Pith. sign in

REVIEW 5 major objections 5 minor 57 references

DecoyDB: A Dataset for Graph Contrastive Learning in Protein-Ligand Binding Affinity Prediction

T0 review · 5 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A dataset of 5.4 million docking decoys, paired with a two-part contrastive loss, improves GNN-based protein-ligand binding affinity prediction across accuracy, label efficiency, and generalization.

desk verdict A useful new decoy dataset and a plausible pretraining recipe, but the generalization claim needs a leak check and the hyperparameters look test-set tuned. read the letter →

arxiv 2507.06366 v1 pith:2RXLKOUA submitted 2025-07-08 cs.LG q-bio.BM

Yupu Zhang , Zelin Xu , Tingsong Xiao , Gustavo Seabra , Yanjun Li , Chenglong Li , Zhe Jiang This is my paper · ORCID
classification cs.LGq-bio.BM
keywords protein-ligandbindingaffinitypredictiongraphcontrastivelearningdecoystructuresself-supervisedpretrainingRMSDAutoDockVinadenoisingscorematchingPDBbind
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Protein-ligand binding affinity prediction is bottlenecked by scarce experimental labels: PDBbind has fewer than 20,000 labeled complexes. This paper claims that the bottleneck can be broken with self-supervised pretraining on unlabeled 3D structures, and proposes DecoyDB, a dataset of 61,104 high-resolution complexes augmented with 5,353,307 AutoDock Vina decoys, each annotated with RMSD from the native pose. It also proposes a contrastive pretraining objective with two negative categories: decoys weighted continuously by RMSD and other real complexes as binary negatives, plus a denoising score matching regularizer. Pretraining with this objective lowers GIGN's RMSE from 1.460 to 1.386 on the PDBbind 2013 core set and from 1.263 to 1.188 on the 2016 core set, and the gains are largest when fine-tuning labels are scarce. If true, this gives a practical route to using the far larger pool of unlabeled protein-ligand structures to improve drug-discovery scoring.

What carries the argument

The carrying mechanism is a contrastive pretraining signal built from docking decoys. For each real complex anchor, near-native decoys (RMSD $\le 2$ Å) act as positive pairs; far-from-native decoys are negative pairs whose contribution to the InfoNCE contrastive loss is weighted by normalized RMSD via $\beta_{k,j} = \alpha\, d_{z_k,z_j}/d_{\max}$ in Eq. (2), so worse poses are pushed farther from the anchor; other real complexes supply a second, binary negative category. A denoising score matching term adds Gaussian noise to ligand coordinates and trains the encoder to score the noised structure as if recovering the clean pose, enshrining the assumption that the native pose is an energy minimum. Together these two losses form $L = L_1 + \mu L_2$, and this total objective is what pre-training optimizes before the regression head is fine-tuned on PDBbind labels.

What would settle it

Randomly permute the RMSD labels among decoys before computing the continuous weights in Eq. (2) while keeping everything else fixed; if test RMSE on the PDBbind 2013 and 2016 core sets does not move, the RMSD-based continuous weighting is not the source of the reported gains. A supporting check would measure whether Vina's docking score actually correlates with RMSD across DecoyDB decoys, since the graded negatives are meant to encode binding quality rather than geometry alone.

Watch

Extended reading notes

Core claim

The paper's central claim is that a GNN encoder pre-trained on DecoyDB with this customized contrastive objective predicts binding affinity more accurately, with fewer labels, and on harder leakage-proof splits than the same encoder trained from scratch or with prior pretraining approaches. The claim is supported by consistent RMSE reductions across three base encoders—GIGN, EGNN, and SchNet—and by ablations showing that both the decoy-negative term and the denoising regularizer contribute, with the two-category loss slightly beating the one-category variant. The paper reads the result as evidence that structure-aware decoys, graded by RMSD, carry usable self-supervised signal about binding quality.

Load-bearing premise

The pre-training signal rests on the assumption that AutoDock Vina's generated poses are realistic enough to stand in for true binding and that a decoy's RMSD from the experimentally observed pose reliably tracks how poor its binding is; if a pose far from the native geometry can still bind well, or a pose close to it binds poorly, the contrastive labels mislead the encoder.

