Pith. sign in

REVIEW 4 major objections 7 minor 48 references

The de Rham-Hodge analysis and modeling of biomolecules

T0 review · 4 major / 7 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper proposes de Rham-Hodge spectral analysis as a unified framework for biomolecular geometry, topology, flexibility, and natural motion, reporting 0.580 average Pearson correlation for B-factor prediction across 364 proteins.

desk verdict A genuine new application of de Rham-Hodge theory to biomolecules with a real DEC implementation, but the B-factor and natural-mode claims rest on assumptions stated without derivation and on thin empirical validation. read the letter →

arxiv 1908.00572 v1 pith:G46CA2WR submitted 2019-08-01 q-bio.BM math.ATmath.DG

Rundong Zhao , Menglun Wang , Yiying Tong , Guo-Wei Wei This is my paper · ORCID
classification q-bio.BMmath.ATmath.DG MSC 58A1458J5092C40
keywords deRham-HodgetheoryHodgeLaplacianbiomolecularflexibilityB-factorpredictionnaturalmodeanalysiscryo-EMdiscreteexteriorcalculusmoleculartopology
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

This paper aims to establish that a single mathematical object—the Laplace-de Rham operator on the volume enclosed by a molecular surface—can play the roles that currently require separate geometric, topological, and elastic-network models of biomolecules. Its headline quantitative claim is that the 0-form Hodge Laplacian spectrum, with a least-squares scale factor, predicts protein B-factors with an average Pearson correlation of 0.580 over 364 proteins, compared with 0.565 for the Gaussian network model. For cryo-EM density maps, the paper augments the 1-form Hodge Laplacian with a Helfrich curvature energy and reads the resulting eigenmodes as natural deformation modes, including motions beyond the small-deformation harmonic regime. A sympathetic reader would care because the framework operates directly on volumetric data and can in principle analyze flexibility and collective motion at atomic, residue, domain, and organelle scales with one continuum formulation.

What carries the argument

The load-bearing object is the discrete Hodge Laplacian $L_k = D_k^T S_{k+1} D_k + S_k D_{k-1} S_{k-1}^{-1} D_{k-1}^T S_k$, assembled by discrete exterior calculus on a tetrahedral mesh of the volume enclosed by the molecular surface. Here $D_k$ are signed incidence matrices satisfying $D_{k+1}D_k = 0$, and $S_k$ are diagonal Hodge star matrices converting primal to dual mesh elements; the generalized eigenvalue problem $L_k\omega_k = \lambda_k S_k\omega_k$ yields the spectral basis. The same construction gives the Helmholtz-Hodge decomposition of vector fields into orthogonal gradient, curl, and harmonic components. The 0-form spectrum drives the B-factor predictor through Eq. (11), and the 1-form spectrum with the added $G^T Q G$ Helfrich boundary term drives the Laplace-de Rham-Helfrich natural-mode operator of Eq. (16).

What would settle it

On the flexibility side, rerun the 364-protein benchmark with Eq. (11) using eigenmodes of the plain scalar Laplacian on the same tetrahedral mesh: if the average Pearson correlation remains near 0.580, the Hodge structure is not the source of the prediction. On the motion side, compare the lowest nontrivial modes of the Laplace-de Rham-Helfrich operator against an ensemble of experimentally observed conformations of one cryo-EM complex; if the predicted modes fail to span the observed displacement, the curvature-energy term is not capturing real collective motion.

Watch

Extended reading notes

Core claim

The paper's central claim is that the spectrum and eigenfields of the Hodge Laplacian on a biomolecular manifold contain the geometric and topological information of the molecule and, through a spectral inversion, its flexibility and natural modes. Topology is read from the zero-eigenvalue harmonic forms: the null space of the tangential 1-form operator counts tunnels and handles, the normal 1-form null space counts cavities, and the 0-form null space counts connected components. Geometry is read from the nonzero spectrum, organized into three independent groups—tangential gradient, normal gradient, and curl—that distinguish proteins by shape. The B-factor model sets the fluctuation at an atom to a times the sum over modes of $(1/\lambda_j)\,\omega_j(r)\,\omega_j(r')^T$ evaluated at that atom, which yields the reported 0.580 average Pearson correlation over the 364-protein benchmark. The natural-mode model replaces the 1-form Laplacian with $E_\mu = d_0\star_0^{-1} d_0^T \star_1 + \star_1^{-1} d_1^T \star_2 d_1 + G^T Q G$, where $Q$ is the quadratic form of the Helfrich curvature energy, and treats its eigenmodes as collective motions of X-ray structures and cryo-EM maps.

