Lipidation-induced bacterial cell membrane translocation of star-peptides

Amal Jayawardena; Andrew Hung; Elnaz Hajizadeh; Greg Qiao

REVIEW 4 major objections 6 minor 8 references

Lipidation-induced bacterial cell membrane translocation of star-peptides

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Lipidation with moderate-length fatty acids stabilizes the helical arms of SNAPP and lowers the free-energy barrier for crossing a bacterial membrane, with C12 causing the strongest bilayer disruption and C18 back-folding to weaken it.

desk verdict Plausible qualitative story about lipidated SNAPPs, but the PMF barrier numbers are internally inconsistent and lack error bars, so the central quantitative claim shouldn't be taken at face value. read the letter →

arxiv 2505.06447 v1 pith:25G4IQ54 submitted 2025-05-09 physics.bio-ph cond-mat.soft

Amal Jayawardena , Andrew Hung , Greg Qiao , Elnaz Hajizadeh This is my paper · ORCID

classification physics.bio-phcond-mat.soft

keywords antimicrobialpeptidesSNAPPlipidationmoleculardynamicsmembranetranslocationpotentialofmeanforcebilayerdisruptionantibioticresistance

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 uses atomistic molecular dynamics to ask whether attaching fatty acids to the arms of SNAPP—an eight-arm star-shaped antimicrobial peptide polymer—helps or hinders its attack on bacterial membranes. It finds that C6 and C12 lipid chains stabilize the arms' α-helical structure, let them insert deeper into the hydrophobic core, and lower the free-energy barrier for crossing a model POPE/POPG bilayer, with C12 producing the largest bilayer thinning and deformation. The C18 chain, by contrast, makes the arms fold back toward the polymer core, and the membrane is barely disturbed. These results indicate that lipid chain length is a tunable design knob for next-generation antibiotics aimed at multidrug-resistant bacteria.

What carries the argument

The central object is SNAPP (Structurally Nanoengineered Antimicrobial Peptide Polymer), an eight-arm star polymer whose arms are alternating lysine/valine peptides (KKVKKVKKVKKVKKV), with fatty acids covalently attached to the terminal valine of each arm. The argument runs on two computed quantities: per-arm pulling-force profiles from steered molecular dynamics, and potential-of-mean-force (PMF) profiles from umbrella sampling reconstructed with WHAM, both along a reaction coordinate through a 4:1 POPE/POPG bilayer. The PMF barrier height is what carries the translocation claim, while α-helical stability is quantified by ellipticity at 222 nm.

What would settle it

Synthesize the C6-, C12-, and C18-lipidated SNAPPs and measure membrane permeabilization or minimum inhibitory concentrations: the mechanism predicts C12-SNAPP should be the strongest membrane disrupter and C18-SNAPP the weakest, and that C6/C12 arms should cross the bilayer with lower free-energy barriers than unmodified arms; observing the opposite ranking would refute it.

Watch

Extended reading notes

Core claim

The central claim is a length-dependent lipidation effect: conjugating hexanoic acid (C6) or lauric acid (C12) to the N-terminus of each SNAPP arm preserves and enhances the α-helical structure and hydrophobic insertion, whereas stearic acid (C18) drives the arms to back-fold into a compact blob and weak membrane engagement. In umbrella-sampling PMFs of single arms, the energy barrier for crossing the bilayer drops from 409.9 kJ/mol for the unmodified arm to 350.9 kJ/mol for C6 and 362.9 kJ/mol for C12, while C18 remains near the unmodified value; pulling-force peaks follow the same trend. Bilayer thickness and upper-leaflet deformation maps rank the disruption as C12 > C6 > C18, with C18 causing negligible thinning. The authors conclude that moderate lipidation is a design principle for membrane-active antimicrobial star polymers and that excessive hydrophobicity is counterproductive.

Load-bearing premise

The PMF and pulling-force numbers come from a single SNAPP arm pulled through the bilayer, not from the intact eight-arm star, so the quantitative conclusions assume that the core and arm-arm cooperation do not change the translocation barrier.

