bayesianflow-for-chem 2.2.1__tar.gz → 2.2.2__tar.gz

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bayesianflow_for_chem
-Version: 2.2.1
+Version: 2.2.2
 Summary: Bayesian flow network framework for Chemistry
 Home-page: https://augus1999.github.io/bayesian-flow-network-for-chemistry/
 Author: Nianze A. Tao
@@ -96,7 +96,7 @@ You can find pretrained models on our [🤗Hugging Face model page](https://hugg
 
 ## Dataset Handling
 
-We provide a Python class [`CSVData`](https://github.com/Augus1999/bayesian-flow-network-for-chemistry/blob/main/bayesianflow_for_chem/data.py#L152) to handle data stored in CSV or similar format containing headers to identify the entities. The following is a quickstart.
+We provide a Python class [`CSVData`](https://github.com/Augus1999/bayesian-flow-network-for-chemistry/blob/main/bayesianflow_for_chem/data.py#L153) to handle data stored in CSV or similar format containing headers to identify the entities. The following is a quickstart.
 
 1. Download your dataset file (e.g., ESOL from [MoleculeNet](https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/delaney-processed.csv)) and split the file:
 ```python

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/bayesianflow_for_chem/__init__.py

@@ -16,7 +16,7 @@ __all__ = [
     "MLP",
     "EnsembleChemBFN",
 ]
-__version__ = "2.2.1"
+__version__ = "2.2.2"
 __author__ = "Nianze A. Tao (Omozawa Sueno)"
 
 

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/bayesianflow_for_chem/cli.py

@@ -18,11 +18,11 @@ from bayesianflow_for_chem.scorer import smiles_valid, Scorer
 from bayesianflow_for_chem.data import (
     VOCAB_COUNT,
     VOCAB_KEYS,
-    AA_VOCAB_COUNT,
-    AA_VOCAB_KEYS,
+    FASTA_VOCAB_COUNT,
+    FASTA_VOCAB_KEYS,
     load_vocab,
     smiles2token,
-    aa2token,
+    fasta2token,
     split_selfies,
     collate,
     CSVData,
@@ -116,6 +116,23 @@ madmadmadmadmadmadmadmadmadmadmadmadmadmadmad
 madmadmadmadmadmadmadmadmadmadmadmadmadmadmad
 """
 
+_END_MESSAGE = r"""
+If you find this project helpful, please cite us:
+1. N. Tao, and M. Abe, J. Chem. Inf. Model., 2025, 65, 1178-1187.
+2. N. Tao, 2024, arXiv:2412.11439.
+"""
+
+_ERROR_MESSAGE = r"""
+Some who believe in inductive logic are anxious to point out, with
+Reichenbach, that 'the principle of induction is unreservedly accepted
+by the whole of science and that no man can seriously doubt this
+principle in everyday life either'. Yet even supposing this were the
+case—for after all, 'the whole of science' might err—I should still
+contend that a principle of induction is superfluous, and that it must
+lead to logical inconsistencies.  
+                        -- Karl Popper --
+"""
+
 _ALLOWED_PLUGINS = [
     "collate_fn",
     "num_workers",
@@ -396,7 +413,7 @@ def main_script(version: str) -> None:
             print("Configuration check passed.")
         return
     if flag_critical != 0:
-        raise RuntimeError
+        raise RuntimeError(_ERROR_MESSAGE)
     print(_MESSAGE.format(version))
     # ####### build tokeniser #######
     tokeniser_config = runtime_config["tokeniser"]
@@ -406,9 +423,9 @@ def main_script(version: str) -> None:
         vocab_keys = VOCAB_KEYS
         tokeniser = smiles2token
     if tokeniser_name == "fasta":
-        num_vocab = AA_VOCAB_COUNT
-        vocab_keys = AA_VOCAB_KEYS
-        tokeniser = aa2token
+        num_vocab = FASTA_VOCAB_COUNT
+        vocab_keys = FASTA_VOCAB_KEYS
+        tokeniser = fasta2token
     if tokeniser_name == "selfies":
         vocab_data = load_vocab(tokeniser_config["vocab"])
         num_vocab = vocab_data["vocab_count"]
@@ -680,6 +697,7 @@ def main_script(version: str) -> None:
             f.write("\n".join(mols))
     # ------- finished -------
     print(" ####### job finished #######")
+    print(_END_MESSAGE)
 
 
 if __name__ == "__main__":

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/bayesianflow_for_chem/data.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 # Author: Nianze A. TAO (Omozawa SUENO)
 """
-Tokenise SMILES/SAFE/SELFIES/protein-sequence strings.