Load-bearing premise

The load-bearing premise is that an atom's thermal fluctuation is the inverse-spectrum sum of the Hodge Laplacian at that atom (Eq. 11), an identification assumed rather than derived; if the spectrum does not represent physical fluctuation, the reported accuracy does not validate the paradigm.

Editorial extensions

If this is right

  • The 0-form spectral model estimates flexibility from a continuum molecular surface without atomic coordinates, so B-factor-like fluctuations become computable for very large complexes and for cryo-EM maps at any resolution.
  • The same discretization returns topology (handles, cavities, connected components) and geometry (three independent spectral groups) from one pipeline, replacing separate persistent-homology and shape-analysis computations.
  • The Laplace-de Rham-Helfrich operator yields natural modes directly from cryo-EM density, with two tunable weights separating divergence energy, curl energy, and curvature energy, and it permits larger anharmonic deformations than harmonic normal mode analysis.
  • Because computational cost scales with mesh size rather than atom count, the approach remains feasible for problems with millions of atoms that are intractable for coordinate-based elastic models.

Reading between the lines

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

  • The authors do not test this, but if Eq. (11) is a general spectral-to-fluctuation correspondence, then any shape model with a Laplace operator—coarse density maps, reconstructed organelles, or surface-only representations—becomes a flexibility predictor without force-field parameters.
  • The harmonic eigenfields of the 1-form Laplacian carry not just the count of handles and cavities but also their spatial location and width; these fields could serve as differentiable topological features for machine learning, going beyond persistent-homology barcodes that discard geometric location.
  • The natural-mode model's physical content could be checked against experimentally observed conformational ensembles of the same complex; a mismatch between the predicted low-order modes and observed displacements would show how much of the predictive content comes from the Helfrich term rather than the Laplacian itself.
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 / 7 minor

Summary. The paper introduces de Rham-Hodge theory as a unified mathematical framework for analyzing biomolecular geometry, topology, flexibility, and natural modes. The authors construct tetrahedral-mesh domains from Gaussian-density level sets of proteins or cryo-EM maps, implement discrete exterior calculus (DEC) with tangential and normal boundary conditions, and compute spectra of Hodge Laplacians. For flexibility, Eq. (11) models the B-factor at an atom as a scale factor times the diagonal of the pseudoinverse of the 0-form Hodge Laplacian evaluated at that atom, and the method is tested on a 364-protein benchmark, reporting an average Pearson correlation of 0.580 versus 0.565 for GNM (Table 1). For natural modes, Eqs. (13)-(16) define a Laplace-de Rham-Helfrich operator that adds a Helfrich curvature energy boundary term to the 1-form Laplacian, and low-frequency eigenmodes are shown qualitatively for cryo-EM maps such as EMD 1258. The paper also illustrates Hodge decomposition of vector fields and topological analysis via harmonic forms.

Significance. If the proposed framework were fully validated, it would be a significant contribution: it offers a common mathematical language for geometry, topology, flexibility, and collective motion, and it can in principle operate directly on cryo-EM maps and at variable resolutions where coordinate-based elastic network models are impractical. The mathematical background on de Rham-Hodge theory and DEC is standard, and the exposition of the three independent spectra and their boundary-condition dualities is useful. The paper also provides a concrete algorithmic pathway—simplicial mesh generation, DEC matrices, and generalized eigenvalue problems—that is reproducible in principle. However, the central empirical claims are not yet established: the B-factor model is introduced as an assumption rather than derived, the reported accuracy gain over GNM is tiny and lacks statistical support, and the natural-mode results are only qualitative. The paper's value is therefore best assessed as a promising methodological proposal whose validation needs substantial additional work.