Editorial extensions

If this is right

If the single-arm barrier reflects the full star, C6 and C12 lipidation should make SNAPP arms cross bacterial membranes with less energy cost than unmodified arms.
C12-SNAPP is predicted to be the most bilayer-disruptive of the three lipidated variants, making it the lead candidate for further antimicrobial testing.
C18 lipidation is predicted to be counterproductive: long chains collapse the arms and weaken membrane deformation, so longer lipids are not better.
Chain length, not just hydrophobicity, becomes a tunable parameter for optimizing star-peptide antibiotics, with an effective window around C6–C12.

Reading between the lines

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

Editorial inference: the PMF calculation omits the eight-arm core and arm-arm cooperativity; the natural next test is the same umbrella-sampling calculation on the intact star, which could change barrier heights even if it preserves the ordering.
A direct experimental check would be to synthesize C6-, C12-, and C18-lipidated SNAPPs and measure MICs or membrane leakage: the simulations predict C12-SNAPP should be the most potent and C18-SNAPP the least.
The C18 back-folding suggests a general design caution: any highly hydrophobic anchor that self-associates may reduce multivalent membrane engagement, so amphiphilic balance matters more than raw hydrophobicity.

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.

Desk Editor's Note

read the letter

Quick take: this is a reasonable computational study that extends the authors' earlier SNAPP simulations to lipidated variants, and the qualitative observations about C18 back-folding are the most interesting part. But the central quantitative claim—that lipidation lowers the translocation free-energy barrier—rests on PMF numbers that are internally inconsistent and have no error bars. I wouldn't trust the barrier ordering as stated, though the paper is worth refereeing.

What's new: applying C6, C12, C18 lipidation to eight-arm SNAPP models in atomistic MD. The C18 back-folding behavior and the contact/bilayer-thickness analyses are concrete and plausible. The methods are standard CHARMM36 with reasonable protocols, and the authors cite the relevant lipidation literature. The qualitative trend (C12 most disruptive, C18 least) is consistent with their snapshots and contact counts.

Where I'd push back: the PMF section. The alt-SNAPP barrier is quoted as 381 kJ/mol at point 'c', but the text says the rise from -28.9 to 381 is 409.9 kJ/mol, and the conclusion reports 409.9 as the barrier. That is an arithmetic inconsistency that matters because the comparisons use these numbers. Second, there are no replicate trajectories, no error bars, and 200 ns per window along a single collective variable. The difference between alt (381) and C12 (362.9) is 18 kJ/mol, which is likely within sampling noise for this setup. Third, the PMFs are computed for single arms, not intact eight-arm stars; the authors acknowledge this in Section 2.3 but the abstract does not. Fourth, the abstract says lipidation reduces the barrier as a general statement, but C18 does not reduce it; the authors later acknowledge this, so the abstract overgeneralizes. Figure 10 also lacks numeric axes, which prevents independent verification.

None of this kills the qualitative membrane-disruption story. But the quantitative PMF claims should be re-analyzed with error estimation, and the numbers corrected and reported consistently.

Who it's for: people working on antimicrobial peptide design and MD simulations of peptide-membrane interactions. It deserves peer review, but I'd expect significant revision. I'd bring it to a reading group only if someone is specifically interested in SNAPPs or lipidation strategies.

Referee Report

4 major / 6 minor

Summary. Jayawardena et al. present all-atom molecular dynamics simulations of three lipidated variants of the antimicrobial star polymer SNAPP (C6, C12, C18) in water and in water/TFE for secondary-structure analysis, next to a POPE/POPG bilayer for mechanism-of-action studies, and umbrella-sampling PMFs for a single SNAPP arm translocating through the same bilayer. They report that lipidation increases arm alpha-helicity, that C12 produces the largest bilayer thinning and deformation, that C18 arms back-fold and minimally deform the membrane, and that the PMF barrier for translocation is lower for C6 and C12 arms than for the unlipidated alt-SNAPP arm, while C18 is comparable. The authors conclude that tuning lipid chain length can optimize SNAPP-based antimicrobials.

