Pith. sign in

REVIEW 4 major objections 5 minor 53 references

CAML: Commutative algebra machine learning -- a case study on protein-ligand binding affinity prediction

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Commutative algebra descriptors push binding-affinity prediction to R=0.858 on a standard benchmark.

desk verdict CAML introduces a genuinely new descriptor family for affinity prediction, but the central benchmark claim is currently unverifiable because the paper contradicts itself about which PDBbind-v2016 split was used. read the letter →

arxiv 2504.18646 v1 pith:CNSNMRCI submitted 2025-04-25 q-bio.BM

Hongsong Feng , Faisal Suwayyid , Mushal Zia , JunJie Wee , Yuta Hozumi , Chunlong Chen , Guo-Wei Wei This is my paper · ORCID
classification q-bio.BM MSC 13P1055N3192C40
keywords persistentStanley-Reisnertheoryfacetpersistencebarcodescommutativealgebramachinelearningprotein-ligandbindingaffinitymetalloprotein-ligandgradientboostingdecisiontreestopologicaldataanalysisstructure-baseddrugdiscovery
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

The paper proposes commutative algebra machine learning (CAML), a featurization strategy for predicting how tightly small molecules bind to proteins. The central move is to encode a protein-ligand complex as a family of simplicial complexes and track algebraic invariants of the associated Stanley-Reisner rings as a filtration parameter grows, producing descriptors that feed gradient-boosted decision trees. On the PDBbind-v2016 benchmark, the full model—element-specific and category-specific commutative algebra descriptors fused with transformer sequence embeddings—reaches Pearson correlation R=0.858 and RMSE=1.669 kcal/mol; on the metalloprotein-ligand benchmark, the category-specific model reaches R=0.755. These results are presented as surpassing earlier persistent homology and persistent spectral models. If the comparisons hold, commutative algebra contributes a new class of geometric descriptors for structure-based drug discovery.

What carries the argument

The carrying object is the facet persistence barcode from persistent Stanley-Reisner theory. From a point cloud of atom coordinates, the method builds Rips or bipartite simplicial complexes, filters them by interatomic distance, and at each filtration level records the facet prime monomial ideals of the Stanley-Reisner ring; the persistent facet Betti number counts how many facet ideals survive from one level to the next. These survival counts, together with persistent graded Betti numbers via Hochster's formula and persistent f- and h-vectors, form the molecular descriptor. Descriptors are computed per element group, per amino-acid category, and for bipartite protein-ligand and metal-involving interactions, then fed to gradient-boosted decision trees; pretrained protein and small-molecule transformers supply a separate sequence-based model whose consensus with the structural model gives the final CAML prediction.

What would settle it

Run the released code on the PDBbind-v2016 split stated in Section 2.1 (3,768 training, 290 test) and check whether R=0.858 and RMSE=1.669 kcal/mol are reproduced; if the correlation falls below the R=0.848 of the persistent-homology baseline on that split, the claimed state-of-the-art result collapses. Similarly, reproducing R=0.755 on the metalloprotein split behind the JPH-GBT comparison would confirm the metal-ion result.

Watch

Extended reading notes

Core claim

The paper's claim is that persistent Stanley-Reisner theory, the commutative algebra of squarefree monomial ideals determined by the facets of a simplicial complex, can serve as a practical featurization engine for machine learning on biomolecular structures. Tracking how facet ideals, graded Betti numbers, and f- and h-vectors change under a distance filtration yields a vectorized fingerprint of a molecular complex, and the paper introduces three algorithmic variants—bipartite complexes, element-specific atom pairings, and category-specific amino-acid groupings—to capture the physics of protein-ligand and metalloprotein-ligand interactions. Combined with a consensus fusion of transformer-based sequence embeddings, this pipeline is reported to give state-of-the-art binding affinity predictions on both benchmarks.

Load-bearing premise

The load-bearing premise is that the reported numbers were computed on the same benchmark split and with the same metric conventions as the earlier models they are compared to; the paper's own Section 2.1 (3,768 training, 290 test) and Table 3 (1,105 training, 195 test) give conflicting split counts for the protein-ligand benchmark, so if the actual split differs from the baselines' split, the headline R=0.858 is not comparable.

