
+
Requires-Dist: datacache>=0.4.4
|
|
26
|
+
Dynamic: license-file
|
|
27
|
+
|
|
28
|
+
[](https://github.com/openvax/pepdata/actions/workflows/tests.yml)
|
|
29
|
+
[](https://openvax.github.io/pepdata/)
|
|
30
|
+
[](https://pypi.org/project/pepdata/)
|
|
31
|
+
|
|
32
|
+
# pepdata
|
|
33
|
+
|
|
34
|
+
pepdata ships amino-acid property tables and pairwise residue matrices
|
|
35
|
+
(BLOSUM, PMBEC, contact energies), turns peptide sequences into n-gram feature
|
|
36
|
+
vectors, and loads MHC-binding and T-cell assay tables from the
|
|
37
|
+
[IEDB](https://www.iedb.org/). It is the [OpenVax](https://github.com/openvax)
|
|
38
|
+
library for peptide-level reference data. It does not predict MHC binding; for
|
|
39
|
+
that, see [mhctools](https://github.com/openvax/mhctools) and
|
|
40
|
+
[mhcflurry](https://github.com/openvax/mhcflurry).
|
|
41
|
+
|
|
42
|
+
## Install
|
|
43
|
+
|
|
44
|
+
```sh
|
|
45
|
+
python -m pip install pepdata
|
|
46
|
+
```
|
|
47
|
+
|
|
48
|
+
pepdata needs Python 3.11 or later. Property tables and matrices ship with the
|
|
49
|
+
package and work offline. The IEDB tables are downloaded on first use and
|
|
50
|
+
cached by [datacache](https://github.com/openvax/datacache).
|
|
51
|
+
|
|
52
|
+
## Quick start
|
|
53
|
+
|
|
54
|
+
```python
|
|
55
|
+
from pepdata.amino_acid_properties import hydropathy
|
|
56
|
+
|
|
57
|
+
peptide = "ACD"
|
|
58
|
+
print([hydropathy[residue] for residue in peptide])
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
```text
|
|
62
|
+
[1.8, 2.5, -3.5]
|
|
63
|
+
```
|
|
64
|
+
|
|
65
|
+
Each property table is a dictionary from single-letter residue to value. Use
|
|
66
|
+
uppercase canonical residue letters; a missing letter raises `KeyError`.
|
|
67
|
+
|
|
68
|
+
## Documentation
|
|
69
|
+
|
|
70
|
+
The [documentation](https://openvax.github.io/pepdata/) has two guides and an
|
|
71
|
+
API reference:
|
|
72
|
+
|
|
73
|
+
1. [Residue and pairwise tables](https://openvax.github.io/pepdata/guides/properties/):
|
|
74
|
+
property dictionaries, BLOSUM and other matrices, alphabets.
|
|
75
|
+
2. [Peptide features and IEDB tables](https://openvax.github.io/pepdata/guides/datasets/):
|
|
76
|
+
n-gram vectors and the MHC-binding and T-cell assay loaders.
|
|
77
|
+
3. [API reference](https://openvax.github.io/pepdata/reference/).
|
|
78
|
+
|
|
79
|
+
| Task | Module or object |
|
|
80
|
+
| --- | --- |
|
|
81
|
+
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
82
|
+
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
83
|
+
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
84
|
+
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
85
|
+
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
86
|
+
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
87
|
+
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
88
|
+
|
|
89
|
+
## Development
|
|
90
|
+
|
|
91
|
+
```sh
|
|
92
|
+
python -m pip install -e . pytest pytest-cov ruff
|
|
93
|
+
./lint.sh # ruff check
|
|
94
|
+
./test.sh # pytest with coverage
|
|
95
|
+
python -m pip install -r requirements-docs.txt
|
|
96
|
+
./docs.sh # mkdocs build --strict, then a rendered-link check
|
|
97
|
+
python scripts/check_docs_examples.py # run docs examples, compare output
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
## License
|
|
101
|
+
|
|
102
|
+
Apache 2.0; see [LICENSE.txt](https://github.com/openvax/pepdata/blob/master/LICENSE.txt).
|
pepdata-1.3.0/README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
1
|
+
[](https://github.com/openvax/pepdata/actions/workflows/tests.yml)
|
|
2
|
+
[](https://openvax.github.io/pepdata/)
|
|
3
|
+
[](https://pypi.org/project/pepdata/)
|
|
4
|
+
|
|
5
|
+
# pepdata
|
|
6
|
+
|
|
7
|
+
pepdata ships amino-acid property tables and pairwise residue matrices
|
|
8
|
+
(BLOSUM, PMBEC, contact energies), turns peptide sequences into n-gram feature
|
|
9
|
+
vectors, and loads MHC-binding and T-cell assay tables from the
|
|
10
|
+
[IEDB](https://www.iedb.org/). It is the [OpenVax](https://github.com/openvax)
|
|
11
|
+
library for peptide-level reference data. It does not predict MHC binding; for
|
|
12
|
+
that, see [mhctools](https://github.com/openvax/mhctools) and
|
|
13
|
+
[mhcflurry](https://github.com/openvax/mhcflurry).
|
|
14
|
+
|
|
15
|
+
## Install
|
|
16
|
+
|
|
17
|
+
```sh
|
|
18
|
+
python -m pip install pepdata
|
|
19
|
+
```
|
|
20
|
+
|
|
21
|
+
pepdata needs Python 3.11 or later. Property tables and matrices ship with the
|
|
22
|
+
package and work offline. The IEDB tables are downloaded on first use and
|
|
23
|
+
cached by [datacache](https://github.com/openvax/datacache).
|
|
24
|
+
|
|
25
|
+
## Quick start
|
|
26
|
+
|
|
27
|
+
```python
|
|
28
|
+
from pepdata.amino_acid_properties import hydropathy
|
|
29
|
+
|
|
30
|
+
peptide = "ACD"
|
|
31
|
+
print([hydropathy[residue] for residue in peptide])
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
```text
|
|
35
|
+
[1.8, 2.5, -3.5]
|
|
36
|
+
```
|
|
37
|
+
|
|
38
|
+
Each property table is a dictionary from single-letter residue to value. Use
|
|
39
|
+
uppercase canonical residue letters; a missing letter raises `KeyError`.
|
|
40
|
+
|
|
41
|
+
## Documentation
|
|
42
|
+
|
|
43
|
+
The [documentation](https://openvax.github.io/pepdata/) has two guides and an
|
|
44
|
+
API reference:
|
|
45
|
+
|
|
46
|
+
1. [Residue and pairwise tables](https://openvax.github.io/pepdata/guides/properties/):
|
|
47
|
+
property dictionaries, BLOSUM and other matrices, alphabets.
|
|
48
|
+
2. [Peptide features and IEDB tables](https://openvax.github.io/pepdata/guides/datasets/):
|
|
49
|
+
n-gram vectors and the MHC-binding and T-cell assay loaders.
|
|
50
|
+
3. [API reference](https://openvax.github.io/pepdata/reference/).
