Search

Optimisation-based predictive models are widely-used to explore the lifting strategies. Existing models incorporated empirical subject-specific posture constraints to improve the prediction accuracy. However, over-reliance on these constraints limits the application of predictive models. This paper proposed a multi-phase optimisation method (MPOM) for two-dimensional sagittally symmetric semi-squat lifting prediction, which decomposes the complete lifting task into three phases-the initial posture, the final posture, and the dynamic lifting phase. The first two phases are predicted with force- and stability-related strategies, and the last phase is predicted with a smoothing-related objective. Box-lifting motions of different box initial heights were collected for validation. The results show that MPOM has better or similar accuracy than the traditional single-phase optimisation (SPOM) of minimum muscular utilisation ratio, and MPOM reduces the reliance on experimental data. MPOM offers the opportunity to improve accuracy at the expense of efforts to determine appropriate weightings in the posture prediction phases.

Understanding the interfacial behaviors of biomolecules is crucial to applications in biomaterials and nanoparticle-based biosensing technologies. In this work, we utilized autoencoder-based graph clustering to analyze discontinuous molecular dynamics (DMD) simulations of lysozyme adsorption on a graphene surface. Our high-throughput DMD simulations integrated with a Go̅-like protein-surface interaction model makes it possible to explore protein adsorption at a large temporal scale with sufficient accuracy. The graph autoencoder extracts a low-dimensional feature vector from a contact map. The sequence of the extracted feature vectors is then clustered, and thus the evolution of the protein molecule structure in the absorption process is segmented into stages. Our study demonstrated that the residue-surface hydrophobic interactions and the π-π stacking interactions play key roles in the five-stage adsorption. Upon adsorption, the tertiary structure of lysozyme collapsed, and the secondary structure was also affected. The folding stages obtained by autoencoder-based graph clustering were consistent with detailed analyses of the protein structure. The combination of machine learning analysis and efficient DMD simulations developed in this work could be an important tool to study biomolecules' interfacial behaviors.

A manual material handling task involves the phases of reaching, lifting, unloading, and standing up (RLUS). Understanding the mechanisms of manual material handling is important for occupational health and the development of assist devices. Predictive models are becoming popular in exploring which performance criterion is appropriate in the lifting phase. However, limited attempts have been performed on the other phases. The associated performance criterion for predicting other phases is unknown. In this study, an optimization model for predicting RLUS has been developed with the multi-objective optimization method. Two performance criteria (minimum dynamic effort and maximum balance) were studied to explore their importance in each phase. The result shows that maximum balance leads to joint angle errors 27.6% and 40.9% smaller than minimum dynamic effort in reaching and unloading phases, but 40.4% and 65.9% larger in lifting and standing up phases. When the two performance criteria are combined, the maximum balance could help improve the predicting accuracy in the reaching, lifting, and unloading phases. These findings suggest that people prefer different performance criteria in different phases. This study helps understand the differences in motion strategies in manual materials handling (MMH), which would be used to develop a more accurate predictive model.

In recent years, convolutional neural networks (CNNs) have become widely adopted for computer vision tasks. FPGAs have been adequately explored as a promising hardware accelerator for CNNs owing to their high performance, energy efficiency, and reconfigurability. However, previous FPGA methods, which are based on the conventional convolutional algorithm, are often bounded by the computational capability of FPGAs. This paper first introduces four convolution algorithms: 6-loop algorithm, general matrix-matrix multiplication (GEMM), Winograd algorithm, and fast Fourier transform (FFT) algorithm. Then, we present the implementations of these algorithms at home and abroad, and also introduce their corresponding optimization techniques.

Fundamental understanding of biomolecular interfacial behavior, such as protein adsorption at the microscopic scale, is critical to broad applications in biomaterials, nanomedicine, and nanoparticle-based biosensing techniques. The goal of achieving both computational efficiency and accuracy presents a major challenge for simulation studies at both atomistic and molecular scales. In this work, we developed a unique, accurate, high-throughput simulation method which, by integrating discontinuous molecular dynamics (DMD) simulations with the Go-like protein-surface interaction model, not only solves the dynamics efficiently, but also describes precisely the protein intramolecular and intermolecular interactions at the atomistic scale and the protein-surface interactions at the coarse-grained scale. Using our simulation method and in-house developed software, we performed a systematic study of α-helical ovispirin-1 peptide adsorption on a graphene surface, and our study focused on the effect of surface hydrophobic interactions and π-π stacking on protein adsorption. Our DMD simulations were consistent with full-atom molecular dynamics simulations and showed that a single ovispirin-1 peptide lay down on the flat graphene surface with randomized secondary structure due to strong protein-surface interactions. Peptide aggregates were formed with an internal hydrophobic core driven by strong interactions of hydrophobic residues in the bulk environment. However, upon adsorption, the hydrophobic graphene surface can break the hydrophobic core by denaturing individual peptide structures, leading to disassembling the aggregate structure and further randomizing the ovispirin-1 peptide's secondary structures.

The strategy of solar-driven steam generation to collect purified water shows great potential in seawater desalination and domestic wastewater treatment. However, there is still a technical gap in achieving high cost-effectiveness and providing sustainable solutions for rapid evaporation to obtain high-quality and safe water. In this study, a natural rich biomass pectin (Nicandra physalodes (Linn.) Gaertn. polysaccharide) and polypyrrole photo-thermal material were introduced into polyvinyl alcohol network to achieve the construction of a low-cost hybrid hydrogel solar evaporator (total material cost of $7.89 /m2). The mixed evaporator works stably in a long-span acid-base range (pH 1 ~ 14) and salinity range (0 ~ 320 g/kg) with a water purification capacity of 2.42 (kg/m2* h) under 1 sun, and its daily water purification capacity can reach 24.4 kg/m2 under natural light. This paper provides a new possibility for a highly efficient and cost-effective water desalination system with guaranteed water quality.

