Article
Chemistry, Multidisciplinary
Chak Kui Wong, Chuyan Tang, John S. Schreck, Jonathan P. K. Doye
Summary: This study demonstrates the application of coarse-grained modeling combined with umbrella sampling to compute the free-energy landscapes associated with mechanical deformations of large DNA nanostructures. The approach is illustrated with examples of DNA nanotubes and DNA origami sheets. The ability to compute such landscapes is crucial for DNA mechanotechnology and understanding stress accumulation during the self-assembly process of origamis.
Article
Computer Science, Artificial Intelligence
Song Tang, Yan Zou, Zihao Song, Jianzhi Lyu, Lijuan Chen, Mao Ye, Shouming Zhong, Jianwei Zhang
Summary: This study proposes a new source data-free unsupervised domain adaptation method, which addresses the limitations of existing methods in defining geometric structures and depicting semantic relationships by conducting semantic consistency learning on a manifold.
Review
Chemistry, Multidisciplinary
Pablo G. Argudo, Juan J. Giner-Casares
Summary: Proteins and peptide fragments play a crucial role in self-assembly for nanostructures, with intrinsically disordered proteins and protein regions showing significant biological activity. Experimental techniques and computational modeling procedures are used for characterization, leading to a wide variety of nanostructures and promising performance in biotechnological applications. Exciting possibilities for IDPs and IDRs in nanotechnology with relevant biological applications are demonstrated.
NANOSCALE ADVANCES
(2021)
Article
Computer Science, Artificial Intelligence
Yuanhong Liu, Zebiao Hu, Yansheng Zhang
Summary: Locally linear embedding (LLE) is an effective tool for feature extraction from datasets. However, most existing algorithms assume that the dataset is in a Euclidean space, which is not the case for most of the original data space. This study addresses the issues in conventional LLE by constructing a symmetric positive definite manifold and estimating its tangent space, integrating local and global discriminant information into LLE, and performing feature extraction in the tangent space.
Article
Engineering, Multidisciplinary
John R. Jungck, Stephen Brittain, Donald Plante, James Flynn
Summary: The researchers have discovered three new hybrid classes based on their experience with 4D printing, indicating that self-assembly can involve self-folding, environment-dependent folding, and chaotic alternating processes. By using geometric and topological perspectives, they have developed design principles and tested them through experiments and models to demonstrate five different processes.
Article
Physics, Fluids & Plasmas
Fatih Yasar, Alan J. Ray, Ulrich H. E. Hansmann
Summary: Simulations of protein folding and protein association typically take much longer timescales than can be covered in all-atom molecular dynamics simulations. To address this challenge, researchers have introduced a replica-exchange-based multiscale sampling technique that combines the faster sampling in coarse-grained simulations with the potentially higher accuracy of all-atom simulations. By testing the efficiency of this technique in simulations of the Trp-cage protein and comparing the landscapes of wild-type and A2T mutant A beta(1-42) peptides, the researchers found a mechanism by which a small hydrophobic alanine (A) mutation to a bulky polar threonine (T) may interfere with the self-assembly of A beta fibrils.
Article
Chemistry, Multidisciplinary
V. S. Marangoni, M. C. F. Costa, P. R. Ng, H. T. L. Nguyen, M. Trushin, A. Carvalho, X. Zhao, S. J. Pennycook, R. K. Donato, A. H. Castro Neto
Summary: The synthesis of composite nanofibres using functionalised graphene structures in liquid medium as building blocks involves simultaneous scrolling and reacting 2D electrolytes, resulting in a dimensional reduction of 2D materials into one-dimensional nanostructures. The spontaneous self-assembly and cross-linking processes enable the production of nanofibres without the need for fibrillation techniques, such as wet-spinning or external templates.
MATERIALS TODAY CHEMISTRY
(2021)
Article
Materials Science, Multidisciplinary
Weifeng Zhao, Huihui Zhou, Gai Zhang, Aijie Ma, Weixing Chen
Summary: Microspheres composed of H2SO4-protonated graphitic carbon nitride (g-C3N4) nanosheets were prepared through a template-free dialysis method, showing improved photocatalytic performance towards Rhodamine B degradation. Soft solution assembly of g-C3N4 nanosheets in a one pot manner demonstrated a new strategy for constructing g-C3N4 nanostructures.