|
|
51
|
+
|
|
52
|
+
| Task | Module or object |
|
|
53
|
+
| --- | --- |
|
|
54
|
+
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
55
|
+
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
56
|
+
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
57
|
+
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
58
|
+
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
59
|
+
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
60
|
+
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
61
|
+
|
|
62
|
+
## Development
|
|
63
|
+
|
|
64
|
+
```sh
|
|
65
|
+
python -m pip install -e . pytest pytest-cov ruff
|
|
66
|
+
./lint.sh # ruff check
|
|
67
|
+
./test.sh # pytest with coverage
|
|
68
|
+
python -m pip install -r requirements-docs.txt
|
|
69
|
+
./docs.sh # mkdocs build --strict, then a rendered-link check
|
|
70
|
+
python scripts/check_docs_examples.py # run docs examples, compare output
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
## License
|
|
74
|
+
|
|
75
|
+
Apache 2.0; see [LICENSE.txt](https://github.com/openvax/pepdata/blob/master/LICENSE.txt).
|
|
@@ -42,7 +42,7 @@ def local_path(auto_download=True):
|
|
|
42
42
|
return download()
|
|
43
43
|
raise ValueError(
|
|
44
44
|
("MHC data file %s does not exist locally,"
|
|
45
|
-
" call pepdata.mhc.download() to get a copy from IEDB") % path)
|
|
45
|
+
" call pepdata.iedb.mhc.download() to get a copy from IEDB") % path)
|
|
46
46
|
return path
|
|
47
47
|
|
|
48
48
|
def delete():
|
|
@@ -108,8 +108,6 @@ def load_dataframe(
|
|
|
108
108
|
# To deal with this, drop any columns which are all NaN
|
|
109
109
|
df = df.dropna(axis=1, how="all")
|
|
110
110
|
|
|
111
|
-
print(df.head())
|
|
112
|
-
|
|
113
111
|
n = len(df)
|
|
114
112
|
|
|
115
113
|
mhc_group_key = "MHC Restriction"
|
|
@@ -55,16 +55,31 @@ class PeptideVectorizer(object):
|
|
|
55
55
|
}
|
|
56
56
|
|
|
57
57
|
def fit_transform(self, amino_acid_strings):
|
|
58
|
+
"""
|
|
59
|
+
Learn the n-gram vocabulary of the given sequences and vectorize them.
|
|
60
|
+
|
|
61
|
+
Parameters
|
|
62
|
+
----------
|
|
63
|
+
amino_acid_strings : list of str
|
|
64
|
+
Peptide or protein sequences.
|
|
65
|
+
|
|
66
|
+
Returns
|
|
67
|
+
-------
|
|
68
|
+
numpy.ndarray
|
|
69
|
+
Array of shape (len(amino_acid_strings), number of distinct
|
|
70
|
+
n-grams) with n-gram counts, or per-row frequencies (each row
|
|
71
|
+
divided by its total) when normalize_row is True.
|
|
72
|
+
"""
|
|
58
73
|
self.count_vectorizer = \
|
|
59
74
|
make_count_vectorizer(self.reduced_alphabet, self.max_ngram)
|
|
60
75
|
|
|
61
76
|
if self.training_already_reduced:
|
|
62
77
|
c = make_count_vectorizer(None, self.max_ngram)
|
|
63
|
-
X = c.fit_transform(amino_acid_strings).
|
|
78
|
+
X = c.fit_transform(amino_acid_strings).toarray()
|
|
64
79
|
self.count_vectorizer.vocabulary_ = c.vocabulary_
|
|
65
80
|
else:
|
|
66
81
|
c = self.count_vectorizer
|
|
67
|
-
X = c.fit_transform(amino_acid_strings).
|
|
82
|
+
X = c.fit_transform(amino_acid_strings).toarray()
|
|
68
83
|
|
|
69
84
|
if self.normalize_row:
|
|
70
85
|
X = normalize(X, norm='l1')
|
|
@@ -74,8 +89,24 @@ class PeptideVectorizer(object):
|
|
|
74
89
|
self.fit_transform(amino_acid_strings)
|
|
75
90
|
|
|
76
91
|
def transform(self, amino_acid_strings):
|
|
92
|
+
"""
|
|
93
|
+
Vectorize sequences with the vocabulary learned by fit.
|
|
94
|
+
|
|
95
|
+
n-grams that weren't seen during fitting are not counted.
|
|
96
|
+
|
|
97
|
+
Parameters
|
|
98
|
+
----------
|
|
99
|
+
amino_acid_strings : list of str
|
|
100
|
+
Peptide or protein sequences.
|
|
101
|
+
|
|
102
|
+
Returns
|
|
103
|
+
-------
|
|
104
|
+
numpy.ndarray
|
|
105
|
+
Array of shape (len(amino_acid_strings), size of the fitted
|
|
106
|
+
vocabulary), normalized per row as in fit_transform.
|
|
107
|
+
"""
|
|
77
108
|
assert self.count_vectorizer, "Must call 'fit' before 'transform'"
|
|
78
|
-
X = self.count_vectorizer.transform(amino_acid_strings).
|
|
109
|
+
X = self.count_vectorizer.transform(amino_acid_strings).toarray()
|
|
79
110
|
if self.normalize_row:
|
|
80
111
|
X = normalize(X, norm='l1')
|
|
81
112
|
return X
|
|
@@ -83,5 +83,5 @@ def read_pmbec_coefficients(
|
|
|
83
83
|
return d
|
|
84
84
|
|
|
85
85
|
# dictionary of PMBEC coefficient accessed like pmbec_dict["V"]["R"]
|
|
86
|
-
pmbec_dict = read_pmbec_coefficients(key_type="row")
|
|
86
|
+
pmbec_dict = read_pmbec_coefficients(key_type="row", verbose=False)
|
|
87
87
|
pmbec_matrix = dict_to_amino_acid_matrix(pmbec_dict)
|
|
@@ -0,0 +1,102 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: pepdata
|
|
3
|
+
Version: 1.3.0
|
|
4
|
+
Summary: Immunological peptide datasets and amino acid properties
|
|
5
|
+
Author-email: Alex Rubinsteyn <alex.rubinsteyn@mssm.edu>
|
|
6
|
+
Project-URL: Homepage, https://github.com/openvax/pepdata
|
|
7
|
+
Project-URL: Bug Tracker, https://github.com/openvax/pepdata/issues
|
|
8
|
+
Classifier: Development Status :: 4 - Beta
|
|
9
|
+
Classifier: Environment :: Console
|
|
10
|
+
Classifier: Operating System :: OS Independent
|
|
11
|
+
Classifier: Intended Audience :: Science/Research
|
|
12
|
+
Classifier: License :: OSI Approved :: Apache Software License
|
|
13
|
+
Classifier: Programming Language :: Python
|
|
14
|
+
Classifier: Programming Language :: Python :: 3.11
|
|
15
|
+
Classifier: Programming Language :: Python :: 3.12
|
|
16
|
+
Classifier: Programming Language :: Python :: 3.13
|
|
17
|
+
Classifier: Programming Language :: Python :: 3.14
|
|
18
|
+
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
|
|
19
|
+
Requires-Python: >=3.11
|
|
20
|
+
Description-Content-Type: text/markdown
|
|
21
|
+
License-File: LICENSE.txt
|
|
22
|
+
Requires-Dist: numpy<3.0.0,>=2.0.0
|
|
23
|
+
Requires-Dist: pandas<4,>=3
|
|
24
|
+
Requires-Dist: scikit-learn>=0.14.1
|
|
25
|
+
Requires-Dist: datacache>=0.4.4
|
|
26
|
+
Dynamic: license-file
|
|
27
|
+
|
|
28
|
+
[](https://github.com/openvax/pepdata/actions/workflows/tests.yml)
|
|
29
|
+
[](https://openvax.github.io/pepdata/)
|
|
30
|
+
[](https://pypi.org/project/pepdata/)
|
|
31
|
+
|
|
32
|
+
# pepdata
|
|
33
|
+
|
|
34
|
+
pepdata ships amino-acid property tables and pairwise residue matrices
|
|
35
|
+
(BLOSUM, PMBEC, contact energies), turns peptide sequences into n-gram feature
|
|
36
|
+
vectors, and loads MHC-binding and T-cell assay tables from the
|
|
37
|
+
[IEDB](https://www.iedb.org/). It is the [OpenVax](https://github.com/openvax)
|
|
38
|
+
library for peptide-level reference data. It does not predict MHC binding; for
|
|
39
|
+
that, see [mhctools](https://github.com/openvax/mhctools) and
|
|
40
|
+
[mhcflurry](https://github.com/openvax/mhcflurry).