A belt-type passive exoskeleton equipped with multiple working modes has been designed to support the back muscles during manual lifting tasks. Our concept is to develop a wearable assistive device that can provide motion-based assistance like the existing passive devices in the down phase of lifting, but in the up phase, can supply more load-based power. To achieve this goal, we designed two purely mechanical control mechanisms that can preload the load-based assist force and release it when the wearer intends to erect the trunk. This paper presents mathematical proof using a linked segment model and moment balance equations in the sagittal plane. Simulation proof is also provided based on a simple musculoskeletal model which executed sagittal plane lifting with a 20kg load in hands in stoop posture under three different conditions (no assist, classic mode, enhanced mode). The subsequent results suggest that the device substantially reduces the muscle force and lumbar moment by 27.7-43.5%, and in the enhanced mode, the efficacy is better than the classic mode. No such a quasi-passive back-support exoskeleton was found at the time of writing. This design concept is promising to help reduce the risk of back injuries in heavy load lifting and works that need workers to keep a forward bending and static holding postures such as vehicle assembly.

Poly-proline-arginine (poly-PR) and poly-glycine-arginine (poly-GR) proteins are believed to be the most toxic dipeptide repeat (DPR) proteins that are expressed by the hexanucleotide repeat expansion mutation in C9ORF72, which are associated with amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) diseases. Their structural information and mechanisms of toxicity remain incomplete, however. Using molecular dynamics simulation and all-atom model of proteins, we study folding and aggregation of both poly-PR and poly-GR. The results indicate formation of double-helix structure during the aggregation of poly-PR into dimers, whereas no stable aggregate is formed during the aggregation of poly-GR; the latter only folds into α-helix and double-helix structures that are similar to those formed in the folding of poly-glycine-alanine (poly-GA) protein. Our findings are consistent with the experimental data indicating that poly-PR and poly-GR are less likely to aggregate because of the hydrophilic arginine residues within their structures. Such characteristics could, however, in some respect facilitate migration of the DPR proteins between and within cells and, at the same time, give proline residues the benefits of activating the receptors that regulate ionotropic effect in neurons, resulting in death or malfunction of neurons because of the abnormal increase or decrease of the ion transmission. This may explain the neurotoxicities of poly-PR and poly-GR associated with many neurodegenerative diseases. To our knowledge, this is the first molecular dynamics simulation of the phenomena involving poly-PR and poly-GR proteins.

Poly-glycine-alanine (poly-GA) proteins are widely believed to be one of the main toxic dipeptide repeat molecules associated with amyotrophic lateral sclerosis (ALS) and frontotemporal dementia diseases. Using discontinuous molecular dynamics simulation and an all-atom model of the proteins, we study folding, stability, and aggregation of poly-GA. The results demonstrate that poly-GA is an aggregation-prone protein that, after a long enough time, forms β-sheet-rich aggregates that match recent experiment data and that two unique helical structures are formed very frequently, namely, β-helix and double-helix. The details of the two structures are analyzed. The analysis indicates that such helical structures are stable and share the characteristics of both α-helices and β-sheets. Molecular simulations indicate that identical phenomena also occur in the aggregation of poly-glycine-arginine (poly-GR). Therefore, we hypothesize that proteins of type (GX)n in which X may be any non-glycine amino acid and n is the repeat length may share the same folding structures of β-helix and double-helix and that it is the glycine in the repeat that contributes the most to this characteristic. Molecular dynamics simulation with continuous interaction potentials and explicit water molecules as the solvent supports the hypothesis. To our knowledge, this is the first molecular dynamics simulation of the phenomena involving poly-GA and poly-GR proteins.

In this paper, the second in a series devoted to molecular modeling of protein aggregation, a mesoscale model of proteins together with extensive discontinuous molecular dynamics simulation is used to study the phenomenon in a confined medium. The medium, as a model of a crowded cellular environment, is represented by a spherical cavity, as well as cylindrical tubes with two aspect ratios. The aggregation process leads to the formation of β sheets and eventually fibrils, whose deposition on biological tissues is believed to be a major factor contributing to many neuro-degenerative diseases, such as Alzheimer's, Parkinson's, and amyotrophic lateral sclerosis diseases. Several important properties of the aggregation process, including dynamic evolution of the total number of the aggregates, the mean aggregate size, and the number of peptides that contribute to the formation of the β sheets, have been computed. We show, similar to the unconfined media studied in Paper I [S. Zheng et al., J. Chem. Phys. 145, 134306 (2016)], that the computed properties follow dynamic scaling, characterized by power laws. The existence of such dynamic scaling in unconfined media was recently confirmed by experiments. The exponents that characterize the power-law dependence on time of the properties of the aggregation process in spherical cavities are shown to agree with those in unbounded fluids at the same protein density, while the exponents for aggregation in the cylindrical tubes exhibit sensitivity to the geometry of the system. The effects of the number of amino acids in the protein, as well as the size of the confined media, have also been studied. Similarities and differences between aggregation in confined and unconfined media are described, including the possibility of no fibril formation, if confinement is severe.