OSCR

A Bond-Level Sequence Framework for Molecular Representation Learning with Structural Constraints.

Code ↔ Paper

3 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 3 matches · 1 of them tie a paragraph to a whole file, not to given lines: a weak match, whose lines are not tinted
  1. [1] § 3. Materials and Methods › 3.2. Materials › 3.2.2. Implementation Details ↔ utils/config.py, the whole file · a weak match · score 0.57 · masking ratios, hyperparameters, schedules, triplet, seeds, layer
  2. [2] § 3. Materials and Methods › 3.1. Methodology › 3.1.1. Preliminary ↔ datasets/dataset.py, lines 32–75 · score 0.56 · formal charges, atomic, aromaticity, conjugation, ring, vector
  3. [3] § 3. Materials and Methods › 3.1. Methodology › 3.1.1. Preliminary ↔ scripts/preprocess.py, lines 46–62 · score 0.56 · formal charges, atomic, aromaticity, conjugation, ring, vector

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 152 lines · 4.3 KB · no license · 1 match

  1. # utils/config.py
  2. import os
  3. import json
  4. import time
  5. import torch
  7. class Config:
  8. seed = 202
  9. device = "cuda" if torch.cuda.is_available() else "cpu"
  10. num_workers = 0
  11. pin_memory = False
  13. # -----------------------
  14. # 1) Data
  15. # -----------------------
  16. train_csv = "data/pretrain/10K.csv"
  17. smiles_col = "SMILES"
  18. vocab_path = "data/splits/vocab_directed.json"
  19. target_cache_path = "data/pretrain/cache/data_directed.pt"
  22. max_seq_len = 64
  23. mask_ratio_1 = 0.15
  24. mask_ratio_2 = 0.30
  26. # -----------------------
  27. # 2) Model (Backbone)
  28. # -----------------------
  29. embed_dim = 256#512 #256
  30. num_heads = 32#64#32
  31. hidden_dim = 1024
  32. num_layers = 4
  33. dropout = 0.1
  35. global_attr_dim = 4
  36. adj_ratio = 0.85
  38. # -----------------------
  39. # 3) Optim / Schedule
  40. # -----------------------
  41. epochs = 20
  42. batch_size = 256
  43. lr = 1e-5
  44. weight_decay = 0.01
  45. grad_clip_norm = 5.0
  46. best_ema_beta = 0.9
  48. # -----------------------
  49. # 4) Pretrain Loss Hyperparams
  50. # -----------------------
  51. contrastive_temperature = 0.1
  52. triplet_margin = 1e-6
  53. triplet_loss_weight = 0.1
  55. # -----------------------
  56. # 5) Output
  57. # -----------------------
  58. out_root = "out/pretrain"
  59. run_name = "" # 为空则自动时间戳
  60. save_every_epochs = 1
  61. resume = True
  62. log_every_steps = 50
  64. # -----------------------
  65. # 6) Finetune (Classification)
  66. # -----------------------
  67. ft_dataset_name = "tox21"
  68. ft_data_root = "data/finetune"
  69. ft_epochs = 50
  70. ft_batch_size = 32
  71. ft_lr = 2e-5
  72. ft_encoder_lr_ratio = 0.05
  73. ft_weight_decay = 0.01
  74. ft_dropout = 0.2
  75. ft_freeze_epochs = 5
  76. unfreeze_last_n = -1
  77. ft_patience = 10
  78. ft_pos_weight_mode = "sqrt"
  79. pretrain_ckpt = "out/pretrain/best.pt"
  80. ft_vocab_path_un = "data/splits/vocab_undirected.json"
  81. ft_vocab_path_di = "data/splits/vocab_directed.json"
  83. mask_strategy = "structured" # random 或 structured
  84. use_consistency = True
  85. use_triplet = True
  86. lambda_cl = 1.0
  87. lambda_tri = 1.0
  91. # ============= 🆕 新增/修改的方法 =============
  93. @classmethod
  94. def override_from_json(cls, json_path: str) -> None:
  95. """从 JSON 文件加载参数并覆盖当前配置"""
  96. if not json_path or not os.path.exists(json_path):
  97. print(f"[Config] Warning: Config file not found: {json_path}")
  98. return
  100. print(f"[Config] Loading params from {json_path} ...")
  101. with open(json_path, "r", encoding="utf-8") as f:
  102. params = json.load(f)
  104. cls.override(params)
  106. @classmethod
  107. def override(cls, params: dict) -> None:
  108. """使用字典批量覆盖配置"""
  109. for k, v in params.items():
  110. if hasattr(cls, k):
  111. # 类型安全检查（可选）：防止把 float 覆盖成 str
  112. original_val = getattr(cls, k)
  113. if isinstance(original_val, float) and isinstance(v, int):
  114. v = float(v)
  116. setattr(cls, k, v)
  117. # 只有在值真的改变时才打印，避免刷屏
  118. if str(original_val) != str(v):
  119. print(f" -> Override: {k} = {v}")
  120. else:
  121. print(f" -> [Ignore] Unknown key: {k}")
  123. @classmethod
  124. def get_run_name(cls) -> str:
  125. return cls.run_name or time.strftime("%m.%d_%H-%M-%S", time.localtime())
  127. @classmethod
  128. def get_run_dir(cls) -> str:
  129. return os.path.join(cls.out_root, cls.get_run_name())
  131. @classmethod
  132. def ensure_dirs(cls) -> str:
  133. run_dir = cls.get_run_dir()
  134. os.makedirs(run_dir, exist_ok=True)
  135. os.makedirs(os.path.join(run_dir, "ckpt"), exist_ok=True)
  136. return run_dir
  138. @classmethod
  139. def dump(cls, run_dir: str) -> None:
  140. """保存最终配置，方便复现"""
  141. cfg = {}
  142. for k, v in vars(cls).items():
  143. if k.startswith("_"): continue
  144. if callable(v) or isinstance(v, (classmethod, staticmethod)): continue
  145. try:
  146. json.dumps(v)
  147. cfg[k] = v
  148. except TypeError:
  149. cfg[k] = str(v)
  151. with open(os.path.join(run_dir, "config.json"), "w", encoding="utf-8") as f:
  152. json.dump(cfg, f, ensure_ascii=False, indent=2)