Editorial extensions

If this is right

  • On the standard protein-ligand benchmark (PDBbind-v2016), the full CAML model reaches Pearson R=0.858 and RMSE=1.669 kcal/mol, above the compared persistent homology model TopBP-DL (R=0.848), persistent spectral model PerSpect-ML (R=0.843), and path-spectral model PPS-ML (R=0.840).
  • On the metalloprotein-ligand benchmark, CAML(CS) reaches R=0.755 and CAML(ES) R=0.745, exceeding the compared JPH-GBT model at R=0.742.
  • The element-specific and category-specific variants are independently strong (R=0.836 and R=0.834), their consensus rises to R=0.845, and adding the transformer sequence model lifts the consensus to R=0.858, indicating that structural and sequence information are complementary.
  • Because the descriptors are built directly from atom coordinates and sequences, the same CAML pipeline transfers to other biomolecular property prediction problems without redesigning the featurization.
  • For metalloprotein complexes, the bipartite-complex construction explicitly represents protein-metal and metal-ligand interactions, allowing the model to use 138 atom-pair combination types rather than treating all atoms alike.

Reading between the lines

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

  • If the conflicting split counts in the paper (Section 2.1: 3,768 training / 290 test; Table 3: 1,105 training / 195 test) mean the final model was evaluated on a split different from the baselines', the reported gain over TopBP-DL may not be reproducible on the baseline split; the paper does not reconcile the two counts.
  • A reader cannot tell from the paper alone how much of the 0.858 result comes from the commutative-algebra descriptors versus from the transformer sequence model or from consensus averaging; ablating each channel on the same split would separate those contributions.
  • The bipartite and category-specific constructions should transfer naturally to other two-component biomolecular systems, such as protein-protein, protein-DNA, or protein-RNA complexes, where pairwise interaction categories matter but standard persistent homology treats all atom pairs alike.
  • The persistent facet Betti curves are continuous functions of the filtration; making them differentiable inputs to a deep network is a natural next step that the paper does not explore.
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

4 major / 5 minor

Summary. The paper introduces commutative algebra machine learning (CAML) for protein-ligand and metalloprotein-ligand binding affinity prediction. Building on persistent Stanley-Reisner theory (PSRT), the authors propose three featurization schemes: element-specific commutative algebra, category-specific commutative algebra, and commutative algebra on bipartite complexes. These descriptors are combined with gradient-boosted decision trees and, in the final PDBbind model, with transformer-based sequence embeddings (ESM-2 and a small-molecule transformer). The authors report Pearson correlation R = 0.858 and RMSE = 1.669 kcal/mol on PDBbind-v2016, and R = 0.755 on a metalloprotein-ligand benchmark, claiming state-of-the-art performance against persistent homology and persistent spectral baselines. The manuscript also provides a GitHub link for code and data availability.

Significance. If the reported benchmark results are valid, this paper demonstrates that commutative-algebra descriptors, specifically PSRT-derived facet Betti numbers, add predictive signal for structure-based affinity prediction beyond the compared persistent homology and persistent spectral methods. The work introduces a genuinely new descriptor class and pairs it with established machine learning tools, and it reports averages over 20 independent runs for the main results. However, the central claim is a comparative benchmark claim, and it currently rests on unresolved inconsistencies about the exact test set used for PDBbind-v2016 and about the metalloprotein dataset split. These issues must be resolved before the claimed state-of-the-art status can be accepted.