Editorial extensions

If this is right

  • Pre-trained models should be preferred over from-scratch training for binding affinity prediction when experimental labels are limited, since the reported gains are largest at 2,000–6,000 fine-tuning labels.
  • DecoyDB can serve as a pretraining corpus that is disjoint from PDBbind fine-tuning data, removing the leakage that previous decoy-based pretraining built from PDBbind complexes suffered.
  • The continuous RMSD weighting matters: replacing the two-category loss with a one-category loss raises RMSE slightly on both core sets, so grading negatives by deviation adds signal.
  • The DSM regularization term is not redundant with contrastive learning; removing it degrades performance, so both objectives should be kept in the pretraining loss.
  • Generalization to a leakage-proof split improves more than to a random split, implying the pretraining learns transferable interaction structure rather than memorizing benchmark biases.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because DecoyDB annotates every decoy with RMSD but not with an independent energy estimate, a natural untested extension is to re-weight negatives by a consensus of docking scores or by an energy function rather than by RMSD alone.
  • The dataset's scale (5.35M decoys over 61K complexes) could support other structure-based tasks such as pose ranking or docking-scoring benchmarks, although the experiments here only evaluate affinity prediction.
  • The improvement pattern suggests the method might transfer to other GNN architectures beyond the three tested, since the loss operates on embeddings rather than on architecture-specific features; this remains untested.
  • A specific testable extension is to shuffle decoy RMSD labels during pretraining: if core-set RMSE does not change, the continuous weighting is not the active ingredient.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper introduces DecoyDB, a large-scale dataset of 61,104 protein-ligand complexes drawn from the PDB and 5,353,307 AutoDock Vina-generated decoy poses annotated with RMSD to the native ligand pose. It also proposes a graph contrastive pretraining framework that combines a two-category contrastive loss (with continuous negative weighting based on decoy RMSD and discrete negatives from other real complexes) with a denoising score matching (DSM) regularization. The pretrained encoders (GIGN, EGNN, SchNet, TorchMD-Net) are fine-tuned on PDBbind and evaluated on the PDBbind 2013/2016 coresets and the leakage-proof LP-PDBbind split. The paper claims consistent improvements in prediction accuracy, label efficiency, and generalizability from DecoyDB pretraining.

Significance. The dataset is a potentially valuable community resource: it is substantially larger than existing decoy-augmented pretraining sets, includes continuous RMSD annotations that enable finer-grained contrastive supervision, and is released with code and data. If the evaluation is demonstrated to be leakage-free and the loss functions are correctly specified, the reported gains across multiple GNN backbones and the label-efficiency results would constitute a useful contribution to structure-based binding-affinity prediction. However, the generalization claim currently rests on an unverified overlap assumption for LP-PDBbind, and several central equations are not correctly specified as written, so the results cannot be fully assessed in the present form.

major comments (5)
  1. [§5.5 and §5 (Dataset description)] The generalization claim in Table 4 rests entirely on the LP-PDBbind split, but the manuscript never states that LP-PDBbind complexes (or the decoys derived from them) were excluded from DecoyDB pretraining. The dataset description only reports removing overlapping samples from the PDBbind2016/2013 fine-tuning sets and removing the two coresets from pretraining. Since DecoyDB is built from PDB structures and LP-PDBbind is a reorganized PDBbind-derived split, structural overlap is plausible; if any LP-PDBbind test complex appears in DecoyDB, the Table 4 improvement (1.496 to 1.371) could reflect memorization rather than generalization. Please provide an explicit overlap analysis between DecoyDB and the LP-PDBbind train/validation/test splits, and remove any overlapping complexes from pretraining before reporting the generalization numbers.
  2. [§5.2, Figure 3] Figure 3 and the surrounding text report RMSE 'across the test datasets' while varying hyperparameters α and μ over the range 0.4 to 2.0. If α and μ were chosen by minimizing test-set RMSE, then the final Table 2 results for OURS are not independent test evaluations, and the comparison against baselines is unfair because the baselines are not similarly selected on the test coresets. The authors must state clearly whether α and μ were selected on the validation split (the 1,000-complex held-out set) or on the test coresets; if the latter, the experimental protocol must be revised, for example by selecting on validation and re-reporting the test numbers.
  3. [§4.2, Eq. (2)] Equation (2) defines β_{k,j} using d_{z_k,z_j}, which the text describes as 'the RMSD between embeddings z_k and z_j.' However, the RMSD computed in §4.1 is a spatial distance between ligand poses in input space, and it has no meaning for embedding vectors. The continuous negative weighting should reference the decoy's RMSD to the native pose (for example, d_{s_k,s_j} in input space), not an embedding distance. As written, the equation is not implementable and conflates input-space geometry with embedding-space geometry.
  4. [§4.2, Eq. (4)] Equation (4) defines the denoising score matching loss as the expectation of the vector difference ∂log f(s′)/∂x′ − (x − x′)/σ², with no squared norm. This is a vector-valued quantity and cannot serve as a scalar loss; standard DSM is E[||∂log f(s′)/∂x′ − (x − x′)/σ²||²]. The missing squared L2 norm is a mathematical error in a central component of the proposed method. Please correct the equation and confirm that the implementation matches the corrected form.
  5. [§4.2, Eq. (1)] Equation (1) is not the InfoNCE loss as claimed: the denominator sums over j ≠ i, excluding the positive sample i, whereas standard InfoNCE includes the positive sample in the denominator to normalize the softmax over all samples. Additionally, the expression 'sim(z_k,z_i^T/τ)' is dimensionally ill-formed; it should presumably be sim(z_k,z_i)/τ. Please restate the loss precisely and clarify whether the implementation follows the equation as written or the standard InfoNCE form, since this changes the training objective.
minor comments (5)
  1. [§4.2, Eq. (1)] The superscript T on z_i in Eq. (1) is misplaced; the similarity should be written as sim(z_k, z_i)/τ.
  2. [§5, Dataset description] There is a duplicated word in 'by by randomly selecting'; it should read 'by randomly selecting.'
  3. [Figure 3] The caption states that panels (a) and (c) show μ while panels (b) and (d) show α, but the panel labels and axis values could be misaligned; please clarify which panel corresponds to which hyperparameter and which dataset.
  4. [§4.1] The RMSD calculation is described only as 'the spatial distances between the atoms of the decoy’s ligand and the corresponding atoms in the original ligand'; please specify whether this is computed over heavy atoms, how symmetry/atom-mapping ambiguities are handled, and whether any alignment is performed before the RMSD is calculated.
  5. [Appendix B] The description of ConBAP states that it uses binary negative samples from decoys, but no ablation is reported that replaces the proposed continuous RMSD weighting with binary (unweighted) decoy negatives while keeping the same dataset; such an ablation would isolate the benefit of continuous weighting.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: pretraining and fine-tuning are empirically separated; self-citations are background only, and the LP-PDBbind and Vina concerns are evaluation risks, not circularity.