Article
Biochemical Research Methods
Zhen Tian, Yue Yu, Haichuan Fang, Weixin Xie, Maozu Guo
Summary: In this study, a novel approach called SCSMDA is proposed for predicting microbe-drug associations. SCSMDA employs structure-enhanced contrastive learning and self-paced negative sampling to enhance the embeddings of microbes and drugs and select informative negative samples for training the classifier. The results show that SCSMDA outperforms other baseline methods in MDA prediction.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Mathematics, Interdisciplinary Applications
Tung-Huan Su, Jimmy Gaspard Jean, Chuin-Shan Chen
Summary: This paper presents a generalized data-driven identification (DDI) approach, the local-convexity DDI (LCDDI) method, for constructing high-quality material databases. The LCDDI method learns the local structures of material data and produces optimal material data points for solving stresses with given strains. Numerical experiments demonstrate the effectiveness of the LCDDI method in handling complex heterogeneous strain fields. Convergence studies show significant improvement in the results.
COMPUTATIONAL MECHANICS
(2023)
Article
Computer Science, Information Systems
Jianzhe Zhao, Mengbo Yang, Ronglin Zhang, Wuganjing Song, Jiali Zheng, Jingran Feng, Stan Matwin
Summary: This paper introduces a novel privacy-enhanced federated learning framework, Optimal LDP-FL, which achieves local differential privacy protection through client self-sampling and data perturbation mechanisms. The proposed method significantly reduces computational and communication costs while improving privacy and model accuracy.
Article
Biophysics
Heini Ijas, Tim Liedl, Veikko Linko, Gregor Posnjak
Summary: DNA self-assembly and DNA origami are reliable methods for organizing materials. We use temperature-controlled sample holders and standard spectrometers to monitor the assembly process in real time and determine the folding and melting temperatures of different DNA structures. Additionally, we investigate the digestion of DNA structures in the presence of DNase I.
BIOPHYSICAL JOURNAL
(2022)
Article
Chemistry, Multidisciplinary
Jianhua Wang, Yuhui Wei, Ping Zhang, Yue Wang, Qinglin Xia, Xiaoguo Liu, Shihua Luo, Jiye Shi, Jun Hu, Chunhai Fan, Bin Li, Lihua Wang, Xingfei Zhou, Jiang Li
Summary: In this study, the folding processes of several multidomain DNA origami structures were visualized using atomic force microscopy under ambient annealing conditions in solution, revealing the coexistence of diverse transitional structures that might result in the same prescribed products. Based on experimental observations and energy landscape simulations, the heterogeneity of the folding pathways of multidomain DNA origami structures was proposed.
Article
Chemistry, Physical
Klara Markova, Antonin Kunka, Klaudia Chmelova, Martin Havlasek, Petra Babkova, Sergio M. Marques, Michal Vasina, Joan Planas-Iglesias, Radka Chaloupkova, David Bednar, Zbynek Prokop, Jiri Damborsky, Martin Marek
Summary: Enzyme functionality depends on its unique three-dimensional structure, with computational algorithms being used to stabilize proteins for research and biotech applications. In a unique case, introducing eleven stabilizing mutations affected the protein folding energy landscape and resulted in advantageous catalytic domain-swapped dimers.
Article
Geosciences, Multidisciplinary
Haitham Abdulmohsin Afan, Ayman Yafouz, Ahmed H. Birima, Ali Najah Ahmed, Ozgur Kisi, Barkha Chaplot, Ahmed El-Shafie
Summary: This research investigates two sampling approaches in river streamflow prediction using deep learning algorithms. The results show that stratified-deep learning (SDL) outperforms linear-deep learning (LDL) in terms of accuracy across multiple assessment criteria.