major comments (4)
  1. [§II.B, Eq. (11)] The B-factor model is load-bearing but is introduced with 'We assume' and no derivation. The sum over (1/lambda_j) omega_j(r) omega_j(r')^T is the diagonal of the pseudoinverse of the Hodge Laplacian, which is the equilibrium Green's function for a scalar field with Dirichlet energy. In GNM this quantity has a clear harmonic-spring Hamiltonian interpretation; here the field is a geometric density constructed from Gaussian kernels, not an atomic displacement field, so the identification of this Green's function with thermal atomic fluctuations (Debye-Waller factors) is not justified. Without a statistical-mechanics or elasticity derivation, or an independent dynamical test (e.g., comparison with molecular dynamics mean-square fluctuations), the reported correlation does not validate the de Rham-Hodge paradigm for flexibility. The authors should either supply such a derivation or explicitly reframe Eq. (11) as an empirical descriptor and provide the corresponding evidence.
  2. [Table 1 and §II.B (Flexibility analysis)] The hyperparameters c (level set), r (grid spacing), and d (cutoff radius) are chosen on the same 364-protein benchmark and the best result is reported as the headline accuracy. This is effectively model selection on the test set. Given that the improvement over GNM is only 0.580 versus 0.565 in average Pearson correlation, the claim of outperformance is not supported without cross-validation, a separate held-out test, or a statistical significance test over the 364 proteins. The paper should report the variability across proteins (e.g., standard deviation or distribution of per-protein correlations) and compare against GNM under identical evaluation protocols.
  3. [§II.B, Eqs. (13)-(16)] The Laplace-de Rham-Helfrich operator is asserted rather than derived. Equation (14) defines Q = ∂²V/∂X² but does not specify the function space or the discretization of this second variation, and Eq. (15) introduces X = Gω without defining the matrix G or explaining how the surface displacement is represented as a 1-form. Consequently, the claim that low-frequency eigenmodes of Eq. (16) are physical natural modes, and that the first three are translations, is not backed by a mathematical derivation or by quantitative comparison to known motions (e.g., overlap with ANM modes or domain motions). The authors should at least specify the discrete construction of G and Q, and validate the modes against a ground-truth deformation or an established elastic model.
  4. [§II.B (Natural mode analysis), Fig. 9] The natural-mode results are purely qualitative: the paper shows eigenmode visualizations for EMD 1258 but provides no quantitative measure of whether these modes correspond to functionally relevant or experimentally observed motions. The statement that 'the first three eigenmodes are associated with 3D translational motions' is also not demonstrated in the text or figures. A quantitative validation—for example, modal overlap with a known conformational change or with ANM-computed modes—is needed to support the central claim that the operator predicts natural modes.
minor comments (7)
  1. [§II.B, Eq. (11)] The notation omega_j^k(r)(omega_j^k(r'))^T is confusing for 0-forms, where the quantity is a scalar and the transpose is superfluous; for 1-forms and 2-forms the intended product (outer product or pointwise inner product) should be stated explicitly.
  2. [§II.B (Flexibility analysis)] The text says 'The cutoff radius is set to 7 Å' but later states the best parameters are c = 0.4, r = 1.6 Å, d = 4.0 Å, and Table 1 reports results at cutoff radius 4.0 Å. This inconsistency should be resolved.
  3. [§II.B (Natural mode analysis)] The sentence 'In contrast, the Laplace-de Rham-Helfrich operator does preserve these properties' appears to contradict the immediately preceding claim that the original operator admits an orthogonal Hodge decomposition while the new operator does not; likely 'does not preserve' is intended.
  4. [General] There are several typographical issues, including 'eigenfucntions' (Section II.A), 'PPC' for 'PCC' (Fig. 7 caption), and 'cutoff radius' throughout; these should be corrected.
  5. [§II.A (Reduction and analysis)] Several textual references such as 'as shown in 3 b' and '3 e left' are incomplete or missing the figure label; the cross-references to Figure 3 panels should be made consistent.
  6. [Table 1] The GNM baseline value 0.565 is taken from Ref. [35], but the implementation details (e.g., spring constant, cutoff) used for GNM are not stated here; for a fair comparison these should be specified or the baseline rerun with the same protein set and protocol.
  7. [Conclusion and Methods] The paper does not mention availability of code or data. Given that the method involves many implementation choices (mesh generation, DEC assembly, boundary-condition handling), a reproducibility statement or release of the code would materially strengthen the contribution.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the flexibility and natural-mode models are explicit assumptions evaluated against external data, and the parameter grid search is an overfitting concern, not a circular one.