|
|
41
|
+
|
|
42
|
+
## Install
|
|
43
|
+
|
|
44
|
+
```sh
|
|
45
|
+
python -m pip install pepdata
|
|
46
|
+
```
|
|
47
|
+
|
|
48
|
+
pepdata needs Python 3.11 or later. Property tables and matrices ship with the
|
|
49
|
+
package and work offline. The IEDB tables are downloaded on first use and
|
|
50
|
+
cached by [datacache](https://github.com/openvax/datacache).
|
|
51
|
+
|
|
52
|
+
## Quick start
|
|
53
|
+
|
|
54
|
+
```python
|
|
55
|
+
from pepdata.amino_acid_properties import hydropathy
|
|
56
|
+
|
|
57
|
+
peptide = "ACD"
|
|
58
|
+
print([hydropathy[residue] for residue in peptide])
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
```text
|
|
62
|
+
[1.8, 2.5, -3.5]
|
|
63
|
+
```
|
|
64
|
+
|
|
65
|
+
Each property table is a dictionary from single-letter residue to value. Use
|
|
66
|
+
uppercase canonical residue letters; a missing letter raises `KeyError`.
|
|
67
|
+
|
|
68
|
+
## Documentation
|
|
69
|
+
|
|
70
|
+
The [documentation](https://openvax.github.io/pepdata/) has two guides and an
|
|
71
|
+
API reference:
|
|
72
|
+
|
|
73
|
+
1. [Residue and pairwise tables](https://openvax.github.io/pepdata/guides/properties/):
|
|
74
|
+
property dictionaries, BLOSUM and other matrices, alphabets.
|
|
75
|
+
2. [Peptide features and IEDB tables](https://openvax.github.io/pepdata/guides/datasets/):
|
|
76
|
+
n-gram vectors and the MHC-binding and T-cell assay loaders.
|
|
77
|
+
3. [API reference](https://openvax.github.io/pepdata/reference/).
|
|
78
|
+
|
|
79
|
+
| Task | Module or object |
|
|
80
|
+
| --- | --- |
|
|
81
|
+
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
82
|
+
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
83
|
+
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
84
|
+
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
85
|
+
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
86
|
+
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
87
|
+
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
88
|
+
|
|
89
|
+
## Development
|
|
90
|
+
|
|
91
|
+
```sh
|
|
92
|
+
python -m pip install -e . pytest pytest-cov ruff
|
|
93
|
+
./lint.sh # ruff check
|
|
94
|
+
./test.sh # pytest with coverage
|
|
95
|
+
python -m pip install -r requirements-docs.txt
|
|
96
|
+
./docs.sh # mkdocs build --strict, then a rendered-link check
|
|
97
|
+
python scripts/check_docs_examples.py # run docs examples, compare output
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
## License
|
|
101
|
+
|
|
102
|
+
Apache 2.0; see [LICENSE.txt](https://github.com/openvax/pepdata/blob/master/LICENSE.txt).
|
|
@@ -34,6 +34,9 @@ exclude = ["test", "test.*"]
|
|
|
34
34
|
"Homepage" = "https://github.com/openvax/pepdata"
|
|
35
35
|
"Bug Tracker" = "https://github.com/openvax/pepdata/issues"
|
|
36
36
|
|
|
37
|
+
[tool.ruff]
|
|
38
|
+
target-version = "py311"
|
|
39
|
+
|
|
37
40
|
[tool.ruff.lint]
|
|
38
41
|
select = ["E", "F"]
|
|
39
42
|
ignore = ["F821", "E501", "F841", "E731", "E741", "E722", "E721"]
|
|
@@ -0,0 +1,68 @@
|
|
|
1
|
+
import pytest
|
|
2
|
+
|
|
3
|
+
from pepdata.amino_acid_alphabet import canonical_amino_acid_letters
|
|
4
|
+
from pepdata.blosum import (
|
|
5
|
+
blosum30_dict,
|
|
6
|
+
blosum30_matrix,
|
|
7
|
+
blosum50_dict,
|
|
8
|
+
blosum50_matrix,
|
|
9
|
+
blosum62_dict,
|
|
10
|
+
blosum62_matrix,
|
|
11
|
+
)