full rationale

This is an empirical systems paper rather than a derivation, and I cannot exhibit any load-bearing step in which a claimed prediction reduces to its own input. The pretraining signal is defined by AutoDock Vina decoys annotated with RMSD, and the contrastive loss in Equations (1)-(3) uses those RMSD values as continuous weights. That is a training objective, not a fitted parameter renamed as a prediction: the reported results are RMSE and Pearson correlation on external PDBbind core sets and on the LP-PDBbind split, none of which is an algebraic function of the contrastive loss or of the RMSD annotations. The DSM regularizer in Equation (4) is imported from prior work and is an auxiliary loss; nothing in the paper reduces the fine-tuning target to the DSM objective. The self-citations that exist (e.g., Rezaei et al. 2022 and Li et al. 2022b, which include co-authors) appear only as background context and are not used to justify the central accuracy, label-efficiency, or generalizability claims. The most serious concerns in the paper are not circularity. First, the LP-PDBbind generalization experiment may be leaky: the paper states that overlapping samples were removed from fine-tuning datasets and that the PDBbind2013/2016 coresets were removed from pretraining, but it never states that LP-PDBbind test complexes were excluded from DecoyDB pretraining. If such complexes were present, the Table 4 result would reflect memorization rather than generalization. That is a data-hygiene/evaluation-validity risk, not a self-referential derivation. Second, the decoy labels come from AutoDock Vina, whose scoring function is itself trained on data overlapping PDBbind; this is a legitimate concern about the independence of the pretraining signal, but it is a concern about external validity, not about the paper's argument being circular by construction. Because the test sets are external, the core benchmarks are self-contained, and the minor self-citations are not load-bearing, the appropriate circularity score is low.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the validity of computationally generated decoys as training signal, the choice of RMSD as a continuous negative proxy, several unreported hyperparameter values, and the assumption that overlap removal fully prevents leakage. No new physical entities are introduced.

free parameters (4)
  • alpha (α) = not reported; sensitivity sweep 0.4-2.0
    Balances decoy negatives vs other-complex negatives in β (Eq. 2); appears selected by test set RMSE (Fig. 3).
  • mu (μ) = not reported; sensitivity sweep 0.4-2.0
    Weights the DSM regularization in the total loss (Eq. 5); appears selected by test set RMSE (Fig. 3).
  • temperature (τ) = not specified
    Temperature in InfoNCE loss (Eq. 1); value not reported in the paper.
  • noise scale (σ) = not specified
    Variance of Gaussian noise in the DSM regularization (Eq. 4); value not reported.
assumptions (5)
  • domain assumption AutoDock Vina with exhaustiveness 8 generates chemically plausible, diverse binding poses; RMSD to native pose is a valid continuous measure of binding quality.
    Underpins the entire positive/negative pair construction (§4.1, §4.2 Eq. 2). If false, pretraining signal is misleading.
  • domain assumption The native crystallized complex is the minimum-energy binding pose, so denoising ligand coordinates yields a useful structural prior.
    Motivates the DSM regularization in §4.2 (Eq. 4).
  • domain assumption Distance thresholds (5 Å for interaction edges, 10 Å for pocket atoms) capture biochemically relevant interactions.
    Graph construction in §4.1 and Appendix A.2.
  • domain assumption Removing samples overlapping with the fine-tuning datasets eliminates leakage between pretraining and downstream evaluation.
    Stated in §5; relies on exact ID matching, not sequence or structural similarity, so hidden homology may remain.
  • standard math Standard supervised and contrastive training assumptions: optimizer convergence, early stopping, and five-run averages give stable estimates.
    Appendix A.3; standard practice though not guaranteed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DecoyDB: A Dataset for Graph Contrastive Learning in Protein-Ligand Binding Affinity Prediction." pith.science (2026). https://pith.science/paper/2RXLKOUA