config.py at commit 5b5028e, no license · at the source

Overview

Authors: Haoran Fan1, Haoqiang Qi1, Xin Huang1,2, Dongyang Zhu1,2, Na Wang1,2, Ting Wang1,2, Hongxun Hao1,2
ORCID iDs: Xin Huang, Hongxun Hao
  1. National Engineering Research Center of Industrial Crystallization Technology, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China; (H.F.); (H.Q.); (X.H.); (D.Z.)
  2. Engineering Research Center of Green Refining Process, Ministry of Education, Tianjin University, Tianjin 300072, China
Institutions: Tianjin University (China)
Journal: Molecules (Basel, Switzerland), volume 31, issue 11, article 1972
Dates: received 12 May 2026; accepted 3 June 2026; published online 5 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3390/molecules31111972 · PMID 42280272 · PMCID PMC13257498 · OpenAlex W7163639147
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: cellular / molecular (subfield)
Methods: Statistics, Machine learning
Keywords: molecular property prediction, transformer, bond-level representation, self-supervised learning, structural degeneracy
Topic: Computational Drug Discovery Methods (Computational Theory and Mathematics, Computer Science), according to OpenAlex
Funding: National Natural Science Foundation of China (22478281, 22578314)
Citations: not cited yet (Europe PMC); 27 references in the paper

Abstract

Molecular property prediction is a fundamental task in drug discovery and materials design. While graph neural networks (GNNs) and SMILES-based Transformers have made significant strides, the former are often limited by local message-passing bottlenecks such as over-squashing, while the latter frequently lack explicit topological constraints and suffer from severe vocabulary imbalance. In this work, we revisit the granularity of molecular modeling and propose a representation learning framework built upon bond-level sequences. Our framework models molecules as sequences of directed bond tokens and introduces a structure-aware hybrid attention mechanism. By imposing hard topological constraints on a subset of attention heads to reinforce local connectivity while preserving global receptive fields in the remaining heads, the design is intended to separate short-range chemical bonding from long-range contextual dependencies. For pre-training, we implemented a multi-scale consistency learning paradigm, which utilizes an atom-centric group masking strategy to induce a hierarchical loss of local structural information and employs contrastive and triplet losses to ensure identity consistency across varying scales of structural degradation. Furthermore, by incorporating macro-scale physicochemical descriptors (e.g., LogP, TPSA) as global anchors, we examined how the inclusion of global attribute bias can provide weak physicochemical priors during pre-training, while its effect during downstream fine-tuning remains task-dependent. Experimental results demonstrate that our lightweight model, with approximately 3.5 million parameters, exhibits a dataset-dependent performance profile across MoleculeNet benchmarks and shows promising behavior on selected topology-sensitive tasks, particularly MUV. Ablation studies further analyze the contribution of bond-level connectivity, the stage-dependent dynamics of global attribute bias, structured masking, and pre-training configurations. Ultimately, this work provides an alternative representation design for molecular modeling, offering a parameter-efficient option for future molecular learning systems alongside traditional SMILES-based and graph-based formulations.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 3 matches between paragraphs and lines of code.

Haoran-Fan/bondformer

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 5b5028ee4f3b0df07ee10bc5bb9cd174b59f8366, 11 March 2026
Languages: Python (20)
Size: 30 files, 20 scripts
Software Heritage: not archived
Found in: “Data Availability Statement”
Holds: README, environment (environment.yml)
Not found: license file, CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (15 files), NumPy (6 files), pandas (5 files), RDKit (2 files), scikit-learn (2 files)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
21 files
At the source: github.com/Haoran-Fan/bondformer

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 20 scripts, each with its path and the digest of its content;
  • 3 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The source code, environment specifications, and a self-contained demonstration dataset supporting the findings of this study are publicly available in the official repository to ensure reproducibility and open access: [https://github.com/Haoran-Fan/bondformer] (accessed on 2 June 2026).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 7 authors, 5 keywords, 1 funder, 8 references.