Significance. The manuscript uses standard, well-documented methods (CHARMM36, CHARMM-GUI membrane builder, GROMACS, WHAM) and provides a systematic comparison of four lipidation states with 1 microsecond bilayer simulations and 50-window umbrella sampling. The qualitative observations — C6 and C12 tails insert into the hydrophobic core, C18 back-folds, and the bilayer deformation maps for the full stars — are plausible and visually supported. The paper does not, however, provide statistical uncertainties for the PMFs, and the reported barrier values are internally inconsistent; the quantitative conclusions are therefore not yet established.

major comments (4)

[Section 3.3.2 and Conclusion] The energy barrier for alt-SNAPP is internally inconsistent: the text says the peak at point 'c' is 381 kJ/mol, while also stating the rise from 'b' (-28.9 kJ/mol) to 'c' is 409.9 kJ/mol, and the conclusion reports the alt-SNAPP barrier as 409.9 kJ/mol. The C6 (350.9) and C12 (362.9) values are described as barriers from 'b' to their respective peaks. The definition of 'barrier' must be made consistent (peak relative to bulk water versus rise from the adsorbed minimum) and all variants compared on the same basis; as written, the headline comparison is not reproducible.
[Section 3.3.2, Figure 10] No error bars, replicate umbrella-sampling runs, or block/bootstrap estimates are provided for the PMFs. With a single run per variant and 200 ns of sampling per window along a single reaction coordinate, the reported 18-30 kJ/mol differences (alt vs C6, alt vs C12) cannot be distinguished from sampling noise. Because the central claim that lipidation lowers the translocation barrier rests on these differences, an uncertainty analysis is load-bearing and should be added.
[Section 2.3, Abstract, Conclusion] The PMF and pulling-force profiles are computed for single isolated SNAPP arms (Section 2.3), yet the abstract and conclusion attribute the barrier reduction to 'SNAPPs' and discuss translocation of the full star polymer. The central core and the multivalent cooperativity among the eight arms are not represented in the free-energy calculation. The claims should be explicitly restricted to single-arm behavior, or simulations of the intact eight-arm star under umbrella sampling should be provided.
[Abstract and Section 3.3.2] The abstract states that 'lipidation reduces the free energy barrier for translocation' as a general statement, but the C18-SNAPP arm has a barrier comparable to or higher than alt-SNAPP (the text says 'comparable to that of alt-SNAPP'). The conclusion acknowledges this; the abstract and general design-principle wording need to be qualified to C6 and C12 lipidation, with C18 treated as an exception.

minor comments (6)

[Section 1 (Introduction)] The text describes the C18 lipid as 'capric acid'; capric acid is C10, whereas the simulations use stearic acid (C18). Please correct this.
[Conclusion] The phrase 'ad 362.9 kJ/mol' should be 'and 362.9 kJ/mol'.
[Figures 4f, 5f, 6f, 7] The word 'unabected' should be 'unaffected' in the figure captions.
[Figure 10] The PMF plot is referenced with specific values in the text, but the axes are not visibly labeled with numerical values in the figure; please add quantitative axes and, if possible, error bands.
[Section 2.2] The sentence 'implemented within the GROMACS helix command' is unclear; please specify the exact GROMACS tool name and version, since the standard distribution does not use a command literally named 'helix'.
[Section 2.1] All arms are initialized with alpha-helical secondary structure; this modeling choice should be explicitly discussed as a limitation when comparing helical content among variants.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: lipidation effects are direct MD/WHAM simulation outputs; self-citation is non-load-bearing and the internal PMF inconsistencies are correctness issues, not circular reasoning.