@misc{pith2026250706366,
  author       = {Pith},
  title        = {Pith review of: DecoyDB: A Dataset for Graph Contrastive Learning in Protein-Ligand Binding Affinity Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2RXLKOUA}},
  note         = {Machine review of arXiv:2507.06366}
}
read the original abstract

Predicting the binding affinity of protein-ligand complexes plays a vital role in drug discovery. Unfortunately, progress has been hindered by the lack of large-scale and high-quality binding affinity labels. The widely used PDBbind dataset has fewer than 20K labeled complexes. Self-supervised learning, especially graph contrastive learning (GCL), provides a unique opportunity to break the barrier by pre-training graph neural network models based on vast unlabeled complexes and fine-tuning the models on much fewer labeled complexes. However, the problem faces unique challenges, including a lack of a comprehensive unlabeled dataset with well-defined positive/negative complex pairs and the need to design GCL algorithms that incorporate the unique characteristics of such data. To fill the gap, we propose DecoyDB, a large-scale, structure-aware dataset specifically designed for self-supervised GCL on protein-ligand complexes. DecoyDB consists of high-resolution ground truth complexes (less than 2.5 Angstrom) and diverse decoy structures with computationally generated binding poses that range from realistic to suboptimal (negative pairs). Each decoy is annotated with a Root Mean Squared Deviation (RMSD) from the native pose. We further design a customized GCL framework to pre-train graph neural networks based on DecoyDB and fine-tune the models with labels from PDBbind. Extensive experiments confirm that models pre-trained with DecoyDB achieve superior accuracy, label efficiency, and generalizability.

Figures

Figures reproduced from arXiv: 2507.06366 by the authors.

Figure 1
Figure 1. The data construction pipeline of DecoyDB. 0 5000 10000 15000 20000 25000 Number of Atoms 0.0 0.5 1.0 1.5 2.0 Number of Complexes ×10 4 (a) 0 20 40 60 80 100 Number of Decoys 0 1 2 3 4 Number of Complexes ×10 4 (b) 0 5 10 15 20 25 RMSD 0 2 4 6 8 Number of Complexes ×10 5 (c) 0 5 10 15 20 25 RMSD 0.0 0.2 0.4 0.6 0.8 1.0 Cumulative Frequency (d) [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Statistical analysis of DecoyDB. (a) Distribution of the number of atoms in each protein-ligand complex. (b) Distribution of the number of decoys per complex. (c) Distribution of RMSD values for decoy complexes. (d) Cumulative distribution of RMSD values for decoy complexes. complex), we defined a grid box around the ligand with a 5Å padding in each dimension to ensure sufficient space for ligand flexibility during … view at source ↗
Figure 3
Figure 3. Sensitivity analysis of two key hyper-parameters, [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: (a) is the training and validation loss curve during pre-training. (b) is the ablation study. (c) is the [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 52 canonical work pages

  1. [1]

    Berman, John Westbrook, Zukang Feng, Gary Gilliland, T

    Helen M. Berman, John Westbrook, Zukang Feng, Gary Gilliland, T. N. Bhat, Helge Weissig, Ilya N. Shindyalov, and Philip E. Bourne. The protein data bank. Nucleic Acids Research , 28(1):235--242, 01 2000

    work page 2000

  2. [2]

    A simple framework for contrastive learning of visual representations

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual representations. In International conference on machine learning , pages 1597--1607. PMLR, 2020

    2020

  3. [3]

    Comprehensive analysis of kinase inhibitor selectivity

    Mindy I Davis, Jeremy P Hunt, Sanna Herrgard, Pietro Ciceri, Lisa M Wodicka, Gabriel Pallares, Michael Hocker, Daniel K Treiber, and Patrick P Zarrinkar. Comprehensive analysis of kinase inhibitor selectivity. Nature biotechnology , 29(11):1046--1051, 2011

    work page 2011

  4. [4]