Article
Chemistry, Physical
Maximilian Topel, Ayesha Ejaz, Allison Squires, Andrew L. Ferguson
Summary: Single-molecule Fo''rster resonance energy transfer (smFRET) is used to track the real-time dynamics of molecules. The Takens' Delay Embedding Theorem guarantees that the atomistic dynamics of a system can be represented by a time-delayed embedding of scalar observables. A method called Single-molecule TAkens Reconstruction (STAR) is used to learn the transformation between atomic coordinates and delay-embedded distances accessible to smFRET. STAR has been applied to reconstruct molecular configurations with high accuracy. In this work, the role of signal-to-noise ratio, data volume, and time resolution in simulated smFRET data is investigated to assess the performance of STAR under experimental conditions.
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
(2023)
Editorial Material
Chemistry, Physical
Andrew L. Ferguson, Jim Pfaendtner
JOURNAL OF PHYSICAL CHEMISTRY A
(2023)
Editorial Material
Chemistry, Physical
Andrew L. Ferguson, Jim Pfaendtner
JOURNAL OF PHYSICAL CHEMISTRY B
(2023)
Editorial Material
Chemistry, Physical
Andrew L. Ferguson, Jim Pfaendtner
JOURNAL OF PHYSICAL CHEMISTRY C
(2023)
Article
Chemistry, Physical
Yutao Ma, Rohan Kapoor, Bineet Sharma, Allen P. Liu, Andrew L. Ferguson
Summary: In this work, an active learning screening method is constructed to identify peptide vesicle sequences with high thermodynamic stabilities. These peptide vesicles are more robust and stable compared to traditional lipid vesicles.
MOLECULAR SYSTEMS DESIGN & ENGINEERING
(2023)
Article
Biochemistry & Molecular Biology
Mingfei Zhao, Shuai Zhang, Renyu Zheng, Sarah Alamdari, Christopher J. Mundy, Jim Pfaendtner, Lilo D. Pozzo, Chun-Long Chen, James J. De Yoreo, Andrew L. Ferguson
Summary: In this study, the mechanical properties of amphiphilic diblock peptoids were investigated using molecular dynamics simulations and atomic force microscopy. The computational predictions were found to be in good agreement with experimental measurements. Additionally, a theoretical model was developed to analyze the stability of nanotubes.
Editorial Material
Chemistry, Physical
Andrew L. Ferguson, Jim Pfaendtner
JOURNAL OF PHYSICAL CHEMISTRY A
(2023)
Article
Chemistry, Physical
Michael S. S. Jones, Zachary A. A. McDargh, Rafal P. P. Wiewiora, Jesus A. A. Izaguirre, Huafeng Xu, Andrew L. L. Ferguson
Summary: All atom molecular dynamics (MD) simulations are limited by short time steps, but Markov state modeling (MSM) and latent space simulators (LSS) offer alternative approaches to extend the time scales and improve sampling of rare events and metastable states. In this work, LSS models are developed for both single-molecule and multi-molecule systems, generating long continuous trajectories with reduced computational cost. These trajectories provide valuable insights for therapeutic design and optimization of molecules and improve precision in studying DNA folding dynamics.
JOURNAL OF PHYSICAL CHEMISTRY A
(2023)
Article
Chemistry, Physical
Kirill Shmilovich, Andrew L. Ferguson
Summary: The efficacy of collective variable (CV) enhanced sampling techniques relies on the selection of good CVs correlated with the long-time dynamical behavior of the system. In this work, we introduce Girsanov Reweighting Enhanced Sampling Technique (GREST) as an adaptive sampling scheme that discovers slow CVs and drives sampling of the system. GREST is demonstrated on various systems and provides a publicly available Python package.
JOURNAL OF PHYSICAL CHEMISTRY A
(2023)
Editorial Material
Chemistry, Physical
Andrew L. Ferguson, Jim Pfaendtner
JOURNAL OF PHYSICAL CHEMISTRY B
(2023)
Article
Multidisciplinary Sciences
Brennan Ashwood, Michael S. Jones, Andrew L. Ferguson, Andrei Tokmakoff
Summary: DNA duplex stability is affected by cooperative interactions between adjacent nucleotides and is destabilized by abasic site, which splits the cooperativity in a short duplex into two segments and introduces additional barriers to hybridization. This study used temperature-jump infrared spectroscopy and molecular dynamics simulations to investigate the destabilization mechanism and hybridization dynamics caused by an abasic site in small DNA duplexes.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Article
Chemistry, Multidisciplinary
Walter Alvarado, Vasundhara Agrawal, Wing Shun Li, Vinayak P. P. Dravid, Vadim Backman, Juan J. J. de Pablo, Andrew L. L. Ferguson
Summary: We developed a denoising autoencoder (DAE) that outperforms other algorithms in providing high-resolution STEM images of nucleosomes and organized domains within chromatin dense regions. The DAE removes noise commonly found in high-angle annular dark field (HAADF) STEM experiments and learns structural features driven by chromatin folding physics. It allows for the resolution of alpha-tetrahedron tetranucleosome motifs that mediate DNA accessibility and does not find evidence for the suggested 30 nm fiber as the higher-order structure of chromatin fiber.