full rationale

The paper's claimed derivations do not reduce to their own inputs by construction. Equation (11) is introduced explicitly as an assumption ('We assume that the de Rham-Hodge B factor at the ith atom estimated by the Lk is given by...'), not as a first-principles derivation, and the experimental PDB B-factors enter only as external targets; the fitted scale a is a single per-protein factor that does not affect Pearson correlation, so the reported 0.580 versus 0.565 comparison is not forced by the fit. The parameters c, r, and d are selected on the same 364-protein benchmark by grid search (Table 1, Fig. 7g), which is a model-selection and generalization concern, not a circular reduction of the prediction to its inputs. The natural-mode model in Eqs. (13)-(16) is also an explicit ansatz: a Helfrich curvature energy is added to the 1-form Laplace-de Rham operator to define a new operator E_mu, and the displayed modes are eigenvectors of that defined operator; there is no experimental mode set being reproduced through fitted parameters such that the output would equal the input by construction. The topological statements about null-space dimensions rely on standard de Rham-Hodge mathematics rather than on the paper's own fitted values. The cited prior work [35, 42, 34] supplies the FRI surface construction, the benchmark set, and a GNM baseline, but these are inputs and comparisons, not load-bearing conclusions whose content is identical to the target claims; no self-citation is used to exclude alternatives or to import a uniqueness theorem. Accordingly, any doubts about the physical justification of Eq. (11) or about in-sample parameter tuning should be treated as correctness and validation risks, not as circularity.

Assumptions & free parameters 9 free parameters · 6 assumptions · 0 invented entities

The framework imports classical de Rham-Hodge theory and discrete exterior calculus as standard mathematics, plus two application-specific assumptions: the molecular level set is a valid manifold and the inverse-Hodge-Laplacian diagonal equals flexibility. The latter is explicitly assumed in Eq. (11). The natural mode operator adds Helfrich curvature energy through Eqs. (13)-(16) without derivation. No new physical entities, particles, forces, or conserved quantities are introduced.

free parameters (9)
  • B-factor scale a = not reported (least-squares fit in Eq. 11)
    Global proportionality constant between inverse Laplacian diagonal and experimental B-factors; does not affect Pearson correlation but is fitted to data.
  • Level set parameter c = 0.4 (best on 364-protein benchmark)
    Controls the isosurface of the Gaussian density (Eq. 10) and hence the molecular domain; chosen by maximizing average Pearson correlation in Table 1 and Fig. 7g.
  • Grid spacing r = 1.6 Å (best)
    Controls tetrahedral mesh density; selected on the same benchmark; finer meshes improve PCC but cost more.
  • B-factor cutoff radius d = 4.0 Å (best)
    Linear-regression radius around each C-alpha atom in Eq. (11); selected on the benchmark; the paper reports lesser influence at 5 Å.
  • Gaussian kernel width eta = twice the atomic van der Waals radius
    Scale in density estimator Eq. (5); inherited from FRI, chosen by hand.
  • Gaussian exponent kappa = 2 (Gaussian kernel)
    Kernel in Eq. (8); chosen, not fitted.
  • Laplace-de Rham-Helfrich weight lambda = unspecified (chosen with mu > lambda > 1)
    Weights divergence energy in Eq. (17); no values or sensitivity analysis given.
  • Bending rigidity mu = unspecified
    Scales Helfrich curvature energy Eq. (13); no value given in paper.
  • Spontaneous curvature H0 = unspecified
    Appears in Eq. (13); never defined or quantified for the molecules.
assumptions (6)
  • standard math Hodge decomposition and de Rham theorem hold on compact molecular manifolds with boundary under the chosen boundary conditions.
    Used throughout Section II.A to split fields and identify harmonic forms with topology. These are classical results.
  • domain assumption The level set of the Gaussian density (Eq. 10) is a smooth compact manifold sufficiently representing the macromolecular surface.
    All spectral and topological analysis is computed on this constructed domain; the level set value changes the shape and the spectra.
  • domain assumption The diagonal discrete Hodge star on a tetrahedral mesh gives a faithful approximation of the continuous L2 inner product and spectra.
    Section III.B uses diagonal volume ratios for S_k; the paper notes higher-order Hodge stars exist but treats the diagonal version as sufficient, without convergence analysis for the biomolecular shapes.
  • ad hoc to paper B-factors are proportional to the diagonal of the pseudoinverse of the 0-form Hodge Laplacian (Eq. 11).
    Stated as 'We assume' in Section II.B; no derivation from statistical mechanics or elasticity is provided.
  • ad hoc to paper Helfrich curvature energy on the molecular boundary is the relevant potential for internal natural modes and couples through G^T Q G (Eqs. 13-16).
    The Laplace-de Rham-Helfrich operator is assembled by adding this curvature quadratic form to the 1-form Laplacian; no derivation from molecular force fields is given.
  • ad hoc to paper The low-frequency eigenmodes of the Laplace-de Rham-Helfrich operator represent physical natural modes, with the first three modes being translations.
    Section II.B interprets the modes in Fig. 9 as natural modes without comparison to NMA/ANM or experimental displacement data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The de Rham-Hodge analysis and modeling of biomolecules." pith.science (2026). https://pith.science/paper/G46CA2WR