|
|
12
|
+
|
|
13
|
+
BLOSUM_TABLES = {
|
|
14
|
+
30: (blosum30_dict, blosum30_matrix),
|
|
15
|
+
50: (blosum50_dict, blosum50_matrix),
|
|
16
|
+
62: (blosum62_dict, blosum62_matrix),
|
|
17
|
+
}
|
|
18
|
+
|
|
19
|
+
# Published self-substitution scores for alanine, tryptophan and cysteine.
|
|
20
|
+
DIAGONAL_SCORES = {
|
|
21
|
+
30: {"A": 4, "W": 20, "C": 17},
|
|
22
|
+
50: {"A": 5, "W": 15, "C": 13},
|
|
23
|
+
62: {"A": 4, "W": 11, "C": 9},
|
|
24
|
+
}
|
|
25
|
+
|
|
26
|
+
|
|
27
|
+
@pytest.mark.parametrize("number", sorted(BLOSUM_TABLES))
|
|
28
|
+
def test_blosum_covers_canonical_residues(number):
|
|
29
|
+
table, _ = BLOSUM_TABLES[number]
|
|
30
|
+
for row in canonical_amino_acid_letters:
|
|
31
|
+
assert set(canonical_amino_acid_letters) <= set(table[row])
|
|
32
|
+
|
|
33
|
+
|
|
34
|
+
@pytest.mark.parametrize("number", sorted(BLOSUM_TABLES))
|
|
35
|
+
def test_blosum_diagonal_scores(number):
|
|
36
|
+
table, _ = BLOSUM_TABLES[number]
|
|
37
|
+
for residue, score in DIAGONAL_SCORES[number].items():
|
|
38
|
+
assert table[residue][residue] == score
|
|
39
|
+
for residue in canonical_amino_acid_letters:
|
|
40
|
+
row = [table[residue][other] for other in canonical_amino_acid_letters]
|
|
41
|
+
assert table[residue][residue] == max(row)
|
|
42
|
+
|
|
43
|
+
|
|
44
|
+
@pytest.mark.parametrize("number", [30, 62])
|
|
45
|
+
def test_blosum_is_symmetric(number):
|
|
46
|
+
table, _ = BLOSUM_TABLES[number]
|
|
47
|
+
for x in canonical_amino_acid_letters:
|
|
48
|
+
for y in canonical_amino_acid_letters:
|
|
49
|
+
assert table[x][y] == table[y][x]
|
|
50
|
+
|
|
51
|
+
|
|
52
|
+
def test_blosum50_asymmetry_is_limited_to_shipped_w_t_entry():
|
|
53
|
+
# The shipped BLOSUM50 file has W->T = -4 but T->W = -3 (published value -3).
|
|
54
|
+
asymmetric = {
|
|
55
|
+
(x, y)
|
|
56
|
+
for x in canonical_amino_acid_letters
|
|
57
|
+
for y in canonical_amino_acid_letters
|
|
58
|
+
if blosum50_dict[x][y] != blosum50_dict[y][x]
|
|
59
|
+
}
|
|
60
|
+
assert asymmetric == {("T", "W"), ("W", "T")}
|
|
61
|
+
|
|
62
|
+
|
|
63
|
+
@pytest.mark.parametrize("number", sorted(BLOSUM_TABLES))
|
|
64
|
+
def test_blosum_matrix_matches_dict_in_canonical_order(number):
|
|
65
|
+
table, matrix = BLOSUM_TABLES[number]
|
|
66
|
+
for i, x in enumerate(canonical_amino_acid_letters):
|
|
67
|
+
for j, y in enumerate(canonical_amino_acid_letters):
|
|
68
|
+
assert matrix[i, j] == table[x][y]
|
|
@@ -10,10 +10,29 @@
|
|
|
10
10
|
# See the License for the specific language governing permissions and
|
|
11
11
|
# limitations under the License.
|
|
12
12
|
|
|
13
|
-
|
|
13
|
+
import os
|
|
14
|
+
import subprocess
|
|
15
|
+
import sys
|
|
14
16
|
|
|
15
17
|
from pepdata import iedb
|
|
16
18
|
|
|
19
|
+
ALLELE_FIXTURE = os.path.join(os.path.dirname(__file__), "data", "alleles_test_fixture.xml")
|
|
20
|
+
|
|
21
|
+
|
|
22
|
+
def test_load_alleles_in_a_fresh_interpreter():
|
|
23
|
+
# pytest itself imports xml.etree, which hid a bare "import xml" in
|
|
24
|
+
# pepdata.iedb.alleles. Parse in a new interpreter so only pepdata's own
|
|
25
|
+
# imports are loaded.
|
|
26
|
+
script = (
|
|
27
|
+
"import pepdata.iedb.alleles as alleles\n"
|
|
28
|
+
"alleles.local_path = lambda force_download=False: %r\n"
|
|
29
|
+
"print(len(alleles.load_alleles()))\n" % ALLELE_FIXTURE
|
|
30
|
+
)
|
|
31
|
+
result = subprocess.run(
|
|
32
|
+
[sys.executable, "-c", script], capture_output=True, text=True)
|
|
33
|
+
assert result.returncode == 0, result.stderr
|
|
34
|
+
assert int(result.stdout) > 0
|
|
35
|
+
|
|
17
36
|
def test_iedb_human_class1_allele():
|
|
18
37
|
allele_dict = iedb.alleles.load_alleles_dict()
|
|
19
38
|
allele = allele_dict["HLA-C*07:02"]
|
|
@@ -12,8 +12,13 @@
|
|
|
12
12
|
# See the License for the specific language governing permissions and
|
|
13
13
|
# limitations under the License.
|
|
14
14
|
|
|
15
|
+
import pytest
|
|
16
|
+
|
|
15
17
|
from pepdata import iedb
|
|
16
18
|
|
|
19
|
+
# Bound at import, before conftest's autouse fixture replaces it with a fixture path.
|
|
20
|
+
original_mhc_local_path = iedb.mhc.local_path
|
|
21
|
+
|
|
17
22
|
def test_mhc_hla_a2():
|
|
18
23
|
"""
|
|
19
24
|
IEDB MHC: Test that HLA restriction actually decreases number of results and
|
|
@@ -28,3 +33,17 @@ def test_mhc_hla_a2():
|
|
|
28
33
|
assert len(df_a2_combined) <= len(df_a2_1) + len(df_a2_2), \
|
|
29
34
|
"Expected %d <= %d + %d" % \
|
|
30
35
|
(len(df_a2_combined), len(df_a2_1), len(df_a2_2))
|
|
36
|
+
|
|
37
|
+
|
|
38
|
+
def test_mhc_load_dataframe_prints_nothing(capsys):
|
|
39
|
+
# nrows=999 avoids the memoized results of the other tests.
|
|
40
|
+
df = iedb.mhc.load_dataframe(nrows=999)
|
|
41
|
+
assert len(df) > 0
|
|
42
|
+
assert capsys.readouterr().out == ""
|
|
43
|
+
|
|
44
|
+
|
|
45
|
+
def test_mhc_missing_file_error_names_the_download_function(monkeypatch, tmp_path):
|
|
46
|
+
monkeypatch.setattr(
|
|
47
|
+
iedb.mhc.cache, "local_path", lambda **kwargs: str(tmp_path / "missing.csv"))
|
|
48
|
+
with pytest.raises(ValueError, match=r"pepdata\.iedb\.mhc\.download\(\)"):
|
|
49
|
+
original_mhc_local_path(auto_download=False)