major comments (4)
  1. [Section 2.1 and Table 3] The test-set definition for the PDBbind-v2016 benchmark is internally contradictory. Section 2.1 states that PDBbind-v2016 has clearly defined training (3,768 complexes) and test sets (290 complexes), but Table 3 lists PDBbind-v2016 as 1,300 total, 1,105 training, and 195 test. These two descriptions cannot describe the same evaluation. If the actual split is the 195-complex one in Table 3, then the reported R = 0.858 is not obtained on the same PDBbind-v2016 test set used by TopBP-DL (R = 0.848), and the headline comparison is not established. The authors must clarify which split was used and, if the 195-complex split is correct, rerun or explicitly compare on the standard 290-complex refined-core test set.
  2. [Section 2.1 and Table 1] There is a numeric inconsistency for the CAML(ES,CS) model. The text in Section 2.1 reports R = 0.853 for this model, while Table 1 lists 0.845. Since the paper describes CAML(ES,CS)+Transformer as a consensus of CAML(ES,CS) and the Transformer model, both the component performance and the final consensus value need to be unambiguous. This discrepancy should be corrected and the reported values verified against the code outputs.
  3. [Section 2.2, Section 3.1, and Table 3] The metalloprotein-ligand benchmark has the same kind of comparability problem. Section 2.2 and Section 3.1 state that the dataset is from reference [40], but Table 3 cites reference [48] for the metalloprotein-ligand row. If the training/test split follows [40] while the JPH-GBT baseline in Table 2 is taken from [48], the comparison to JPH-GBT R = 0.742 may not be apples-to-apples. The authors need to state explicitly which split was used and confirm that the baseline values correspond to that same split.
  4. [Section 2.1 and Table 1] Table 1 reports only average values over 20 runs, without standard deviations or any measure of run-to-run variability, for the PDBbind-v2016 results. The paper's metalloprotein Table 2 includes error bars, so the omission in Table 1 is noticeable. Given that the claimed improvements over TopBP-DL (0.858 vs 0.848), PerSpect-ML (0.843), and PPS-ML (0.840) are small in absolute terms, reporting variability is necessary to assess whether these differences are meaningful.
minor comments (5)
  1. [Throughout] The manuscript contains several typographical errors, including 'Stanely' in the Section 3.2 heading, 'comutative' in the Section 3.3 heading, 'Rensiner' in Section 2.1, 'dscriptors' in Section 3.5, 'o.745' in Section 2.2, and 'we we' in Section 3. These should be corrected.
  2. [Section 3.6] The RMSE formula appears malformed as printed: the display reads sqrt(1/n M sum ...), which is not a well-formed expression. The intended formula is presumably sqrt((1/M) sum (y_e - y_p)^2), and this should be typeset correctly.
  3. [Section 3.4.2] The sentence 'The final molecular descriptors are obtained by first vector among the 256 embedding vectors' is unclear. The authors should specify whether they take the first token embedding, an average over all token embeddings, or some other aggregation.
  4. [Section 3.3] The description of the PCA-based vectorization would benefit from an explicit statement of which features are concatenated for each atom-group point cloud, e.g., facet Betti numbers for 0- and 1-simplices and their rates, and how the filtration parameters (1 to 12 Å, step 0.5 Å) map to the final feature vector length.
  5. [Section 3.5 and Table 5] The consensus weighting between CAML(ES,CS) and the Transformer model is not described. Since the final CAML model is defined as the consensus of these two models, the paper should state how the consensus prediction is computed (e.g., simple averaging of predicted values, weighted averaging, or rank averaging).

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: CAML is an empirical benchmark study; descriptors and labels are external, and no predicted quantity reduces to its own inputs.

full rationale

The paper's central claim is an empirical performance comparison: CAML achieves R=0.858 on PDBbind-v2016 and R=0.755 on the metalloprotein set relative to independently published baselines. The input features are generated from molecular structures through persistent Stanley-Reisner theory as specified in Section 3.2, element-specific and category-specific atom pairings in Section 3.3, and pretrained external NLP models (ESM-2 and the molecular transformer). The regression targets are experimental binding affinities from external benchmark datasets. No equation defines a predicted quantity in terms of the target labels, no fitted parameter is renamed as a prediction, and the self-citations to prior work by the same authors (e.g., [31] for PSRT and [37] for element-specific modeling) are not used as evidence for the headline numerical results; the method itself is described in the paper and tested against external baselines. The internal inconsistencies between the PDBbind-v2016 test-set sizes in Section 2.1 (290) and Table 3 (195), and between the metalloprotein dataset attributions in Section 2.2 ([40]) and Table 3 ([48]), are reproducibility and benchmark-comparability concerns about correctness, not circularity, because they do not make any prediction equal to its input by construction.