@misc{pith2026190800572,
  author       = {Pith},
  title        = {Pith review of: The de Rham-Hodge analysis and modeling of biomolecules},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/G46CA2WR}},
  note         = {Machine review of arXiv:1908.00572}
}
read the original abstract

Recent years have witnessed a trend that advanced mathematical tools, such as algebraic topology, differential geometry, graph theory, and partial differential equations, have been developed for describing biological macromolecules. These tools have considerably strengthened our ability to understand the molecular mechanism of macromolecular function, dynamics and transport from their structures. However, currently, there is no unified mathematical theory to analyze, describe and characterize biological macromolecular geometry, topology, flexibility and natural mode at a variety of scales. We introduce the de Rham-Hodge theory, a landmark of 20th Century's mathematics, as a unified paradigm for analyzing biological macromolecular geometry and algebraic topology, for predicting macromolecular flexibility, and for modeling macromolecular natural modes at a variety of scales. In this paradigm, macromolecular geometric characteristic and topological invariants are revealed by de Rham-Hodge spectral analysis. By using the Helmholtz-Hodge decomposition, every macromolecular vector field is split into orthogonal divergence-free, curl-free, and harmonic components with a distinct physical interpretation. We organize the eigenvalues and eigenvectors of the 0-form Laplace-de Rham operator into one of the most accurate protein B-factor predictors. By combining the 1-form Laplace-de Rham operator and the Helfrich-type curvature energy, we predict the natural modes of both X-ray protein structures and cryo-EM maps. We construct accurate and efficient three-dimensional discrete exterior calculus algorithms for the aforementioned modeling and analysis of biological macromolecules. Using extensive experiments, we validate that the proposed paradigm is one of the most versatile and powerful tools for biological macromolecular studies.

Figures

Figures reproduced from arXiv: 1908.00572 by the authors.

Figure 1
Figure 1. Illustration of tangential spectra of a cryo-EM map EMD 7972 Topologically, EMD 7972 has 6 handles and 2 cavities. The left column is the original shape and its anatomy showing the topological complexity. On the right-hand side of the parenthesis, the first row shows tangential harmonic eigen fields, the second row shows tangential gradient eigen fields, and the third row shows tangential curl eigen fields. 6 [PITH… view at source ↗
Figure 2
Figure 2. Illustration of the normal spectra of protein and DNA complex 6D6V Topolog￾ically, the crystal structure of 6D6V has 1 handle. The left column shows the secondary structure and the solvent excluded surface (SES). On the right-hand side, the first two rows show normal gradient eigen fields, and the last two rows show normal curl eigen fields. 9 [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Illustration of Hodge Laplacian spectra This figure shows the properties of 3 spectral groups, namely, tangential gradient eigen fields (T), normal gradient eigen fields (N), and curl eigen fields (C), for EMD 8962. a shows the original input surface and 3 distinct spectral groups. b shows the cross section of a typical tangential gradient eigen field and the distribution of eigenvalues for group T. c shows the cros… view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Illustration of topological analysis. a Eigen fields by null space of tangential Laplace￾de Rham operators correspond to handles. b Eigen fields by null space of normal Laplace-de Rham operators correspond to cavities. Eigenvalue No. Eigenvalue 0 20 40 60 80 100 0 0.05…
Figure 5
Figure 5. Figure 5: Illustration of geometric analysis The geometry of different molecules (PDB IDs: 2Z5H (a), 6HU5 (b), and 5HY9 (c)) can be captured by three groups of different Hodge Laplacian spectra with clear separations shown in d. Note that the color of the line plot corresponds t…
Figure 6
Figure 6. Figure 6: Biological flow decomposition Illustration of a synthetic vector field in EMD 1590 that is decomposed into several mutually orthogonal components based on different boundary conditions. Geometric analysis and topological analysis based on the de Rham-Hodge theory can b…
Figure 7
Figure 7. Figure 7: Illustration of the procedure for flexibility analysis. We use protein 3VZ9 as an example to demonstrate our procedure from a to f. a shows the input protein crystal structure. b shows that only C-alpha atoms (yellow spheres) are considered in this case. In fact, our m…
Figure 8
Figure 8. Figure 8: Illustration of B-factor prediction. We use proteins 3F2Z, 3VZ9 and 2COV as examples to show our predictions compared with the experiments. The blue lines are the ground truth from experiments. The orange lines are predictions from with our method. Natural mode analysi…
Figure 9
Figure 9. Figure 9: Natural modes of EMD 1258. The 0-th, 4-th, 8-th and 12-th natural modes are shown. strain the elastic motion of biological macromolecules. There are many options, such as Willmore energy, which minimize the difference between two principle curvatures. Additionally, Hel…
Figure 10
Figure 10. Figure 10: Illustration of the primal and dual elements of the tetrahedral mesh. All the red vertices are mesh primal vertices. All the indigo vertices are dual vertices at circumcenter of each tet. All the gray edges are primal edges. All the pink edges are dual edges connectin…
Figure 11
Figure 11. Figure 11: Illustration of orientation. Pre-assigned orientation is colored in red. Induced orien￾tation by ∂ is colored in green. The vertices are assumed to have a positive pre-assigned orientation. Therefore, the induced orientation from edge orientation is +1 at the head and…
Figure 12
Figure 12. Figure 12: Illustration of cohomology. This figure illustrates the relation by exterior derivative and Hodge star operators. The assembly of Laplacian operator Lk is just starting from primal k-forms, multiplying matrices along the circular direction. L0 = D0 T S1 D0 L1 = D1 T S…
Figure 13
Figure 13. Figure 13: Illustration of Laplacian assembly. Laplacian matrix is assembled by matrix multi￾plication with pre-assembled discrete exterior derivative and discrete Hodge star matrices. conforming with the induced orientation. One can easily observe that the discrete exterior der…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 44 canonical work pages