+Tokenise SMILES/SAFE/SELFIES/FASTA strings.
 """
 import os
 import re
@@ -14,7 +14,7 @@ from torch.utils.data import Dataset
 
 __filedir__ = Path(__file__).parent
 
-SMI_REGEX_PATTERN = (
+_SMI_REGEX_PATTERN = (
     r"(\[|\]|H[e,f,g,s,o]?|"
     r"L[i,v,a,r,u]|"
     r"B[e,r,a,i,h,k]?|"
@@ -31,11 +31,11 @@ SMI_REGEX_PATTERN = (
     r"\(|\)|\.|=|#|-|\+|\\|\/|:|"
     r"~|@|\?|>>?|\*|\$|\%[0-9]{2}|[0-9])"
 )
-SEL_REGEX_PATTERN = r"(\[[^\]]+]|\.)"
-AA_REGEX_PATTERN = r"(A|B|C|D|E|F|G|H|I|K|L|M|N|P|Q|R|S|T|V|W|Y|Z|-|.)"
-smi_regex = re.compile(SMI_REGEX_PATTERN)
-sel_regex = re.compile(SEL_REGEX_PATTERN)
-aa_regex = re.compile(AA_REGEX_PATTERN)
+_SEL_REGEX_PATTERN = r"(\[[^\]]+]|\.)"
+_FAS_REGEX_PATTERN = r"(A|B|C|D|E|F|G|H|I|J|K|L|M|N|O|P|Q|R|S|T|U|V|W|X|Y|Z|-|\*|\.)"
+_smi_regex = re.compile(_SMI_REGEX_PATTERN)
+_sel_regex = re.compile(_SEL_REGEX_PATTERN)
+_fas_regex = re.compile(_FAS_REGEX_PATTERN)
 
 
 def load_vocab(
@@ -65,11 +65,12 @@ _DEFUALT_VOCAB = load_vocab(__filedir__ / "_data/vocab.txt")
 VOCAB_KEYS: List[str] = _DEFUALT_VOCAB["vocab_keys"]
 VOCAB_DICT: Dict[str, int] = _DEFUALT_VOCAB["vocab_dict"]
 VOCAB_COUNT: int = _DEFUALT_VOCAB["vocab_count"]
-AA_VOCAB_KEYS = (
-    VOCAB_KEYS[0:3] + "A B C D E F G H I K L M N P Q R S T V W Y Z - .".split()
+FASTA_VOCAB_KEYS = (
+    VOCAB_KEYS[0:3]
+    + "A B C D E F G H I K L M N P Q R S T V W Y Z - . J O U X *".split()
 )
-AA_VOCAB_COUNT = len(AA_VOCAB_KEYS)
-AA_VOCAB_DICT = dict(zip(AA_VOCAB_KEYS, range(AA_VOCAB_COUNT)))
+FASTA_VOCAB_COUNT = len(FASTA_VOCAB_KEYS)
+FASTA_VOCAB_DICT = dict(zip(FASTA_VOCAB_KEYS, range(FASTA_VOCAB_COUNT)))
 
 
 def smiles2vec(smiles: str) -> List[int]:
@@ -81,21 +82,21 @@ def smiles2vec(smiles: str) -> List[int]:
     :return: tokens w/o `<start>` and `<end>`
     :rtype: list
     """
-    tokens = [token for token in smi_regex.findall(smiles)]
+    tokens = [token for token in _smi_regex.findall(smiles)]
     return [VOCAB_DICT[token] for token in tokens]
 
 
-def aa2vec(aa_seq: str) -> List[int]:
+def fasta2vec(fasta: str) -> List[int]:
     """
-    Protein sequence tokenisation using a dataset-independent regex pattern.
+    FASTA sequence tokenisation using a dataset-independent regex pattern.