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

The central claim rests on standard Stanley-Reisner algebra, on hand-chosen descriptor scales and cutoffs, on a dataset split that is stated inconsistently, and on transfer from pretrained language models. No new physical entities are postulated.

free parameters (4)
  • Filtration range and step for PSRT descriptors = 1-12 Å, step 0.5 Å
    Section 3.3; all persistent facet Betti numbers are computed inside this hand-chosen window, and no sensitivity analysis is reported.
  • Atom cutoff distances = 12 Å for protein atoms, 15 Å for metal atoms
    Section 3.3; the cutoffs determine which atoms enter each point cloud, and changing them changes every descriptor.
  • GBDT hyperparameters = estimators 20,000/30,000; max depth 7; min sample split 5; learning rate 0.002; subsample 0.8
    Section 3.5, Table 5; values are selected by the authors rather than derived from data, and no hyperparameter search or ablation is shown.
  • Consensus weighting between CAML(ES,CS) and Transformer = not stated
    The final CAML prediction is a 'consensus' of two models, but the averaging or weighting rule is never defined; this free choice directly affects the reported R=0.858.
assumptions (4)
  • standard math Hochster's formula and the Hilbert-series identities for Stanley-Reisner rings are valid.
    Section 3.2, equations (8)-(16); standard results used without proof.
  • domain assumption Persistent facet Betti numbers of Rips and bipartite complexes encode binding-relevant interactions.
    Section 3.3; no physical or statistical justification is given beyond analogy to persistent homology.
  • domain assumption The PDBbind-v2016 and metalloprotein-ligand split definitions are exactly as stated.
    Section 3.1; the text and Table 3 disagree on the PDBbind test size, so the evaluation protocol is uncertain.
  • domain assumption Pretrained ESM-2 and molecular transformer embeddings transfer to affinity prediction without leakage.
    Section 3.4; the embeddings are averaged and fed to GBDT, but no leakage or domain-shift analysis is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of CAML: Commutative algebra machine learning -- a case study on protein-ligand binding affinity prediction." pith.science (2026). https://pith.science/paper/CNSNMRCI

@misc{pith2026250418646,
  author       = {Pith},
  title        = {Pith review of: CAML: Commutative algebra machine learning -- a case study on protein-ligand binding affinity prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CNSNMRCI}},
  note         = {Machine review of arXiv:2504.18646}
}
read the original abstract

Recently, Suwayyid and Wei have introduced commutative algebra as an emerging paradigm for machine learning and data science. In this work, we integrate commutative algebra machine learning (CAML) for the prediction of protein-ligand binding affinities. Specifically, we apply persistent Stanley-Reisner theory, a key concept in combinatorial commutative algebra, to the affinity predictions of protein-ligand binding and metalloprotein-ligand binding. We introduce three new algorithms, i.e., element-specific commutative algebra, category-specific commutative algebra, and commutative algebra on bipartite complexes, to address the complexity of data involved in (metallo) protein-ligand complexes. We show that the proposed CAML outperforms other state-of-the-art methods in (metallo) protein-ligand binding affinity predictions.

Figures

Figures reproduced from arXiv: 2504.18646 by the authors.

Figure 1
Figure 1. a. A comparison of the predictions from our CAML model with other published models [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. a. The prediction performance of our CAML models on the metalloprotein-ligand bind [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the filtration process of the persistent commutative algebra. Given a point [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 43 canonical work pages

  1. [40]

    Metalprognet: a structure-based deep graph model for metalloprotein–ligand interaction predictions

    Dejun Jiang, Zhaofeng Ye, Chang-Yu Hsieh, Ziyi Yang, Xujun Zhang, Yu Kang, Hongyan Du, Zhenxing Wu, Jike Wang, Yundian Zeng, et al. Metalprognet: a structure-based deep graph model for metalloprotein–ligand interaction predictions. Chemical Science, 14(8):2054–2069, 2023

    work page 2023

  2. [48]

    Join persistent homology (jph)-based machine learning for metalloprotein–ligand binding affinity prediction