  1. [1]

    D. N. Arnold, R. S. Falk, and R. Winther. Finite element exterior calculus, homological tech- niques, and applications. Acta numerica, 15:1–155, 2006

    work page 2006

  2. [2]

    A. R. Atilgan, S. Durell, R. L. Jernigan, M. Demirel, O. Keskin, and I. Bahar. Anisotropy of fluctuation dynamics of proteins with an elastic network model. Biophysical journal, 80(1):505– 515, 2001

    work page 2001

  3. [3]

    A. R. Atilgan, S. R. Durrell, R. L. Jernigan, M. C. Demirel, O. Keskin, and I. Bahar. Anisotropy of fluctuation dynamics of proteins with an elastic network model. Biophys. J. , 80:505 – 515, 2001

    work page 2001

  4. [4]

    Bahar, A

    I. Bahar, A. R. Atilgan, and B. Erman. Direct evaluation of thermal fluctuations in proteins using a single-parameter harmonic potential. Folding and Design, 2:173 – 181, 1997

    work page 1997

  5. [5]

    P. W. Bates, G. W. Wei, and S. Zhao. Minimal molecular surfaces and their applications. Journal of Computational Chemistry , 29(3):380–91, 2008

    work page 2008

  6. [6]

    Bhatia, G

    H. Bhatia, G. Norgard, V. Pascucci, and P.-T. Bremer. The helmholtz-hodge decompositiona survey. IEEE Transactions on visualization and computer graphics , 19(8):1386–1404, 2013

    work page 2013

  7. [7]

    J. F. Blinn. A generalization of algebraic surface drawing. ACM Trans. Graph. , 1:235–256, 1982

    work page 1982

  8. [8]

    Bossavit

    A. Bossavit. Whitney forms: A class of finite elements for three-dimensional computations in electromagnetism. IEE Proceedings A (Physical Science, Measurement and Instrumentation, Management and Education, Reviews) , 135(8):493–500, 1988

    work page 1988

Show all 48 references
  1. [9]

    B. R. Brooks, R. E. Bruccoleri, B. D. Olafson, D. States, S. Swaminathan, and M. Karplus. Charmm: A program for macromolecular energy, minimization, and dynamics calculations. J. Comput. Chem., 4:187–217, 1983

    1983

  2. [10]

    Z. X. Cang and G. W. Wei. TopologyNet: Topology based deep convolutional and multi- task neural networks for biomolecular property predictions. PLOS Computational Biology , 13(7):e1005690, https://doi.org/10.1371/journal.pcbi.1005690, 2017

    doi

  3. [11]