    Autodock vina 1.2

    Jerome Eberhardt, Diogo Santos-Martins, Andreas F Tillack, and Stefano Forli. Autodock vina 1.2. 0: New docking methods, expanded force field, and python bindings. Journal of chemical information and modeling , 61(8):3891--3898, 2021

    work page 2021

  5. [5]

    Knowledge graph-enhanced molecular contrastive learning with functional prompt

    Yin Fang, Qiang Zhang, Ningyu Zhang, Zhuo Chen, Xiang Zhuang, Xin Shao, Xiaohui Fan, and Huajun Chen. Knowledge graph-enhanced molecular contrastive learning with functional prompt. Nature Machine Intelligence , 5(5):542--553, 2023

    work page 2023

  6. [6]

    Potentialnet for molecular property prediction

    Evan N Feinberg, Debnil Sur, Zhenqin Wu, Brooke E Husic, Huanghao Mai, Yang Li, Saisai Sun, Jianyi Yang, Bharath Ramsundar, and Vijay S Pande. Potentialnet for molecular property prediction. ACS central science , 4(11):1520--1530, 2018

    work page 2018

  7. [7]

    Three-dimensional convolutional neural networks and a cross-docked data set for structure-based drug design

    Paul G Francoeur, Tomohide Masuda, Jocelyn Sunseri, Andrew Jia, Richard B Iovanisci, Ian Snyder, and David R Koes. Three-dimensional convolutional neural networks and a cross-docked data set for structure-based drug design. Journal of chemical information and modeling , 60(9):4200--4215, 2020

    work page 2020

  8. [8]

    The mm/pbsa and mm/gbsa methods to estimate ligand-binding affinities

    Samuel Genheden and Ulf Ryde. The mm/pbsa and mm/gbsa methods to estimate ligand-binding affinities. Expert opinion on drug discovery , 10(5):449--461, 2015

    work page 2015

Show all 57 references
  1. [9]

    T-mgcl: Molecule graph contrastive learning based on transformer for molecular property prediction

    Xiaoyu Guan and Daoqiang Zhang. T-mgcl: Molecule graph contrastive learning based on transformer for molecular property prediction. IEEE/ACM Transactions on Computational Biology and Bioinformatics , 2023

    2023

  2. [10]

    Momentum contrast for unsupervised visual representation learning

    Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. Momentum contrast for unsupervised visual representation learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 9729--9738, 2020

    2020

  3. [11]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll \'a r, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 16000--16009, 2022

    2022

  4. [12]

    Interactiongraphnet: A novel and efficient deep graph representation learning framework for accurate protein–ligand interaction predictions

    Dejun Jiang, Chang-Yu Hsieh, Zhenxing Wu, Yu Kang, Jike Wang, Ercheng Wang, Ben Liao, Chao Shen, Lei Xu, Jian Wu, Dongsheng Cao, and Tingjun Hou. Interactiongraphnet: A novel and efficient deep graph representation learning framework for accurate protein–ligand interaction pre...

    2021

  5. [13]

    Unsupervised protein-ligand binding energy prediction via neural euler's rotation equation

    Wengong Jin, Siranush Sarkizova, Xun Chen, Nir Hacohen, and Caroline Uhler. Unsupervised protein-ligand binding energy prediction via neural euler's rotation equation. Advances in Neural Information Processing Systems , 36, 2024

    2024

  6. [14]

    Statistical mechanics of fluid mixtures

    John G Kirkwood. Statistical mechanics of fluid mixtures. The Journal of chemical physics , 3(5):300--313, 1935

    1935

  7. [15]

    Docking and scoring in virtual screening for drug discovery: methods and applications

    Douglas B Kitchen, H \'e l \`e ne Decornez, John R Furr, and J \"u rgen Bajorath. Docking and scoring in virtual screening for drug discovery: methods and applications. Nature reviews Drug discovery , 3(11):935--949, 2004

    2004

  8. [16]

    Modelling drug-target binding affinity using a bert based graph neural network

    Mark Lennox, Neil Robertson, and Barry Devereux. Modelling drug-target binding affinity using a bert based graph neural network. Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual In...