full rationale

The paper's central claims—that C6/C12 lipidation lowers the PMF barrier and that C12-SNAPP causes the largest bilayer deformation—are direct outputs of atomistic MD simulations (GROMACS, CHARMM36) combined with umbrella sampling and WHAM. The PMF values are computed from simulated trajectories along the reaction coordinate, and the bilayer-thickness/deformation maps are obtained with VMD membrane analysis; none of these quantities is fitted to the claimed conclusion or defined in terms of it. The self-citation to ref. 13 is used to compare the unlipidated alt-SNAPP behavior with earlier simulations and to support model plausibility, but the lipidation-dependent results are generated de novo in the present simulations and do not reduce to that citation. The single-arm PMF caveat is an extrapolation/scope limitation, and the discrepancy between the quoted alt-SNAPP barrier (381 kJ/mol at point c) and the stated rise of 409.9 kJ/mol is an internal-consistency/correctness issue; neither is a case of a prediction being equivalent to its input by construction. No circular step is exhibited, so the appropriate score is 0.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on several modeling choices: the approximate core, the initial helical arms, the force field, and the single-arm PMF. None of these are fitted to experimental data for lipidated SNAPPs, but they are assumptions that could affect the conclusions.

free parameters (2)

Core sphere carbon count = 400
The PAMAM dendrimer core is approximated as a hydrophobic sphere of 400 carbon atoms (Section 2.1), chosen by hand to mimic a nonpolar surface. The real core is a PAMAM dendrimer with different size, charge, and flexibility.
Initial arm secondary structure = alpha-helical
Arms were initialized as alpha-helical (Section 2.1), which may bias the observed helicity and insertion behavior; the paper uses this to 'observe environment-dependent conformational changes,' but it is not an equilibrium starting point.

assumptions (4)

domain assumption CHARMM36 force field accurately models peptide-lipid interactions for these lipidated constructs
The paper relies on CHARMM36 (Section 2.1) without testing its accuracy for the unnatural lipidated peptide bonds used here.
domain assumption POPE/POPG 4:1 bilayer models the Gram-negative inner membrane
Section 2.1 uses this composition to replicate the inner membrane; real membranes have additional components and asymmetry.
domain assumption Single-arm PMF represents whole SNAPP translocation
Section 2.3 and 3.3.2 compute PMFs for isolated arms and use them to draw conclusions about SNAPP translocation.
ad hoc to paper Hydrophobic sphere core represents the PAMAM dendrimer core
Section 2.1: 'The core was approximated as a hydrophobic sphere, composed of 400 carbon atoms to mimic a nonpolar spherical surface.' This is an ad hoc simplification without validation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Lipidation-induced bacterial cell membrane translocation of star-peptides." pith.science (2026). https://pith.science/paper/25G4IQ54

@misc{pith2026250506447,
  author       = {Pith},
  title        = {Pith review of: Lipidation-induced bacterial cell membrane translocation of star-peptides},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/25G4IQ54}},
  note         = {Machine review of arXiv:2505.06447}
}

read the original abstract

The rapid emergence of multidrug-resistant (MDR) bacteria demands development of novel and effective antimicrobial agents. Structurally Nanoengineered Antimicrobial Peptide Polymers (SNAPPs), characterized by their unique star-shaped architecture and potent multivalent interactions, represent a promising solution. This study leverages molecular dynamics simulations to investigate the impact of lipidation on SNAPPs' structural stability, membrane interactions, and antibacterial efficacy. We show that lipidation with hexanoic acid (C6), lauric acid (C12), and stearic acid (C18) enhances the {\alpha}-helical stability of SNAPP arms, facilitating deeper insertion into the hydrophobic core of bacterial membranes. Among the variants, C12-SNAPP exhibits the most significant bilayer disruption, followed by C6-SNAPP, whereas the excessive hydrophobicity of C18-SNAPP leads to pronounced arm back-folding towards the core, reducing its effective interaction with the bilayer and limiting its bactericidal performance. Additionally, potential of mean force (PMF) analysis reveals that lipidation reduces the free energy barrier for translocation through the bilipid membrane compared to non-lipidated SNAPPs. These findings underscore the critical role of lipidation in optimizing SNAPPs for combating MDR pathogens. By fine-tuning lipid chain lengths, this study provides a framework for designing next-generation antimicrobial agents to address the global antibiotic resistance crisis, advancing modern therapeutic strategies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

8 extracted references · 6 canonical work pages

[1]

Introduction The rapid rise of multidrug-resistant (MDR) infections has become a global health crisis, posing a significant threat to public health and modern medicine. The World Health Organization (WHO) has declared antimicrobial resistance (AMR) as a critical priority, predicting that by 2050, AMR could lead to 10 million deaths annually if no effectiv...