    Z. X. Cang and G. W. Wei. Integration of element specific persistent homology and machine learning for protein-ligand binding affinity prediction . International Journal for Numerical Methods in Biomedical Engineering, 34(2):DOI: 10.1002/cnm.2914, 2018

    doi

  4. [12]

    Carlsson, A

    G. Carlsson, A. Zomorodian, A. Collins, and L. J. Guibas. Persistence barcodes for shapes. International Journal of Shape Modeling , 11(2):149–187, 2005

    2005

  5. [13]

    M. Chen, B. Tu, and B. Lu. Triangulated manifold meshing method preserving molecular surface topology. J. Mole. Graph. Model. , 38:411–418, 2012. 25

    2012

  6. [14]

    Cheng and X

    H.-L. Cheng and X. Shi. Quality mesh generation for molecular skin surfaces using restricted union of balls. Computational Geometry, 42(3):196–206, 2009

    2009

  7. [15]

    R. B. Corey and L. Pauling. Molecular models of amino acids, peptides, and proteins. Review of Scientific Instruments , 24(8):621–627, 1953

    1953

  8. [16]

    J. G. De La Torre and V. A. Bloomfield. Hydrodynamic properties of macromolecular com- plexes. i. translation. Biopolymers: Original Research on Biomolecules , 16(8):1747–1763, 1977

    1977

  9. [17]

    Demlow and A

    A. Demlow and A. N. Hirani. A posteriori error estimates for finite element exterior calculus: the de rham complex. Found. Comput. Math., 14(6):1337–1371, 2014

    2014

  10. [18]

    Desbrun, A

    M. Desbrun, A. N. Hirani, M. Leok, and J. E. Marsden. Discrete exterior calculus. arXiv preprint math/0508341, 2005

    2005 arXiv

  11. [19]

    T. K. Dey, F. Fan, and Y. Wang. An efficient computation of handle and tunnel loops via reeb graphs. ACM Trans. Graph., 32(4):32, 2013

    2013

  12. [20]

    Q. Du, C. Liu, and X. Wang. A phase field approach in the numerical study of the elastic bending energy for vesicle membranes. Journal of Computational Physics , 198(2):450–468, 2004

    2004

  13. [21]

    B. S. Duncan and A. D. Olson. Shape analysis of molecular surfaces. Biopolymers, 33 2:231–8, 1993

    1993

  14. [22]

    Edelsbrunner and J

    H. Edelsbrunner and J. Harer. Computational topology: An introduction. American Mathemat- ical Soc., 2010

    2010

  15. [23]

    X. Feng, K. Xia, Y. Tong, and G.-W. Wei. Geometric modeling of subcellular structures, organelles and large multiprotein complexes. International Journal for Numerical Methods in Biomedical Engineering, 28:1198–1223, 2012

    2012

  16. [24]

    N. Go, T. Noguti, and T. Nishikawa. Dynamics of a small globular protein in terms of low- frequency vibrational modes. Proc. Natl. Acad. Sci., 80:3696 – 3700, 1983

    1983

  17. [25]

    Haslam, T

    D. Haslam, T. Zeng, R. Li, and J. He. Exploratory studies detecting secondary structures in medium resolution 3d cryo-em images using deep convolutional neural networks. In Proceedings of the 2018 ACM International Conference on Bioinformatics, Computational Biology, and Healt...

    2018

  18. [26]

    D. R. Hekstra, K. I. White, M. A. Socolich, R. W. Henning, V. ˇSrajer, and R. Ranganathan. Electric-field-stimulated protein mechanics. Nature, 540(7633):400, 2016

    2016

  19. [27]

    Helfrich

    W. Helfrich. Elastic properties of lipid bilayers: Theory and possible experiments. Zeitschrift f¨ ur Naturforschung Teil C, 28:693 – 703, 1973

    1973

  20. [28]

    A. N. Hirani. Discrete exterior calculus. PhD thesis, California Institute of Technology, 2003

    2003

  21. [29]

    Lee and F

    B. Lee and F. M. Richards. The interpretation of protein structures: estimation of static accessibility. Journal of molecular biology , 55(3):379–IN4, 1971

    1971

  22. [30]

    Levitt, C

    M. Levitt, C. Sander, and P. S. Stern. Protein normal-mode dynamics: Trypsin inhibitor, crambin, ribonuclease and lysozyme. J. Mol. Biol. , 181(3):423 – 447, 1985. 26