|
|
@@ -0,0 +1,119 @@
|
|
|
1
|
+
# Copyright (c) 2014. Mount Sinai School of Medicine
|
|
2
|
+
#
|
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
4
|
+
# you may not use this file except in compliance with the License.
|
|
5
|
+
# You may obtain a copy of the License at
|
|
6
|
+
#
|
|
7
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
|
8
|
+
#
|
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
12
|
+
# See the License for the specific language governing permissions and
|
|
13
|
+
# limitations under the License.
|
|
14
|
+
|
|
15
|
+
import pickle
|
|
16
|
+
|
|
17
|
+
import numpy as np
|
|
18
|
+
import pytest
|
|
19
|
+
|
|
20
|
+
from pepdata import PeptideVectorizer
|
|
21
|
+
from pepdata.reduced_alphabet import hp2
|
|
22
|
+
|
|
23
|
+
# isoforms of two different proteins a, b
|
|
24
|
+
|
|
25
|
+
a1 = (
|
|
26
|
+
"MSPHPTALLGLVLCLAQTIHTQEEDLPRPSISAEPGTVIPLGSHVTFVCRGPVGVQTFRLERESRSTYND"
|
|
27
|
+
"TEDVSQASPSESEARFRIDSVSEGNAGPYRCIYYKPPKWSEQSDYLELLVKETSGGPDSPDTEPGSSAGPT"
|
|
28
|
+
"QRPSDNSHNEHAPASQGLKAEHLYILIGVSVVFLFCLLLLVLFCLHRQNQIKQGPPRSKDEEQKPQQRPDL"
|
|
29
|
+
"AVDVLERTADKATVNGLPEKDRETDTSALAAGSSQEVTYAQLDHWALTQRTARAVSPQSTKPMAESITYAA"
|
|
30
|
+
"VARH"
|
|
31
|
+
)
|
|
32
|
+
|
|
33
|
+
a2 = (
|
|
34
|
+
"MSLMVVSMACVGFFLLQGAWPHEGVHRKPSLLAHPGPLVKSEETVILQCWSDVRFEHFLLHREGKYKDTLH"
|
|
35
|
+
"LIGEHHDGVSKANFSIGPMMQDLAGTYRCYGSVTHSPYQLSAPSDPLDIVITGLYEKPSLSAQPGPTVLAG"
|
|
36
|
+
"ESVTLSCSSRSSYDMYHLSREGEAHERRFSAGPKVNGTFQADFPLGPATHGGTYRCFGSFRDSPYEWSNSS"
|
|
37
|
+
"DPLLVSVTGNPSNSWPSPTEPSSKTGNPRHLHVLIGTSVVKIPFTILLFFLLHRWCSNKKNAAVMDQEPAG"
|
|
38
|
+
"NRTVNSEDSDEQDHQEVSYA"
|
|
39
|
+
)
|
|
40
|
+
|
|
41
|
+
a3 = (
|
|
42
|
+
"MSLMVVSMACVGFFLLEGPWPHVGGQDKPFLSAWPGTVVSEGQHVTLQCRSRLGFNEFSLSKEDGMPVPEL"
|
|
43
|
+
"YNRIFRNSFLMGPVTPAHAGTYRCCSSHPHSPTGWSAPSNPVVIMVTGVHRKPSLLAHPGPLVKSEETVIL"
|
|
44
|
+
"QCWSDVRFEHFLLHREGKYKDTLHLIGEHHDGVSKANFSIGPMMQDLAGTYRCYGSVTHSPYQLSAPSDPL"
|
|
45
|
+
"DIVITGLYEKPSLSAQPGPTVLAGESVTLSCSSRSSYDMYHLSREGEAHERRFSAGPKVNGTFQADFPLGP"
|
|
46
|
+
"ATHGGTYRCFGSFRDSPYEWSNSSDPLLVSVTAFLSVKSSGHKYIY"
|
|
47
|
+
)
|
|
48
|
+
|
|
49
|
+
A = [a1, a2, a3]
|
|
50
|
+
|
|
51
|
+
b1 = (
|
|
52
|
+
"MPKGRAGSLPTTSIGWRFQLWFLGLTCPERHLARRLKNNSFYPFVQQEPNVFVLEYYLDTLWKGMLLFII"
|
|
53
|
+
"SVVLVSFSSLREVQKQETWVFLVYGVGVGLWLVISSLPRRRLVLNHTRGVYHFSIQGRTVCQGPLHLVYV"
|
|
54
|
+
"RLALSSDAHGRCFFHLVLGGHRLEPLVLVQLSEHYEQMEYLGRYIARKLNINYFDYLATSYRHVVRHWPP"
|
|
55
|
+
"PGAGTVMGKSPMGHKPSSSQSSLEV"
|
|
56
|
+
)
|
|
57
|
+
|
|
58
|
+
b2 = (
|
|
59
|
+
"MPKGRAGSLPTTSIGWRFQLWFLGLTCPERHLARRLKNNSFYPFVQQEPNVFVLEYYLDTLWKGMLLFII"
|
|
60
|
+
"SVVLVSFSSLREVQKQETWVFLVYGVGVGLWLVISSLPRRRLVLNHTRGVYHFSIQGRTVCQGPLHLVYV"
|
|
61
|
+
"RLALSSDAHGRCFFHLVLGGHRLEPLVLVQLSEHYEQMEYLGRYIARKLNINYFDYLATSYRHVVRHWPPP"
|
|
62
|
+
"GAGTVMGKSPMGHKPSSSQSSLEV"
|
|
63
|
+
)
|
|
64
|
+
|
|
65
|
+
B = [b1, b2]
|
|
66
|
+
|
|
67
|
+
|
|
68
|
+
@pytest.mark.parametrize("normalize_row", [True, False])
|
|
69
|
+
def test_fit_transform_returns_one_array_row_per_sequence(normalize_row):
|
|
70
|
+
vectorizer = PeptideVectorizer(max_ngram=2, normalize_row=normalize_row)
|
|
71
|
+
X = vectorizer.fit_transform(A + B)
|
|
72
|
+
assert type(X) is np.ndarray
|
|
73
|
+
assert X.shape == (len(A + B), len(vectorizer.count_vectorizer.vocabulary_))
|
|
74
|
+
Y = vectorizer.transform(A + B)
|
|
75
|
+
assert type(Y) is np.ndarray
|
|
76
|
+
np.testing.assert_allclose(Y, X)
|
|
77
|
+
|
|
78
|
+
|
|
79
|
+
def test_default_normalize_row_gives_row_frequencies():
|
|
80
|
+
# The default path raised "np.matrix is not supported" on current
|
|
81
|
+
# scikit-learn (#62).
|
|
82
|
+
vectorizer = PeptideVectorizer()
|
|
83
|
+
X = vectorizer.fit_transform(["SIINFEKL", "GILGFVFTL"])
|
|
84
|
+
assert X.shape == (2, 10)
|
|
85
|
+
np.testing.assert_allclose(X.sum(axis=1), [1.0, 1.0])
|
|
86
|
+
column_i = vectorizer.count_vectorizer.vocabulary_["i"]
|
|
87
|
+
np.testing.assert_allclose(X[:, column_i], [2 / 8, 1 / 9])
|
|
88
|
+
|
|
89
|
+
|
|
90
|
+
def test_normalize_row_false_gives_counts():
|
|
91
|
+
vectorizer = PeptideVectorizer(normalize_row=False)
|
|
92
|
+
X = vectorizer.fit_transform(["SIINFEKL", "GILGFVFTL"])
|
|
93
|
+
np.testing.assert_array_equal(X.sum(axis=1), [8, 9])
|
|
94
|
+
|
|
95
|
+
|
|
96
|
+
@pytest.mark.parametrize("normalize_row, expected_total", [(False, 13.0), (True, 1.0)])
|
|
97
|
+
def test_transform_counts_only_fitted_ngrams(normalize_row, expected_total):