work page
[2]

MD Simulations In this study, we used GROMACS 2021.v3 molecular dynamics simulation software package32 to perform the simulations detailed in Table A1

Methodology 2.1. MD Simulations In this study, we used GROMACS 2021.v3 molecular dynamics simulation software package32 to perform the simulations detailed in Table A1. The CHARMM36 force field was employed, as it has been widely validated in numerous studies for elucidating membrane properties and peptide-lipid interactions33. The modelled SNAPP chemical...

work page 2021
[17]

Structurally Nanoengineered Antimicrobial Peptide Polymers

DOI: 10.1186/1758-2946-4-17. (35) Martínez, L.; Andrade, R.; Birgin, E. G.; Martínez, J. M. PACKMOL: a package for building initial conﬁgurations for molecular dynamics simulations. J Comput Chem 2009, 30 (13), 2157-2164. DOI: 10.1002/jcc.21224 From NLM. (36) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual molecular dynamics. Journal of Molecular Graphi...

work page arXiv 2009
[2014]

(2) de Kraker, M. E. A.; Stewardson, A. J.; Harbarth, S. Will 10 Million People Die a Year due to Antimicrobial Resistance by 2050? PLOS Medicine 2016, 13 (11), e1002184. DOI: 10.1371/journal.pmed.1002184. (3) Dadgostar, P . Antimicrobial Resistance: Implications and Costs. Infect Drug Resist 2019, 12, 3903-3910. DOI: 10.2147/idr.S234610 From NLM. (4) O'N...

work page doi:10.1371/journal.pmed.1002184 2016
[2016]

P .; Wucher, B

(5) Smith, W. P .; Wucher, B. R.; Nadell, C. D.; Foster, K. R. Bacterial defences: mechanisms, evolution and antimicrobial resistance. Nature Reviews Microbiology 2023, 21 (8), 519-534. (6) Murray, C. J. L.; Ikuta, K. S.; Sharara, F .; Swetschinski, L.; Robles Aguilar, G.; Gray, A.; Han, C.; Bisignano, C.; Rao, P .; Wool, E.; et al. Global burden of bacte...

doi:10.1016/s0140-6736(21)02724-0 2023
[2018]

J.; O'Brien-Simpson, N

(11) Lam, S. J.; O'Brien-Simpson, N. M.; Pantarat, N.; Sulistio, A.; Wong, E. H. H.; Chen, Y.-Y .; Lenzo, J. C.; Holden, J. A.; Blencowe, A.; Reynolds, E. C.; et al. Combating multidrug-resistant Gram-negative bacteria with structurally nanoengineered antimicrobial peptide polymers. Nature Microbiology 2016, 1 (11), 16162. DOI: 10.1038/nmicrobiol.2016.162...

work page doi:10.1038/nmicrobiol.2016.162 2016
[3951]

Impact of lipidation site on the activity of α-helical antimicrobial peptides

(31) Makowska, M.; Wardowska, A.; Bauer, M.; Wyrzykowski, D.; Małuch, I.; Sikorska, E. Impact of lipidation site on the activity of α-helical antimicrobial peptides. Bioorganic Chemistry 2024, 153, 107821. DOI: https://doi.org/10.1016/j.bioorg.2024.107821. (32) Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. GROMAC...

arXiv 2024
[9692]

A.; HASEMAN, J

(53) LOCKWOOD, N. A.; HASEMAN, J. R.; TIRRELL, M. V .; MAYO, K. H. Acylation of SC4 dodecapeptide increases bactericidal potency against Gram-positive bacteria, including drug-resistant strains. Biochemical Journal 2004, 378 (1), 93-103. DOI: 10.1042/bj20031393 (accessed 1/20/2025). (54) Radzishevsky, I. S.; Kovachi, T.; Porat, Y .; Ziserman, L.; Zaknoon,...