    1985

  23. [31]

    J. P. Ma. Usefulness and limitations of normal mode analysis in modeling dynamics of biomolec- ular complexes. Structure, 13:373 – 180, 2005

    2005

  24. [32]

    D. Ming, Y. Kong, M. A. Lambert, Z. Huang, and J. Ma. How to describe protein motion without amino acid sequence and atomic coordinates. Proceedings of the National Academy of Sciences, 99(13):8620–8625, 2002

    2002

  25. [33]

    Natarajan, P

    V. Natarajan, P. Koehl, Y. Wang, and B. Hamann. Visual analysis of biomolecular surfaces. In L. Linsen, H. Hagen, and B. Hamann, editors, Mathematical Methods for Visualization in Medicine and Life Science , pages 237–256. Springer Verlag, 2008

    2008

  26. [34]

    D. D. Nguyen, K. L. Xia, and G. W. Wei. Generalized flexibility-rigidity index. Journal of Chemical Physics, 144:234106, 2016

    2016

  27. [35]

    Opron, K

    K. Opron, K. L. Xia, and G. W. Wei. Fast and anisotropic flexibility-rigidity index for protein flexibility and fluctuation analysis. Journal of Chemical Physics , 140:234105, 2014

    2014

  28. [36]

    F. M. Richards. Areas, volumes, packing, and protein structure. Annual review of biophysics and bioengineering, 6(1):151–176, 1977

    1977

  29. [37]

    F. Tama, W. Wriggers, and C. L. Brooks III. Exploring global distortions of biological macro- molecules and assemblies from low-resolution structural information and elastic network theory. Journal of molecular biology , 321(2):297–305, 2002

    2002

  30. [38]

    Tasumi, H

    M. Tasumi, H. Takenchi, S. Ataka, A. M. Dwidedi, and S. Krimm. Normal vibrations of proteins: Glucagon. Biopolymers, 21:711 – 714, 1982

    1982

  31. [39]

    Xia and G.-W

    K. Xia and G.-W. Wei. A review of geometric, topological and graph theory apparatuses for the modeling and analysis of biomolecular data. arXiv preprint arXiv:1612.01735 , 2016

    2016 arXiv

  32. [40]

    K. L. Xia, X. Feng, Y. Y. Tong, and G. W. Wei. Multiscale geometric modeling of macro- molecules i: Cartesian representation. Journal of Computational Physics , 275:912–936, 2014

    2014

  33. [41]

    K. L. Xia, X. Feng, Y. Y. Tong, and G. W. Wei. Persistent homology for the quantitative prediction of fullerene stability. Journal of Computational Chemistry , 36:408–422, 2015

    2015

  34. [42]

    K. L. Xia, K. Opron, and G. W. Wei. Multiscale multiphysics and multidomain models — Flexibility and rigidity. Journal of Chemical Physics , 139:194109, 2013

    2013

  35. [43]

    K. L. Xia and G. W. Wei. Persistent homology analysis of protein structure, flexibility and folding. International Journal for Numerical Methods in Biomedical Engineering , 30:814–844, 2014

    2014

  36. [44]

    Y. Yao, J. Sun, X. Huang, G. R. Bowman, G. Singh, M. Lesnick, L. J. Guibas, V. S. Pande, and G. Carlsson. Topological methods for exploring low-density states in biomolecular folding pathways. The Journal of chemical physics , 130(14):04B614, 2009

    2009

  37. [45]

    Z. Y. Yu, M. Holst, Y. Cheng, and J. A. McCammon. Feature-preserving adaptive mesh generation for molecular shape modeling and simulation. Journal of Molecular Graphics and Modeling, 26:1370–1380, 2008

    2008

  38. [46]

    R. Zhao, Z. Cang, Y. Tong, and G.-W. Wei. Protein pocket detection via convex hull surface evolution and associated reeb graph. Bioinformatics, 34 17:i830–i837, 2018. 27

    2018

  39. [47]

    R. Zhao, Z. Cang, Y. Tong, and G.-W. Wei. Protein pocket detection via convex hull surface evolution and associated Reeb graph. Bioinformatics, 34 (17):i830–i837, 2018

    2018

  40. [48]

    Zheng, S

    Q. Zheng, S. Yang, and G.-W. Wei. Biomolecular surface construction by pde transform. International journal for numerical methods in biomedical engineering , 28(3):291–316, 2012. 28

    2012

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.