    2021

  9. [17]

    Y. Li, M. A. Rezaei, C. Li, and X. Li. Deepatom: A framework for protein-ligand binding affinity prediction. In 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) , pages 303--310, nov 2019

    2019

  10. [18]

    Machine-learning scoring functions for structure-based drug lead optimization

    Hongjian Li, Kam-Heung Sze, Gang Lu, and Pedro J Ballester. Machine-learning scoring functions for structure-based drug lead optimization. Wiley Interdisciplinary Reviews: Computational Molecular Science , 10(5):e1465, 2020

    2020

  11. [19]

    Geomgcl: Geometric graph contrastive learning for molecular property prediction

    Shuangli Li, Jingbo Zhou, Tong Xu, Dejing Dou, and Hui Xiong. Geomgcl: Geometric graph contrastive learning for molecular property prediction. In Proceedings of the AAAI conference on artificial intelligence , volume 36, pages 4541--4549, 2022

    2022

  12. [20]

    Dyscore: A boosting scoring method with dynamic properties for identifying true binders and nonbinders in structure-based drug discovery

    Yanjun Li, Daohong Zhou, Guangrong Zheng, Xiaolin Li, Dapeng Wu, and Yaxia Yuan. Dyscore: A boosting scoring method with dynamic properties for identifying true binders and nonbinders in structure-based drug discovery. Journal of chemical information and modeling , 62(22):5550...

    2022

  13. [21]

    Leak proof pdbbind: a reorganized dataset of protein-ligand complexes for more generalizable binding affinity prediction

    Jie Li, Xingyi Guan, Oufan Zhang, Kunyang Sun, Yingze Wang, Dorian Bagni, and Teresa Head-Gordon. Leak proof pdbbind: a reorganized dataset of protein-ligand complexes for more generalizable binding affinity prediction. ArXiv , pages arXiv--2308, 2024

    2024

  14. [22]

    A high-quality data set of protein--ligand binding interactions via comparative complex structure modeling

    Xuelian Li, Cheng Shen, Hui Zhu, Yujian Yang, Qing Wang, Jincai Yang, and Niu Huang. A high-quality data set of protein--ligand binding interactions via comparative complex structure modeling. Journal of Chemical Information and Modeling , 64(7):2454--2466, 2024

    2024

  15. [23]

    Predicting drug–target interaction using a novel graph neural network with 3d structure-embedded graph representation

    Jaechang Lim, Seongok Ryu, Kyubyong Park, Yo Joong Choe, Jiyeon Ham, and Woo Youn Kim. Predicting drug–target interaction using a novel graph neural network with 3d structure-embedded graph representation. Journal of Chemical Information and Modeling , 59(9):3981--3988, 2019. ...

    2019

  16. [24]

    Bindingdb: a web-accessible database of experimentally determined protein--ligand binding affinities

    Tiqing Liu, Yuhmei Lin, Xin Wen, Robert N Jorissen, and Michael K Gilson. Bindingdb: a web-accessible database of experimentally determined protein--ligand binding affinities. Nucleic acids research , 35(suppl\_1):D198--D201, 2007

    2007

  17. [25]

    Hiermrl: Hierarchical structure-aware molecular representation learning for property prediction

    Maotao Liu, Yifan Yang, Xu Gong, Li Liu, and Qun Liu. Hiermrl: Hierarchical structure-aware molecular representation learning for property prediction. In 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) , pages 386--389. IEEE, 2022

    2022

  18. [26]

    Graphprompt: Unifying pre-training and downstream tasks for graph neural networks

    Zemin Liu, Xingtong Yu, Yuan Fang, and Xinming Zhang. Graphprompt: Unifying pre-training and downstream tasks for graph neural networks. In Proceedings of the ACM web conference 2023 , pages 417--428, 2023

    2023

  19. [27]

    Enhancing generalizability in protein--ligand binding affinity prediction with multimodal contrastive learning

    Ding Luo, Dandan Liu, Xiaoyang Qu, Lina Dong, and Binju Wang. Enhancing generalizability in protein--ligand binding affinity prediction with multimodal contrastive learning. Journal of Chemical Information and Modeling , 64(6):1892--1906, 2024

    1906

  20. [28]

    Dockingapp rf: a state-of-the-art novel scoring function for molecular docking in a user-friendly interface to autodock vina

    Gabriele Macari, Daniele Toti, Andrea Pasquadibisceglie, and Fabio Polticelli. Dockingapp rf: a state-of-the-art novel scoring function for molecular docking in a user-friendly interface to autodock vina. International Journal of Molecular Sciences , 21(24):9548, 2020

    2020

  21. [29]

    Pubchemqc project: a large-scale first-principles electronic structure database for data-driven chemistry

    Maho Nakata and Tomomi Shimazaki. Pubchemqc project: a large-scale first-principles electronic structure database for data-driven chemistry. Journal of chemical information and modeling , 57(6):1300--1308, 2017

    2017

  22. [30]

    GraphDTA: predicting drug–target binding affinity with graph neural networks

    Thin Nguyen, Hang Le, Thomas P Quinn, Tri Nguyen, Thuc Duy Le, and Svetha Venkatesh. GraphDTA: predicting drug–target binding affinity with graph neural networks . Bioinformatics , 37(8):1140--1147, 10 2020