    Yaxing Wang, Xiang Liu, Yipeng Zhang, Xiangjun Wang, and Kelin Xia. Join persistent homology (jph)-based machine learning for metalloprotein–ligand binding affinity prediction. Journal of Chemical Information and Modeling , 2025

    work page 2025

  3. [1]

    Computer-calculated compounds

    Nic Fleming. Computer-calculated compounds. Nature, 557(7707):S55–7, 2018. 15

    work page 2018

  4. [2]

    Ultra-large library docking for discovering new chemotypes

    Jiankun Lyu, Sheng Wang, Trent E Balius, Isha Singh, Anat Levit, Yurii S Moroz, Matthew J O’Meara, Tao Che, Enkhjargal Algaa, Kateryna Tolmachova, et al. Ultra-large library docking for discovering new chemotypes. Nature, 566(7743):224–229, 2019

    work page 2019

  5. [3]

    Docking and scor- ing in virtual screening for drug discovery: methods and applications

    Douglas B Kitchen, H´ el` ene Decornez, John R Furr, and J¨ urgen Bajorath. Docking and scor- ing in virtual screening for drug discovery: methods and applications. Nature reviews Drug discovery, 3(11):935–949, 2004

    work page 2004

  6. [4]

    Molecular docking: shifting paradigms in drug discovery

    Luca Pinzi and Giulio Rastelli. Molecular docking: shifting paradigms in drug discovery. International journal of molecular sciences , 20(18):4331, 2019

    work page 2019

  7. [5]

    Software for molecular dock- ing: a review

    Nataraj S Pagadala, Khajamohiddin Syed, and Jack Tuszynski. Software for molecular dock- ing: a review. Biophysical reviews, 9:91–102, 2017

    work page 2017

  8. [6]

    Accurate and reliable prediction of relative ligand binding potency in prospective drug discovery by way of a modern free-energy calculation protocol and force field

    Lingle Wang, Yujie Wu, Yuqing Deng, Byungchan Kim, Levi Pierce, Goran Krilov, Dmitry Lupyan, Shaughnessy Robinson, Markus K Dahlgren, Jeremy Greenwood, et al. Accurate and reliable prediction of relative ligand binding potency in prospective drug discovery by way of a modern free-energy calculation protocol and force field. Journal of the American Chemica...

    work page 2015

Show all 53 references
  1. [7]

    Computational methods in drug discovery

    Gregory Sliwoski, Sandeepkumar Kothiwale, Jens Meiler, and Edward W Lowe. Computational methods in drug discovery. Pharmacological reviews, 66(1):334–395, 2014

    2014

  2. [8]

    Highly accurate protein structure prediction with alphafold

    John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ron- neberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin ˇZ´ ıdek, Anna Potapenko, et al. Highly accurate protein structure prediction with alphafold. Nature, 596(7873):583–589, 2021

    2021

  3. [9]

    Accurate prediction of protein structures and interactions using a three-track neural network

    Minkyung Baek, Frank DiMaio, Ivan Anishchenko, Justas Dauparas, Sergey Ovchinnikov, Gyu Rie Lee, Jue Wang, Qian Cong, Lisa N Kinch, R Dustin Schaeffer, et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science, 373(6557):871–...

    2021

  4. [10]

    Evolutionary-scale prediction of atomic-level protein structure with a language model

    Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Nikita Smetanin, Robert Verkuil, Ori Kabeli, Yaniv Shmueli, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science, 379(6637):1123–1130, 2023

    2023

  5. [11]

    Multiobjective tree-based reinforcement learning for estimating tol- erant dynamic treatment regimes

    Yao Song and Lu Wang. Multiobjective tree-based reinforcement learning for estimating tol- erant dynamic treatment regimes. Biometrics, 80(1):ujad017, 2024

    2024

  6. [12]

    Challenges and current status of computational methods for docking small molecules to nucleic acids.European journal of medicinal chemistry , 168:414–425, 2019

    Jiaying Luo, Wanlei Wei, J´ erˆ ome Waldisp¨ uhl, and Nicolas Moitessier. Challenges and current status of computational methods for docking small molecules to nucleic acids.European journal of medicinal chemistry , 168:414–425, 2019