|
|
98
|
+
# SIINFEKA has 15 1- and 2-grams; "A" and "KA" were not seen during fitting.
|
|
99
|
+
vectorizer = PeptideVectorizer(max_ngram=2, normalize_row=normalize_row)
|
|
100
|
+
vectorizer.fit(["SIINFEKL", "SIINFEKV"])
|
|
101
|
+
X = vectorizer.transform(["SIINFEKA"])
|
|
102
|
+
assert X.shape == (1, 16)
|
|
103
|
+
assert X.sum() == pytest.approx(expected_total)
|
|
104
|
+
|
|
105
|
+
|
|
106
|
+
def test_fit_transform_on_already_reduced_training_sequences():
|
|
107
|
+
reduced = ["".join(hp2[residue] for residue in sequence) for sequence in A + B]
|
|
108
|
+
vectorizer = PeptideVectorizer(reduced_alphabet=hp2, training_already_reduced=True)
|
|
109
|
+
X = vectorizer.fit_transform(reduced)
|
|
110
|
+
assert type(X) is np.ndarray
|
|
111
|
+
assert X.shape == (len(A + B), 2)
|
|
112
|
+
np.testing.assert_allclose(X.sum(axis=1), np.ones(len(A + B)))
|
|
113
|
+
|
|
114
|
+
|
|
115
|
+
def test_pickled_vectorizer_transforms_identically():
|
|
116
|
+
vectorizer = PeptideVectorizer(max_ngram=2)
|
|
117
|
+
vectorizer.fit(A)
|
|
118
|
+
restored = pickle.loads(pickle.dumps(vectorizer))
|
|
119
|
+
np.testing.assert_allclose(restored.transform(B), vectorizer.transform(B))
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
import numpy as np
|
|
2
|
+
|
|
3
|
+
from pepdata.amino_acid_alphabet import canonical_amino_acid_letters
|
|
4
|
+
from pepdata.pmbec import (
|
|
5
|
+
pmbec_dict,
|
|
6
|
+
pmbec_matrix,
|
|
7
|
+
)
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
def test_pmbec_covers_canonical_residues():
|
|
11
|
+
assert sorted(pmbec_dict) == sorted(canonical_amino_acid_letters)
|
|
12
|
+
for row in pmbec_dict.values():
|
|
13
|
+
assert sorted(row) == sorted(canonical_amino_acid_letters)
|
|
14
|
+
|
|
15
|
+
|
|
16
|
+
def test_pmbec_matrix_is_symmetric_and_matches_dict():
|
|
17
|
+
assert pmbec_matrix.shape == (20, 20)
|
|
18
|
+
assert np.allclose(pmbec_matrix, pmbec_matrix.T)
|
|
19
|
+
for i, x in enumerate(canonical_amino_acid_letters):
|
|
20
|
+
for j, y in enumerate(canonical_amino_acid_letters):
|
|
21
|
+
assert np.isclose(pmbec_matrix[i, j], pmbec_dict[x][y])
|
|
22
|
+
|
|
23
|
+
|
|
24
|
+
def test_importing_pmbec_prints_nothing():
|
|
25
|
+
import subprocess
|
|
26
|
+
import sys
|
|
27
|
+
|
|
28
|
+
result = subprocess.run(
|
|
29
|
+
[sys.executable, "-c", "import pepdata.pmbec"],
|
|
30
|
+
capture_output=True, text=True, check=True)
|
|
31
|
+
assert result.stdout == ""
|
pepdata-1.2.4/PKG-INFO
DELETED
|
@@ -1,86 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.4
|
|
2
|
-
Name: pepdata
|
|
3
|
-
Version: 1.2.4
|
|
4
|
-
Summary: Immunological peptide datasets and amino acid properties
|
|
5
|
-
Author-email: Alex Rubinsteyn <alex.rubinsteyn@mssm.edu>
|
|
6
|
-
Project-URL: Homepage, https://github.com/openvax/pepdata
|
|
7
|
-
Project-URL: Bug Tracker, https://github.com/openvax/pepdata/issues
|
|
8
|
-
Classifier: Development Status :: 4 - Beta
|
|
9
|
-
Classifier: Environment :: Console
|
|
10
|
-
Classifier: Operating System :: OS Independent
|
|
11
|
-
Classifier: Intended Audience :: Science/Research
|
|
12
|
-
Classifier: License :: OSI Approved :: Apache Software License
|
|
13
|
-
Classifier: Programming Language :: Python
|
|
14
|
-
Classifier: Programming Language :: Python :: 3.11
|
|
15
|
-
Classifier: Programming Language :: Python :: 3.12
|
|
16
|
-
Classifier: Programming Language :: Python :: 3.13
|
|
17
|
-
Classifier: Programming Language :: Python :: 3.14
|
|
18
|
-
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
|
|
19
|
-
Requires-Python: >=3.11
|
|
20
|
-
Description-Content-Type: text/markdown
|
|
21
|
-
License-File: LICENSE.txt
|
|
22
|
-
Requires-Dist: numpy<3.0.0,>=2.0.0
|
|
23
|
-
Requires-Dist: scipy>=0.9
|
|
24
|
-
Requires-Dist: pandas>=0.17
|
|
25
|
-
Requires-Dist: scikit-learn>=0.14.1
|
|
26
|
-
Requires-Dist: progressbar33
|
|
27
|
-
Requires-Dist: biopython>=1.65
|
|
28
|
-
Requires-Dist: datacache>=0.4.4
|
|
29
|
-
Requires-Dist: lxml
|
|
30
|
-
Dynamic: license-file
|
|
31
|
-
|
|
32
|
-
# PepData
|
|
33
|
-
|
|
34
|
-
Look up shipped amino-acid properties and pairwise matrices, encode peptide
|
|
35
|
-
sequences, or load peptide assay tables from IEDB. Property lookups work offline;
|
|
36
|
-
IEDB tables require a download.
|
|
37
|
-
|
|
38
|
-
## Install
|
|
39
|
-
|
|
40
|
-
```sh
|
|
41
|
-
python -m pip install pepdata
|
|
42
|
-
```
|
|
43
|
-
|
|
44
|
-
Python 3.11 or later is required.
|
|
45
|
-
|
|
46
|
-
## Look up a peptide's residue properties
|
|
47
|
-
|
|
48
|
-
```python
|
|
49
|
-
from pepdata.amino_acid_properties import hydropathy
|
|
50
|
-
|
|
51
|
-
peptide = "ACD"
|
|
52
|
-
print([hydropathy[residue] for residue in peptide])
|
|
53
|
-
```
|
|
54
|
-
|
|
55
|
-
```text
|
|
56
|
-
[1.8, 2.5, -3.5]
|
|
57
|
-
```
|
|
58
|
-
|
|
59
|
-
The values come from the shipped residue-property table. The dictionary maps
|
|
60
|
-
single-letter residues to numeric values; it does not predict MHC binding or
|
|
61
|
-
clinical immunogenicity. Use uppercase canonical residue letters. A missing
|
|
62
|
-
letter raises `KeyError`.
|
|
63
|
-
|
|
64
|
-
## Find the interface you need
|
|
65
|
-
|
|
66
|
-
| Task | Module or object |
|
|
67
|
-
| --- | --- |
|
|
68
|
-
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
69
|
-
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
70
|
-
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
71
|
-
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
72
|
-
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
73
|
-
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
74
|
-
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
75
|
-
|
|
76
|
-
See [residue and pairwise tables](docs/guides/properties.md), then
|
|
77
|
-
[peptide features and IEDB tables](docs/guides/datasets.md). The
|
|
78
|
-
[API reference](docs/reference.md) follows those same tasks.