 
-    :param aa_seq: protein (amino acid) sequence
-    :type aa_seq: str
+    :param fasta: protein (amino acid) sequence
+    :type fasta: str
     :return: tokens w/o `<start>` and `<end>`
     :rtype: list
     """
-    tokens = [token for token in aa_regex.findall(aa_seq)]
-    return [AA_VOCAB_DICT[token] for token in tokens]
+    tokens = [token for token in _fas_regex.findall(fasta)]
+    return [FASTA_VOCAB_DICT[token] for token in tokens]
 
 
 def split_selfies(selfies: str) -> List[str]:
@@ -107,7 +108,7 @@ def split_selfies(selfies: str) -> List[str]:
     :return: SELFIES vocab
     :rtype: list
     """
-    return [token for token in sel_regex.findall(selfies)]
+    return [token for token in _sel_regex.findall(selfies)]
 
 
 def smiles2token(smiles: str) -> Tensor:
@@ -115,9 +116,9 @@ def smiles2token(smiles: str) -> Tensor:
     return torch.tensor([1] + smiles2vec(smiles) + [2], dtype=torch.long)
 
 
-def aa2token(aa_seq: str) -> Tensor:
+def fasta2token(fasta: str) -> Tensor:
     # start token: <start> = 1; end token: <end> = 2
-    return torch.tensor([1] + aa2vec(aa_seq) + [2], dtype=torch.long)
+    return torch.tensor([1] + fasta2vec(fasta) + [2], dtype=torch.long)
 
 
 def collate(batch: List[Dict[str, Tensor]]) -> Dict[str, Tensor]:

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/bayesianflow_for_chem/tool.py

@@ -7,7 +7,7 @@ import csv
 import random
 import warnings
 from pathlib import Path
-from typing import List, Dict, Tuple, Union, Optional
+from typing import List, Dict, Tuple, Union, Optional, Literal
 import torch
 import numpy as np
 from torch import cuda, Tensor, softmax
@@ -103,7 +103,7 @@ def test(
     model: ChemBFN,
     mlp: MLP,
     data: DataLoader,
-    mode: str = "regression",
+    mode: Literal["regression", "classification"] = "regression",
     device: Union[str, torch.device, None] = None,
 ) -> Dict[str, float]:
     """
@@ -184,7 +184,9 @@ def test(
 
 
 def split_dataset(
-    file: Union[str, Path], split_ratio: List[int] = [8, 1, 1], method: str = "random"
+    file: Union[str, Path],
+    split_ratio: List[int] = [8, 1, 1],
+    method: Literal["random", "scaffold"] = "random",
 ) -> None:
     """
     Split a dataset.
@@ -526,7 +528,7 @@ class GeometryConverter:
     def smiles2cartesian(
         smiles: str,
         num_conformers: int = 250,
-        rdkit_ff_type: str = "MMFF",
+        rdkit_ff_type: Literal["MMFF", "UFF"] = "MMFF",
         refine_with_crest: bool = False,
         spin: float = 0.0,
     ) -> Tuple[List[str], np.ndarray]:

{bayesianflow_for_chem-2.2.1 → bayesianflow_for_chem-2.2.2}/bayesianflow_for_chem.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bayesianflow_for_chem
-Version: 2.2.1
+Version: 2.2.2
 Summary: Bayesian flow network framework for Chemistry
 Home-page: https://augus1999.github.io/bayesian-flow-network-for-chemistry/
 Author: Nianze A. Tao
@@ -96,7 +96,7 @@ You can find pretrained models on our [🤗Hugging Face model page](https://hugg
 
 ## Dataset Handling
 
-We provide a Python class [`CSVData`](https://github.com/Augus1999/bayesian-flow-network-for-chemistry/blob/main/bayesianflow_for_chem/data.py#L152) to handle data stored in CSV or similar format containing headers to identify the entities. The following is a quickstart.
+We provide a Python class [`CSVData`](https://github.com/Augus1999/bayesian-flow-network-for-chemistry/blob/main/bayesianflow_for_chem/data.py#L153) to handle data stored in CSV or similar format containing headers to identify the entities. The following is a quickstart.
 
 1. Download your dataset file (e.g., ESOL from [MoleculeNet](https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/delaney-processed.csv)) and split the file:
 ```python