    2019

  7. [13]

    Machine learning in chemoin- formatics and drug discovery

    Yu-Chen Lo, Stefano E Rensi, Wen Torng, and Russ B Altman. Machine learning in chemoin- formatics and drug discovery. Drug discovery today, 23(8):1538–1546, 2018

    2018

  8. [14]

    Scientific Reports, 14(1):7526, 2024

    Ai is a viable alternative to high throughput screening: a 318-target study. Scientific Reports, 14(1):7526, 2024

    2024

  9. [15]

    Inactive-enriched machine-learning models exploiting patent data improve structure- based virtual screening for pdl1 dimerizers

    Pablo G´ omez-Sacrist´ an, Saw Simeon, Viet-Khoa Tran-Nguyen, Sachin Patil, and Pedro J Ballester. Inactive-enriched machine-learning models exploiting patent data improve structure- based virtual screening for pdl1 dimerizers. Journal of Advanced Research, 2024. 16

    2024

  10. [16]

    Discovery of novel non-steroidal selective glucocorti- coid receptor modulators by structure-and ign-based virtual screening, structural optimization, and biological evaluation

    Xueping Hu, Jinping Pang, Changwei Chen, Dejun Jiang, Chao Shen, Xin Chai, Liu Yang, Xujun Zhang, Lei Xu, Sunliang Cui, et al. Discovery of novel non-steroidal selective glucocorti- coid receptor modulators by structure-and ign-based virtual screening, structural optimization,...

    2022

  11. [17]

    Machine learning methods for small data challenges in molecular science

    Bozheng Dou, Zailiang Zhu, Ekaterina Merkurjev, Lu Ke, Long Chen, Jian Jiang, Yueying Zhu, Jie Liu, Bengong Zhang, and Guo-Wei Wei. Machine learning methods for small data challenges in molecular science. Chemical Reviews, 123(13):8736–8780, 2023

    2023

  12. [18]

    A review of machine learning methods for imbalanced data challenges in chemistry

    Jian Jiang, Chunhuan Zhang, Lu Ke, Nicole Hayes, Yueying Zhu, Huahai Qiu, Bengong Zhang, Tianshou Zhou, and Guo-Wei Wei. A review of machine learning methods for imbalanced data challenges in chemistry. Chemical Science, 2025

    2025

  13. [19]

    A machine learning approach to predicting protein– ligand binding affinity with applications to molecular docking

    Pedro J Ballester and John BO Mitchell. A machine learning approach to predicting protein– ligand binding affinity with applications to molecular docking. Bioinformatics, 26(9):1169– 1175, 2010

    2010

  14. [20]

    Improving scoring-docking-screening powers of protein– ligand scoring functions using random forest

    Cheng Wang and Yingkai Zhang. Improving scoring-docking-screening powers of protein– ligand scoring functions using random forest. Journal of computational chemistry , 38(3):169– 177, 2017

    2017

  15. [21]

    Development of a polynomial scoring function p3-score for improved scoring and ranking powers

    Chuang Li, Aiwei Zhang, Lifei Wang, Jiaqi Zuo, Caizhen Zhu, Jian Xu, Mingliang Wang, and John ZH Zhang. Development of a polynomial scoring function p3-score for improved scoring and ranking powers. Chemical Physics Letters, 824:140547, 2023

    2023

  16. [22]

    A review of mathematical representations of biomolecular data

    Duc Duy Nguyen, Zixuan Cang, and Guo-Wei Wei. A review of mathematical representations of biomolecular data. Physical Chemistry Chemical Physics , 22(8):4343–4367, 2020

    2020

  17. [23]

    Representability of algebraic topology for biomolecules in machine learning based scoring and virtual screening

    Zixuan Cang, Lin Mu, and Guo-Wei Wei. Representability of algebraic topology for biomolecules in machine learning based scoring and virtual screening. PLoS computational biology, 14(1):e1005929, 2018

    2018

  18. [24]

    Mathematical deep learning for pose and binding affinity prediction and ranking in d3r grand challenges