|
|
79
|
-
|
|
80
|
-
Older pepdata documentation placed the property tables in `amino_acid` and
|
|
81
|
-
described a `StringTransformer` interface. The property dictionaries are
|
|
82
|
-
in `amino_acid_properties`, and there is no `StringTransformer` class.
|
|
83
|
-
|
|
84
|
-
## Documentation development
|
|
85
|
-
|
|
86
|
-
Install `requirements-docs.txt`, run `./docs.sh`, then `python scripts/check_docs_examples.py`.
|
pepdata-1.2.4/README.md
DELETED
|
@@ -1,55 +0,0 @@
|
|
|
1
|
-
# PepData
|
|
2
|
-
|
|
3
|
-
Look up shipped amino-acid properties and pairwise matrices, encode peptide
|
|
4
|
-
sequences, or load peptide assay tables from IEDB. Property lookups work offline;
|
|
5
|
-
IEDB tables require a download.
|
|
6
|
-
|
|
7
|
-
## Install
|
|
8
|
-
|
|
9
|
-
```sh
|
|
10
|
-
python -m pip install pepdata
|
|
11
|
-
```
|
|
12
|
-
|
|
13
|
-
Python 3.11 or later is required.
|
|
14
|
-
|
|
15
|
-
## Look up a peptide's residue properties
|
|
16
|
-
|
|
17
|
-
```python
|
|
18
|
-
from pepdata.amino_acid_properties import hydropathy
|
|
19
|
-
|
|
20
|
-
peptide = "ACD"
|
|
21
|
-
print([hydropathy[residue] for residue in peptide])
|
|
22
|
-
```
|
|
23
|
-
|
|
24
|
-
```text
|
|
25
|
-
[1.8, 2.5, -3.5]
|
|
26
|
-
```
|
|
27
|
-
|
|
28
|
-
The values come from the shipped residue-property table. The dictionary maps
|
|
29
|
-
single-letter residues to numeric values; it does not predict MHC binding or
|
|
30
|
-
clinical immunogenicity. Use uppercase canonical residue letters. A missing
|
|
31
|
-
letter raises `KeyError`.
|
|
32
|
-
|
|
33
|
-
## Find the interface you need
|
|
34
|
-
|
|
35
|
-
| Task | Module or object |
|
|
36
|
-
| --- | --- |
|
|
37
|
-
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
38
|
-
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
39
|
-
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
40
|
-
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
41
|
-
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
42
|
-
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
43
|
-
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
44
|
-
|
|
45
|
-
See [residue and pairwise tables](docs/guides/properties.md), then
|
|
46
|
-
[peptide features and IEDB tables](docs/guides/datasets.md). The
|
|
47
|
-
[API reference](docs/reference.md) follows those same tasks.
|
|
48
|
-
|
|
49
|
-
Older pepdata documentation placed the property tables in `amino_acid` and
|
|
50
|
-
described a `StringTransformer` interface. The property dictionaries are
|
|
51
|
-
in `amino_acid_properties`, and there is no `StringTransformer` class.
|
|
52
|
-
|
|
53
|
-
## Documentation development
|
|
54
|
-
|
|
55
|
-
Install `requirements-docs.txt`, run `./docs.sh`, then `python scripts/check_docs_examples.py`.
|
|
@@ -1,86 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.4
|
|
2
|
-
Name: pepdata
|
|
3
|
-
Version: 1.2.4
|
|
4
|
-
Summary: Immunological peptide datasets and amino acid properties
|
|
5
|
-
Author-email: Alex Rubinsteyn <alex.rubinsteyn@mssm.edu>
|
|
6
|
-
Project-URL: Homepage, https://github.com/openvax/pepdata
|
|
7
|
-
Project-URL: Bug Tracker, https://github.com/openvax/pepdata/issues
|
|
8
|
-
Classifier: Development Status :: 4 - Beta
|
|
9
|
-
Classifier: Environment :: Console
|
|
10
|
-
Classifier: Operating System :: OS Independent
|
|
11
|
-
Classifier: Intended Audience :: Science/Research
|
|
12
|
-
Classifier: License :: OSI Approved :: Apache Software License
|
|
13
|
-
Classifier: Programming Language :: Python
|
|
14
|
-
Classifier: Programming Language :: Python :: 3.11
|
|
15
|
-
Classifier: Programming Language :: Python :: 3.12
|
|
16
|
-
Classifier: Programming Language :: Python :: 3.13
|
|
17
|
-
Classifier: Programming Language :: Python :: 3.14
|
|
18
|
-
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
|
|
19
|
-
Requires-Python: >=3.11
|
|
20
|
-
Description-Content-Type: text/markdown
|
|
21
|
-
License-File: LICENSE.txt
|
|
22
|
-
Requires-Dist: numpy<3.0.0,>=2.0.0
|
|
23
|
-
Requires-Dist: scipy>=0.9
|
|
24
|
-
Requires-Dist: pandas>=0.17
|
|
25
|
-
Requires-Dist: scikit-learn>=0.14.1
|
|
26
|
-
Requires-Dist: progressbar33
|
|
27
|
-
Requires-Dist: biopython>=1.65
|
|
28
|
-
Requires-Dist: datacache>=0.4.4
|
|
29
|
-
Requires-Dist: lxml
|
|
30
|
-
Dynamic: license-file
|
|
31
|
-
|
|
32
|
-
# PepData
|
|
33
|
-
|
|
34
|
-
Look up shipped amino-acid properties and pairwise matrices, encode peptide
|
|
35
|
-
sequences, or load peptide assay tables from IEDB. Property lookups work offline;
|
|
36
|
-
IEDB tables require a download.
|
|
37
|
-
|
|
38
|
-
## Install
|
|
39
|
-
|
|
40
|
-
```sh
|
|
41
|
-
python -m pip install pepdata
|
|
42
|
-
```
|
|
43
|
-
|
|
44
|
-
Python 3.11 or later is required.
|
|
45
|
-
|
|
46
|
-
## Look up a peptide's residue properties
|
|
47
|
-
|
|
48
|
-
```python
|
|
49
|
-
from pepdata.amino_acid_properties import hydropathy
|
|
50
|
-
|
|
51
|
-
peptide = "ACD"
|
|
52
|
-
print([hydropathy[residue] for residue in peptide])
|
|
53
|
-
```
|
|
54
|
-
|
|
55
|
-
```text
|
|
56
|
-
[1.8, 2.5, -3.5]
|
|
57
|
-
```
|
|
58
|
-
|
|
59
|
-
The values come from the shipped residue-property table. The dictionary maps
|
|
60
|
-
single-letter residues to numeric values; it does not predict MHC binding or
|
|
61
|
-
clinical immunogenicity. Use uppercase canonical residue letters. A missing
|
|
62
|
-
letter raises `KeyError`.