    2020

  23. [31]

    Pre-training with fractional denoising to enhance molecular property prediction

    Yuyan Ni, Shikun Feng, Xin Hong, Yuancheng Sun, Wei-Ying Ma, Zhi-Ming Ma, Qiwei Ye, and Yanyan Lan. Pre-training with fractional denoising to enhance molecular property prediction. Nature Machine Intelligence , 6(10):1169--1178, 2024

    2024

  24. [32]

    A probability contrastive learning framework for 3d molecular representation learning

    Jiayu Qin, Jian Chen, Rohan Sharma, Jingchen Sun, and Changyou Chen. A probability contrastive learning framework for 3d molecular representation learning. In The Thirty-eighth Annual Conference on Neural Information Processing Systems , 2024

    2024

  25. [33]

    Improving language understanding by generative pre-training

    Alec Radford, Karthik Narasimhan, Tim Salimans, Ilya Sutskever, et al. Improving language understanding by generative pre-training. 2018

    2018

  26. [34]

    Deep learning in drug design: Protein-ligand binding affinity prediction

    Mohammad A Rezaei, Yanjun Li, Dapeng Wu, Xiaolin Li, and Chenglong Li. Deep learning in drug design: Protein-ligand binding affinity prediction. IEEE/ACM Transactions on Computational Biology and Bioinformatics , 19(01):407--417, 2022

    2022

  27. [35]

    E (n) equivariant graph neural networks

    V ctor Garcia Satorras, Emiel Hoogeboom, and Max Welling. E (n) equivariant graph neural networks. In International conference on machine learning , pages 9323--9332. PMLR, 2021

    2021

  28. [36]

    u tt, Pieter-Jan Kindermans, Huziel Enoc Sauceda Felix, Stefan Chmiela, Alexandre Tkatchenko, and Klaus-Robert M\

    Kristof Sch\" u tt, Pieter-Jan Kindermans, Huziel Enoc Sauceda Felix, Stefan Chmiela, Alexandre Tkatchenko, and Klaus-Robert M\" u ller. Schnet: A continuous-filter convolutional neural network for modeling quantum interactions. In I. Guyon, U. Von Luxburg, S. Bengio, H. Walla...

    2017

  29. [37]

    Misato: machine learning dataset of protein--ligand complexes for structure-based drug discovery

    Till Siebenmorgen, Filipe Menezes, Sabrina Benassou, Erinc Merdivan, Kieran Didi, Andr \'e Santos Dias Mour \ a o, Rados aw Kitel, Pietro Li \`o , Stefan Kesselheim, Marie Piraud, et al. Misato: machine learning dataset of protein--ligand complexes for structure-based drug dis...

    2024

  30. [38]

    Development and evaluation of a deep learning model for protein--ligand binding affinity prediction

    Marta M Stepniewska-Dziubinska, Piotr Zielenkiewicz, and Pawel Siedlecki. Development and evaluation of a deep learning model for protein--ligand binding affinity prediction. Bioinformatics , 34(21):3666--3674, 2018

    2018

  31. [39]

    Development and evaluation of a deep learning model for protein–ligand binding affinity prediction

    Marta M Stepniewska-Dziubinska, Piotr Zielenkiewicz, and Pawel Siedlecki. Development and evaluation of a deep learning model for protein–ligand binding affinity prediction . Bioinformatics , 34(21):3666--3674, 05 2018

    2018

  32. [40]

    Mocl: data-driven molecular fingerprint via knowledge-aware contrastive learning from molecular graph

    Mengying Sun, Jing Xing, Huijun Wang, Bin Chen, and Jiayu Zhou. Mocl: data-driven molecular fingerprint via knowledge-aware contrastive learning from molecular graph. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining , pages 3585--3594, 2021

    2021

  33. [41]

    Making sense of large-scale kinase inhibitor bioactivity data sets: a comparative and integrative analysis

    Jing Tang, Agnieszka Szwajda, Sushil Shakyawar, Tao Xu, Petteri Hintsanen, Krister Wennerberg, and Tero Aittokallio. Making sense of large-scale kinase inhibitor bioactivity data sets: a comparative and integrative analysis. Journal of chemical information and modeling , 54(3)...