Cite

This paper

Fan, H., Qi, H., Huang, X., Zhu, D., Wang, N., Wang, T., & Hao, H. (2026). A Bond-Level Sequence Framework for Molecular Representation Learning with Structural Constraints. Molecules (Basel, Switzerland), 31(11), 1972. https://doi.org/10.3390/molecules31111972

BibTeX

@article{fan2026bond,
author = {Fan, Haoran and Qi, Haoqiang and Huang, Xin and Zhu, Dongyang and Wang, Na and Wang, Ting and Hao, Hongxun},
title = {{A Bond-Level Sequence Framework for Molecular Representation Learning with Structural Constraints}},
journal = {Molecules (Basel, Switzerland)},
year = {2026},
month = jun,
volume = {31},
number = {11},
pages = {1972},
publisher = {Multidisciplinary Digital Publishing Institute (MDPI)},
issn = {1420-3049},
doi = {10.3390/molecules31111972},
url = {https://doi.org/10.3390/molecules31111972},
pmid = {42280272},
pmcid = {PMC13257498}
}

RIS

TY - JOUR
AU - Fan, Haoran
AU - Qi, Haoqiang
AU - Huang, Xin
AU - Zhu, Dongyang
AU - Wang, Na
AU - Wang, Ting
AU - Hao, Hongxun
TI - A Bond-Level Sequence Framework for Molecular Representation Learning with Structural Constraints
T2 - Molecules (Basel, Switzerland)
J2 - Molecules
PY - 2026
DA - 2026/06/05
VL - 31
IS - 11
SP - 1972
SN - 1420-3049
PB - Multidisciplinary Digital Publishing Institute (MDPI)
DO - 10.3390/molecules31111972
UR - https://doi.org/10.3390/molecules31111972
LA - en
ER -

CSL-JSON

{
"id": "10.3390/molecules31111972",
"type": "article-journal",
"title": "A Bond-Level Sequence Framework for Molecular Representation Learning with Structural Constraints",
"container-title": "Molecules (Basel, Switzerland)",
"author": [
{
"family": "Fan",
"given": "Haoran"
},
{
"family": "Qi",
"given": "Haoqiang"
},
{
"family": "Huang",
"given": "Xin"
},
{
"family": "Zhu",
"given": "Dongyang"
},
{
"family": "Wang",
"given": "Na"
},
{
"family": "Wang",
"given": "Ting"
},
{
"family": "Hao",
"given": "Hongxun"
}
],
"container-title-short": "Molecules",
"volume": "31",
"issue": "11",
"page": "1972",
"DOI": "10.3390/molecules31111972",
"PMID": "42280272",
"PMCID": "PMC13257498",
"ISSN": "1420-3049",
"publisher": "Multidisciplinary Digital Publishing Institute (MDPI)",
"URL": "https://doi.org/10.3390/molecules31111972",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
5
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.34133/csbj.0036 [code]
HYG-mol: An Interpretable Multimodal Hypergraph Framework for Molecular Property Prediction.
Journal: Computational and structural biotechnology journal
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular, 2 references
[2] doi:10.1093/bioinformatics/btag153 [code]
MAISNet: a multi-species integrated graph neural network for acetylcholinesterase inhibitor screening.
Journal: Bioinformatics (Oxford, England)
In common: RDKit, PyTorch, scikit-learn, 2 other tools, 1 reference
[3] doi:10.1021/acs.jcim.6c01299 [code]
Targeting BCL-2 through Deep Learning-Based Drug Repurposing: A Multimodal Approach Combining Diffusion-Based Generative Modeling, Neural Relational Inference, and In Vitro Validation.
Journal: Journal of chemical information and modeling
In common: RDKit, PyTorch, pandas, 1 other tool, cellular / molecular, 1 reference
[4] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[5] doi:10.34133/csbj.0184 [code]
Cross-Species Multitask Learning with Molecular and ADME Descriptors for Liver Microsomal Metabolic Stability.
Journal: Computational and structural biotechnology journal
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[6] doi:10.1002/pro.70695 [code]
MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion.
Journal: Protein science : a publication of the Protein Society
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[7] doi:10.3390/ijms27156614 [code]
Candidalysin Inhibits <i>Porphyromonas gingivalis</i> Lipoprotein-Induced IL-1β Production in BV-2 Microglia via Hydrophobic Microbial Interactions.
Journal: International journal of molecular sciences
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[8] doi:10.1038/s41586-026-10670-w [code]
Zero-shot design of drug-binding proteins via neural iterative selection-expansion.
Journal: Nature
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[9] doi:10.1038/s41598-026-53415-5 [code]
Computational design and immunoinformatics validation of a T cell multi-epitope vaccine targeting glioblastoma stem cells.
Journal: Scientific reports
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular
[10] doi:10.1038/s41586-026-10391-0 [code]
Cell-type-targeted mitochondrial transplantation rescues cell degeneration.
Journal: Nature
In common: RDKit, PyTorch, scikit-learn, 2 other tools, cellular / molecular

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.