|
|
63
|
-
|
|
64
|
-
## Find the interface you need
|
|
65
|
-
|
|
66
|
-
| Task | Module or object |
|
|
67
|
-
| --- | --- |
|
|
68
|
-
| Single-residue properties | `pepdata.amino_acid_properties` |
|
|
69
|
-
| BLOSUM substitution matrices | `pepdata.blosum` |
|
|
70
|
-
| PMBEC and contact tables | `pepdata.pmbec`, `pepdata.residue_contact_energies` |
|
|
71
|
-
| Residue alphabets | `pepdata.amino_acid_alphabet`, `pepdata.reduced_alphabet` |
|
|
72
|
-
| N-gram features | `pepdata.PeptideVectorizer` |
|
|
73
|
-
| MHC-binding assay tables | `pepdata.iedb.mhc` |
|
|
74
|
-
| T-cell assay tables | `pepdata.iedb.tcell` |
|
|
75
|
-
|
|
76
|
-
See [residue and pairwise tables](docs/guides/properties.md), then
|
|
77
|
-
[peptide features and IEDB tables](docs/guides/datasets.md). The
|
|
78
|
-
[API reference](docs/reference.md) follows those same tasks.
|
|
79
|
-
|
|
80
|
-
Older pepdata documentation placed the property tables in `amino_acid` and
|
|
81
|
-
described a `StringTransformer` interface. The property dictionaries are
|
|
82
|
-
in `amino_acid_properties`, and there is no `StringTransformer` class.
|
|
83
|
-
|
|
84
|
-
## Documentation development
|
|
85
|
-
|
|
86
|
-
Install `requirements-docs.txt`, run `./docs.sh`, then `python scripts/check_docs_examples.py`.
|
|
@@ -1,61 +0,0 @@
|
|
|
1
|
-
# Copyright (c) 2014. Mount Sinai School of Medicine
|
|
2
|
-
#
|
|
3
|
-
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
4
|
-
# you may not use this file except in compliance with the License.
|
|
5
|
-
# You may obtain a copy of the License at
|
|
6
|
-
#
|
|
7
|
-
# http://www.apache.org/licenses/LICENSE-2.0
|
|
8
|
-
#
|
|
9
|
-
# Unless required by applicable law or agreed to in writing, software
|
|
10
|
-
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
11
|
-
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
12
|
-
# See the License for the specific language governing permissions and
|
|
13
|
-
# limitations under the License.
|
|
14
|
-
|
|
15
|
-
import pickle
|
|
16
|
-
|
|
17
|
-
from pepdata import PeptideVectorizer
|
|
18
|
-
|
|
19
|
-
# isoforms of two different proteins a, b
|
|
20
|
-
|
|
21
|
-
a1 = (
|
|
22
|
-
"MSPHPTALLGLVLCLAQTIHTQEEDLPRPSISAEPGTVIPLGSHVTFVCRGPVGVQTFRLERESRSTYND"
|
|
23
|
-
"TEDVSQASPSESEARFRIDSVSEGNAGPYRCIYYKPPKWSEQSDYLELLVKETSGGPDSPDTEPGSSAGPT"
|
|
24
|
-
"QRPSDNSHNEHAPASQGLKAEHLYILIGVSVVFLFCLLLLVLFCLHRQNQIKQGPPRSKDEEQKPQQRPDL"
|
|
25
|
-
"AVDVLERTADKATVNGLPEKDRETDTSALAAGSSQEVTYAQLDHWALTQRTARAVSPQSTKPMAESITYAA"
|
|
26
|
-
"VARH"
|
|
27
|
-
)
|
|
28
|
-
|
|
29
|
-
a2 = (
|
|
30
|
-
"MSLMVVSMACVGFFLLQGAWPHEGVHRKPSLLAHPGPLVKSEETVILQCWSDVRFEHFLLHREGKYKDTLH"
|
|
31
|
-
"LIGEHHDGVSKANFSIGPMMQDLAGTYRCYGSVTHSPYQLSAPSDPLDIVITGLYEKPSLSAQPGPTVLAG"
|
|
32
|
-
"ESVTLSCSSRSSYDMYHLSREGEAHERRFSAGPKVNGTFQADFPLGPATHGGTYRCFGSFRDSPYEWSNSS"
|
|
33
|
-
"DPLLVSVTGNPSNSWPSPTEPSSKTGNPRHLHVLIGTSVVKIPFTILLFFLLHRWCSNKKNAAVMDQEPAG"
|
|
34
|
-
"NRTVNSEDSDEQDHQEVSYA"
|
|
35
|
-
)
|
|
36
|
-
|
|
37
|
-
a3 = (
|
|
38
|
-
"MSLMVVSMACVGFFLLEGPWPHVGGQDKPFLSAWPGTVVSEGQHVTLQCRSRLGFNEFSLSKEDGMPVPEL"
|
|
39
|
-
"YNRIFRNSFLMGPVTPAHAGTYRCCSSHPHSPTGWSAPSNPVVIMVTGVHRKPSLLAHPGPLVKSEETVIL"
|
|
40
|
-
"QCWSDVRFEHFLLHREGKYKDTLHLIGEHHDGVSKANFSIGPMMQDLAGTYRCYGSVTHSPYQLSAPSDPL"
|
|
41
|
-
"DIVITGLYEKPSLSAQPGPTVLAGESVTLSCSSRSSYDMYHLSREGEAHERRFSAGPKVNGTFQADFPLGP"
|
|
42
|
-
"ATHGGTYRCFGSFRDSPYEWSNSSDPLLVSVTAFLSVKSSGHKYIY"
|
|
43
|
-
)
|
|
44
|
-
|
|
45
|
-
A = [a1, a2, a3]
|
|
46
|
-
|
|
47
|
-
b1 = (
|
|
48
|
-
"MPKGRAGSLPTTSIGWRFQLWFLGLTCPERHLARRLKNNSFYPFVQQEPNVFVLEYYLDTLWKGMLLFII"
|
|
49
|
-
"SVVLVSFSSLREVQKQETWVFLVYGVGVGLWLVISSLPRRRLVLNHTRGVYHFSIQGRTVCQGPLHLVYV"
|
|
50
|
-
"RLALSSDAHGRCFFHLVLGGHRLEPLVLVQLSEHYEQMEYLGRYIARKLNINYFDYLATSYRHVVRHWPP"
|
|
51
|
-
"PGAGTVMGKSPMGHKPSSSQSSLEV"
|
|
52
|
-
)
|
|
53
|
-
|
|
54
|
-
b2 = (
|
|
55
|
-
"MPKGRAGSLPTTSIGWRFQLWFLGLTCPERHLARRLKNNSFYPFVQQEPNVFVLEYYLDTLWKGMLLFII"
|
|
56
|
-
"SVVLVSFSSLREVQKQETWVFLVYGVGVGLWLVISSLPRRRLVLNHTRGVYHFSIQGRTVCQGPLHLVYV"
|
|
57
|
-
"RLALSSDAHGRCFFHLVLGGHRLEPLVLVQLSEHYEQMEYLGRYIARKLNINYFDYLATSYRHVVRHWPPP"
|
|
58
|
-
"GAGTVMGKSPMGHKPSSSQSSLEV"
|
|
59
|
-
)
|
|
60
|
-
|
|
61
|
-
B = [b1, b2]
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|