work page arXiv 2022

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Introduction The rapid rise of multidrug-resistant (MDR) infections has become a global health crisis, posing a significant threat to public health and modern medicine. The World Health Organization (WHO) has declared antimicrobial resistance (AMR) as a critical priority, predicting that by 2050, AMR could lead to 10 million deaths annually if no effectiv...

work page

[2] [2]

MD Simulations In this study, we used GROMACS 2021.v3 molecular dynamics simulation software package32 to perform the simulations detailed in Table A1

Methodology 2.1. MD Simulations In this study, we used GROMACS 2021.v3 molecular dynamics simulation software package32 to perform the simulations detailed in Table A1. The CHARMM36 force field was employed, as it has been widely validated in numerous studies for elucidating membrane properties and peptide-lipid interactions33. The modelled SNAPP chemical...

work page 2021

[3] [17]

Structurally Nanoengineered Antimicrobial Peptide Polymers

DOI: 10.1186/1758-2946-4-17. (35) Martínez, L.; Andrade, R.; Birgin, E. G.; Martínez, J. M. PACKMOL: a package for building initial conﬁgurations for molecular dynamics simulations. J Comput Chem 2009, 30 (13), 2157-2164. DOI: 10.1002/jcc.21224 From NLM. (36) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual molecular dynamics. Journal of Molecular Graphi...

work page arXiv 2009

[4] [2014]

(2) de Kraker, M. E. A.; Stewardson, A. J.; Harbarth, S. Will 10 Million People Die a Year due to Antimicrobial Resistance by 2050? PLOS Medicine 2016, 13 (11), e1002184. DOI: 10.1371/journal.pmed.1002184. (3) Dadgostar, P . Antimicrobial Resistance: Implications and Costs. Infect Drug Resist 2019, 12, 3903-3910. DOI: 10.2147/idr.S234610 From NLM. (4) O'N...

work page doi:10.1371/journal.pmed.1002184 2016

[5] [2016]

P .; Wucher, B

(5) Smith, W. P .; Wucher, B. R.; Nadell, C. D.; Foster, K. R. Bacterial defences: mechanisms, evolution and antimicrobial resistance. Nature Reviews Microbiology 2023, 21 (8), 519-534. (6) Murray, C. J. L.; Ikuta, K. S.; Sharara, F .; Swetschinski, L.; Robles Aguilar, G.; Gray, A.; Han, C.; Bisignano, C.; Rao, P .; Wool, E.; et al. Global burden of bacte...

doi:10.1016/s0140-6736(21)02724-0 2023

[6] [2018]

J.; O'Brien-Simpson, N

(11) Lam, S. J.; O'Brien-Simpson, N. M.; Pantarat, N.; Sulistio, A.; Wong, E. H. H.; Chen, Y.-Y .; Lenzo, J. C.; Holden, J. A.; Blencowe, A.; Reynolds, E. C.; et al. Combating multidrug-resistant Gram-negative bacteria with structurally nanoengineered antimicrobial peptide polymers. Nature Microbiology 2016, 1 (11), 16162. DOI: 10.1038/nmicrobiol.2016.162...

work page doi:10.1038/nmicrobiol.2016.162 2016

[7] [3951]

Impact of lipidation site on the activity of α-helical antimicrobial peptides

(31) Makowska, M.; Wardowska, A.; Bauer, M.; Wyrzykowski, D.; Małuch, I.; Sikorska, E. Impact of lipidation site on the activity of α-helical antimicrobial peptides. Bioorganic Chemistry 2024, 153, 107821. DOI: https://doi.org/10.1016/j.bioorg.2024.107821. (32) Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. GROMAC...

arXiv 2024

[8] [9692]

A.; HASEMAN, J

(53) LOCKWOOD, N. A.; HASEMAN, J. R.; TIRRELL, M. V .; MAYO, K. H. Acylation of SC4 dodecapeptide increases bactericidal potency against Gram-positive bacteria, including drug-resistant strains. Biochemical Journal 2004, 378 (1), 93-103. DOI: 10.1042/bj20031393 (accessed 1/20/2025). (54) Radzishevsky, I. S.; Kovachi, T.; Porat, Y .; Ziserman, L.; Zaknoon,...

work page arXiv 2022