    Duc Duy Nguyen, Zixuan Cang, Kedi Wu, Menglun Wang, Yin Cao, and Guo-Wei Wei. Mathematical deep learning for pose and binding affinity prediction and ranking in d3r grand challenges. Journal of computer-aided molecular design , 33:71–82, 2019

    2019

  19. [25]

    Persistent spectral graph

    Rui Wang, Duc Duy Nguyen, and Guo-Wei Wei. Persistent spectral graph. International Journal for Numerical Methods in Biomedical Engineering , 36(9):e3376, 2020

    2020

  20. [26]

    Persistent spectral–based machine learning (perspect ml) for protein-ligand binding affinity prediction

    Zhenyu Meng and Kelin Xia. Persistent spectral–based machine learning (perspect ml) for protein-ligand binding affinity prediction. Science advances, 7(19):eabc5329, 2021

    2021

  21. [27]

    Multiscale topology-enabled structure-to-sequence transformer for protein–ligand interaction predictions

    Dong Chen, Jian Liu, and Guo-Wei Wei. Multiscale topology-enabled structure-to-sequence transformer for protein–ligand interaction predictions. Nature Machine Intelligence, 6(7):799– 810, 2024

    2024

  22. [28]

    Mathdl: mathematical deep learning for d3r grand challenge 4

    Duc Duy Nguyen, Kaifu Gao, Menglun Wang, and Guo-Wei Wei. Mathdl: mathematical deep learning for d3r grand challenge 4. Journal of computer-aided molecular design , 34:131–147, 2020

    2020

  23. [29]

    Combinatorial Commutative Algebra, volume 227 ofGraduate Texts in Mathematics

    Ezra Miller and Bernd Sturmfels. Combinatorial Commutative Algebra, volume 227 ofGraduate Texts in Mathematics. Springer, 2005. 17

    2005

  24. [30]

    Commutative algebra: with a view toward algebraic geometry , volume 150

    David Eisenbud. Commutative algebra: with a view toward algebraic geometry , volume 150. Springer Science & Business Media, 2013

    2013

  25. [31]

    Persistent stanley–reisner theory

    Faisal Suwayyid and Guo-Wei Wei. Persistent stanley–reisner theory. arXiv preprint arXiv:2503.23482, 2025

    2025 arXiv

  26. [32]

    Richard P. Stanley. Combinatorics and Commutative Algebra , volume 41 of Progress in Math- ematics. Birkh¨ auser, 2nd edition, 1996

    1996

  27. [33]

    Cohen-Macaulay Rings, volume 39 of Cambridge Studies in Advanced Mathematics

    Winfried Bruns and J¨ urgen Herzog. Cohen-Macaulay Rings, volume 39 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, 1998

    1998

  28. [34]

    Monomial ideals, edge ideals of hypergraphs, and their graded betti numbers

    Huy T` ai H` a and Adam Van Tuyl. Monomial ideals, edge ideals of hypergraphs, and their graded betti numbers. Journal of Algebraic Combinatorics , 27:215–245, 2008

    2008

  29. [35]

    Persistent homology—a survey

    Herbert Edelsbrunner and John Harer. Persistent homology—a survey. In Surveys on Discrete and Computational Geometry: Twenty Years Later, volume 453 ofContemporary Mathematics, pages 257–282. American Mathematical Society, 2008

    2008

  30. [36]

    Computing persistent homology

    Afra Zomorodian and Gunnar Carlsson. Computing persistent homology. Discrete & Compu- tational Geometry, 33(2):249–274, 2005

    2005

  31. [37]

    TopologyNet: Topology based deep convolutional and multi- task neural networks for biomolecular property predictions

    Zixuan Cang and Guo-Wei Wei. TopologyNet: Topology based deep convolutional and multi- task neural networks for biomolecular property predictions. PLoS Computational Biology , 13(7):e1005690, 2017

    2017

  32. [38]

    Position: Topological deep learning is the new frontier for relational learning

    Theodore Papamarkou, Tolga Birdal, Michael Bronstein, Gunnar Carlsson, Justin Curry, Yue Gao, Mustafa Hajij, Roland Kwitt, Pietro Lio, Paolo Di Lorenzo, et al. Position: Topological deep learning is the new frontier for relational learning. arXiv preprint arXiv:2402.08871 , 2024