    2014

  34. [42]

    Equivariant transformers for neural network based molecular potentials

    Philipp Th \"o lke and Gianni De Fabritiis. Equivariant transformers for neural network based molecular potentials. In International Conference on Learning Representations , 2022

    2022

  35. [43]

    Directed graph contrastive learning

    Zekun Tong, Yuxuan Liang, Henghui Ding, Yongxing Dai, Xinke Li, and Changhu Wang. Directed graph contrastive learning. Advances in neural information processing systems , 34:19580--19593, 2021

    2021

  36. [44]

    Sunsetting binding moad with its last data update and the addition of 3d-ligand polypharmacology tools

    Swapnil Wagle, Richard D Smith, Anthony J Dominic III, Debarati DasGupta, Sunil Kumar Tripathi, and Heather A Carlson. Sunsetting binding moad with its last data update and the addition of 3d-ligand polypharmacology tools. Scientific Reports , 13(1):3008, 2023

    2023

  37. [45]

    Contrastive learning with stronger augmentations

    Xiao Wang and Guo-Jun Qi. Contrastive learning with stronger augmentations. IEEE transactions on pattern analysis and machine intelligence , 45(5):5549--5560, 2022

    2022

  38. [46]

    The pdbbind database: methodologies and updates

    Renxiao Wang, Xueliang Fang, Yipin Lu, Chao-Yie Yang, and Shaomeng Wang. The pdbbind database: methodologies and updates. Journal of medicinal chemistry , 48(12):4111--4119, 2005

    2005

  39. [47]

    A workflow to create a high-quality protein--ligand binding dataset for training, validation, and prediction tasks

    Yingze Wang, Kunyang Sun, Jie Li, Xingyi Guan, Oufan Zhang, Dorian Bagni, Yang Zhang, Heather A Carlson, and Teresa Head-Gordon. A workflow to create a high-quality protein--ligand binding dataset for training, validation, and prediction tasks. Digital Discovery , 2025

    2025

  40. [48]

    Pre-training of equivariant graph matching networks with conformation flexibility for drug binding

    Fang Wu, Shuting Jin, Yinghui Jiang, Xurui Jin, Bowen Tang, Zhangming Niu, Xiangrong Liu, Qiang Zhang, Xiangxiang Zeng, and Stan Z Li. Pre-training of equivariant graph matching networks with conformation flexibility for drug binding. Advanced Science , 9(33):2203796, 2022

    2022

  41. [49]

    Infogcl: Information-aware graph contrastive learning

    Dongkuan Xu, Wei Cheng, Dongsheng Luo, Haifeng Chen, and Xiang Zhang. Infogcl: Information-aware graph contrastive learning. Advances in Neural Information Processing Systems , 34:30414--30425, 2021

    2021

  42. [50]

    Biolip: a semi-manually curated database for biologically relevant ligand--protein interactions

    Jianyi Yang, Ambrish Roy, and Yang Zhang. Biolip: a semi-manually curated database for biologically relevant ligand--protein interactions. Nucleic acids research , 41(D1):D1096--D1103, 2012

    2012

  43. [51]

    Geometric interaction graph neural network for predicting protein--ligand binding affinities from 3d structures (gign)

    Ziduo Yang, Weihe Zhong, Qiujie Lv, Tiejun Dong, and Calvin Yu-Chian Chen. Geometric interaction graph neural network for predicting protein--ligand binding affinities from 3d structures (gign). The journal of physical chemistry letters , 14(8):2020--2033, 2023

    2020

  44. [52]

    Graph contrastive learning with augmentations

    Yuning You, Tianlong Chen, Yongduo Sui, Ting Chen, Zhangyang Wang, and Yang Shen. Graph contrastive learning with augmentations. Advances in neural information processing systems , 33:5812--5823, 2020

    2020

  45. [53]

    Pre-training via denoising for molecular property prediction

    Sheheryar Zaidi, Michael Schaarschmidt, James Martens, Hyunjik Kim, Yee Whye Teh, Alvaro Sanchez-Gonzalez, Peter Battaglia, Razvan Pascanu, and Jonathan Godwin. Pre-training via denoising for molecular property prediction. In The Eleventh International Conference on Learning R...

    2023

  46. [54]

    Machine learning for sequence and structure-based protein--ligand interaction prediction

    Yunjiang Zhang, Shuyuan Li, Kong Meng, and Shaorui Sun. Machine learning for sequence and structure-based protein--ligand interaction prediction. Journal of Chemical Information and Modeling , 64(5):1456--1472, 2024

    2024

  47. [55]

    Onionnet: a multiple-layer intermolecular-contact-based convolutional neural network for protein--ligand binding affinity prediction

    Liangzhen Zheng, Jingrong Fan, and Yuguang Mu. Onionnet: a multiple-layer intermolecular-contact-based convolutional neural network for protein--ligand binding affinity prediction. ACS omega , 4(14):15956--15965, 2019

    2019

  48. [56]

    High-temperature equation of state by a perturbation method

    Robert W Zwanzig. High-temperature equation of state by a perturbation method. i. nonpolar gases. The Journal of Chemical Physics , 22(8):1420--1426, 1954

    1954

  49. [57]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence '...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.