ACS CENTRAL SCIENCE
(2023)
Article
Chemistry, Physical
Sarah Alamdari, Kaylyn Torkelson, Xiaoqian Wang, Chun-Long Chen, Andrew L. Ferguson, Jim Pfaendtner
Summary: In this study, a simulation method is used to investigate the complex folding structures of various 12-mer polypeptoids, resulting in a predictive model that links sidechain chemistry with secondary structure formation. The results show that the assembly of Nrpe and Nspe sequences into polyproline type-I helices in water is enthalpically driven, with minor entropic gains from isomerization and steric strain due to the presence of the chiral center. The overall assembly into a helix is found to be entropically unfavorable, highlighting the importance of considering competing interactions in the rational design of peptoid secondary structure building blocks.
JOURNAL OF PHYSICAL CHEMISTRY B
(2023)
Article
Biochemical Research Methods
Niksa Praljak, Xinran Lian, Rama Ranganathan, Andrew L. Ferguson
Summary: In this work, a novel and lightweight deep generative model called ProtWave-VAE is proposed, which combines the strengths of variational autoencoders (VAEs) and autoregressive (AR) models. The model can be trained on unaligned protein sequence data and generate variable length protein sequences. Experimental results demonstrate that the model can infer meaningful functional and phylogenetic relationships, and achieve high accuracy in protein design predictions.
ACS SYNTHETIC BIOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Yifeng Tang, Jeremiah Y. Kim, Carman K. M. Ip, Azadeh Bahmani, Qing Chen, Matthew G. Rosenberger, Aaron P. Esser-Kahn, Andrew L. Ferguson
Summary: A machine learning-enabled active learning pipeline was developed to guide the screening and discovery of small molecule immunomodulators that can improve immune responses. By using high throughput screening and data-driven predictive models, novel small molecules with enhanced or suppressed innate immune signaling capacity were discovered, and chemical design rules were extracted.
Article
Chemistry, Physical
B. P. Akhouri, R. Perween, J. R. Solana
Summary: Monte Carlo simulations are used to obtain the equation of state and internal energy of fluids with Mie n-m potentials, and its performance is tested against a third order perturbation theory. The theory is then applied to tune the potentials for real fluids and achieve accurate fit with experimental data.
MOLECULAR SIMULATION
(2024)
Article
Chemistry, Physical
Malaisamy Veerapandian, Nagarajan Hemavathy, Alagesan Karthika, Jayaraman Manikandan, Umashankar Vetrivel, Jeyaraman Jeyakanthan
Summary: This study investigates the conformational stability and flexibility of SpeB enzyme and its interactions with substrate. The research finds that neutral pH 7 and alkaline pH 11 are the optimal conditions for stable binding between SpeB and substrate.
MOLECULAR SIMULATION
(2024)
Article
Chemistry, Physical
Maipelo Nyepetsi, Olayinka A. Oyetunji, Foster Mbaiwa
Summary: Biodiesel, a potential alternative to fossil-based fuels, has limitations such as high viscosity, pour point, and cloud point. This study used ReaxFF molecular dynamics to investigate the decarboxylation of methyl palmitate using different catalysts. The presence of alpha-NiMoO4 and beta-NiMoO4 accelerated the reactions and resulted in higher quantities of stable products. Ni3Mo catalyst showed an initial rapid formation of products followed by a decrease. All reactions followed first-order kinetics, and the catalysts reduced the activation energies.
MOLECULAR SIMULATION
(2024)