    2024 arXiv

  33. [39]

    Pdb-wide collection of binding data: current status of the pdbbind database

    Zhihai Liu, Yan Li, Li Han, Jie Li, Jie Liu, Zhixiong Zhao, Wei Nie, Yuchen Liu, and Renxiao Wang. Pdb-wide collection of binding data: current status of the pdbbind database. Bioin- formatics, 31(3):405–412, 2015

    2015

  34. [41]

    Persistent path-spectral (pps) based machine learning for protein–ligand binding affinity prediction

    Ran Liu, Xiang Liu, and Jie Wu. Persistent path-spectral (pps) based machine learning for protein–ligand binding affinity prediction. Journal of Chemical Information and Modeling , 63(3):1066–1075, 2023

    2023

  35. [42]

    Metalloproteomes: a bioinformatic approach

    Claudia Andreini, Ivano Bertini, and Antonio Rosato. Metalloproteomes: a bioinformatic approach. Accounts of chemical research, 42(10):1471–1479, 2009

    2009

  36. [43]

    Metallomics and the cell: some definitions and general com- ments

    Lucia Banci and Ivano Bertini. Metallomics and the cell: some definitions and general com- ments. In Metallomics and the Cell , pages 1–13. Springer, 2012

    2012

  37. [44]

    De novo metalloprotein design

    Matthew J Chalkley, Samuel I Mann, and William F DeGrado. De novo metalloprotein design. Nature Reviews Chemistry, 6(1):31–50, 2022. 18

    2022

  38. [45]

    To what extent do structural changes in catalytic metal sites affect enzyme function? Journal of inorganic biochemistry , 179:40–53, 2018

    Yana Valasatava, Antonio Rosato, Nicholas Furnham, Janet M Thornton, and Claudia An- dreini. To what extent do structural changes in catalytic metal sites affect enzyme function? Journal of inorganic biochemistry , 179:40–53, 2018

    2018

  39. [46]

    Bio-inorganic chemistry, volume 10

    Robert Walker Hay. Bio-inorganic chemistry, volume 10. Ellis Horwood Chichester, 1984

    1984

  40. [47]

    How do bacterial cells ensure that metalloproteins get the correct metal? Nature Reviews Microbiology, 7(1):25–35, 2009

    Kevin J Waldron and Nigel J Robinson. How do bacterial cells ensure that metalloproteins get the correct metal? Nature Reviews Microbiology, 7(1):25–35, 2009

    2009

  41. [49]

    Rosenet: improving binding affin- ity prediction by leveraging molecular mechanics energies with an ensemble of 3d convolutional neural networks

    Hussein Hassan-Harrirou, Ce Zhang, and Thomas Lemmin. Rosenet: improving binding affin- ity prediction by leveraging molecular mechanics energies with an ensemble of 3d convolutional neural networks. Journal of chemical information and modeling , 60(6):2791–2802, 2020

    2020

  42. [50]

    Nnscore 2.0: a neural-network receptor–ligand scoring function

    Jacob D Durrant and J Andrew McCammon. Nnscore 2.0: a neural-network receptor–ligand scoring function. Journal of chemical information and modeling , 51(11):2897–2903, 2011

    2011

  43. [51]

    Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C Lawrence Zitnick, Jerry Ma, et al. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences.Proceedings of the National A...

    2021

  44. [52]

    Extracting predictive representations from hundreds of millions of molecules.The journal of physical chemistry letters, 12(44):10793– 10801, 2021

    Dong Chen, Jiaxin Zheng, Guo-Wei Wei, and Feng Pan. Extracting predictive representations from hundreds of millions of molecules.The journal of physical chemistry letters, 12(44):10793– 10801, 2021

    2021

  45. [53]

    SVSBI: sequence-based virtual screen- ing of biomolecular interactions

    Li Shen, Hongsong Feng, Yuchi Qiu, and Guo-Wei Wei. SVSBI: sequence-based virtual screen- ing of biomolecular interactions. Communications Biology, 6(1):536, 2023. 19

    2023

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.