This work presents Neural Equivariant Interatomic Potentials (NequIP), an E(3)-equivariant neural network approach for learning interatomic potentials from ab-initio calculations for molecular ...dynamics simulations. While most contemporary symmetry-aware models use invariant convolutions and only act on scalars, NequIP employs E(3)-equivariant convolutions for interactions of geometric tensors, resulting in a more information-rich and faithful representation of atomic environments. The method achieves state-of-the-art accuracy on a challenging and diverse set of molecules and materials while exhibiting remarkable data efficiency. NequIP outperforms existing models with up to three orders of magnitude fewer training data, challenging the widely held belief that deep neural networks require massive training sets. The high data efficiency of the method allows for the construction of accurate potentials using high-order quantum chemical level of theory as reference and enables high-fidelity molecular dynamics simulations over long time scales.
Strained oxide thin films are of interest for accelerating oxide ion conduction in electrochemical devices. Although the effect of elastic strain has been uncovered theoretically, the effect of ...dislocations on the diffusion kinetics in such strained oxides is yet unclear. Here we investigate a 1/2 {100} edge dislocation by performing atomistic simulations in 4-12% doped CeO2 as a model fast ion conductor. At equilibrium, depending on the size of the dopant, trivalent cations and oxygen vacancies are found to simultaneously enrich or deplete either in the compressive or in the tensile strain fields around the dislocation. The associative interactions among the point defects in the enrichment zone and the lack of oxygen vacancies in the depletion zone slow down oxide ion transport. This finding is contrary to the fast diffusion of atoms along the dislocations in metals and should be considered when assessing the effects of strain on oxide ion conductivity.
A simultaneously accurate and computationally efficient parametrization of the potential energy surface of molecules and materials is a long-standing goal in the natural sciences. While atom-centered ...message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited the accessible length-scales. Local methods, conversely, scale to large simulations but have suffered from inferior accuracy. This work introduces Allegro, a strictly local equivariant deep neural network interatomic potential architecture that simultaneously exhibits excellent accuracy and scalability. Allegro represents a many-body potential using iterated tensor products of learned equivariant representations without atom-centered message passing. Allegro obtains improvements over state-of-the-art methods on QM9 and revMD17. A single tensor product layer outperforms existing deep MPNNs and transformers on QM9. Furthermore, Allegro displays remarkable generalization to out-of-distribution data. Molecular simulations using Allegro recover structural and kinetic properties of an amorphous electrolyte in excellent agreement with ab-initio simulations. Finally, we demonstrate parallelization with a simulation of 100 million atoms.
The incidence of intrahepatic cholangiocarcinoma (iCCA) continues to increase worldwide, however its molecular pathogenesis remains poorly understood. Recent studies have implicated microRNAs in iCCA ...progression. In this study, we demonstrated that miR‐885‐5p was significantly decreased in iCCA tissues. Downregulation of miR‐885‐5p was correlated with vascular invasion, lymph node metastasis, unfavorable overall survival, and shorter disease‐free survival. Silencing or overexpressing miR‐885‐5p by lentiviral approaches significantly influenced iCCA cell proliferation and metastasis in vitro and in vivo. Mechanistically, overexpression of miR‐885‐5p inhibited iCCA metastasis and proliferation by directly inhibiting GALNT3 as well as by indirectly promoting the downregulation of insulin‐like growth factor‐2 mRNA‐binding protein 1 (IGF2BP1). Furthermore, miR‐885‐5p inhibited iCCA metastasis by downregulating the PI3K/AKT/MMPs signaling pathway via targeting GALNT3. Collectively, we demonstrated that miR‐885‐5p was an important mediator of iCCA proliferation and metastasis by regulating GALNT3 and IGF2BP1, thus offering a potential target for iCCA treatment.
Tumor recurrence, the chief reason for poor prognosis of glioma, is largely attributed to glioma stem cells (GSCs) and epithelial-mesenchymal transition (EMT). However, the mechanisms among them ...remain unknown. Here, we determined whether leucine-rich repeat-containing G protein-coupled receptor 5 (LGR5), known as a stem cell marker for colon cancer and gastric cancer, can serve as a novel GSC marker involved in EMT and a therapeutic target in glioma.
Stemness properties were examined in FACS-isolated LGR5
/LGR5
cells. Reported stem cell markers, EMT and the Wnt/β-catenin pathway were examined in stable LGR5 knockdown or overexpressed GSCs by Western Blot. The treatment experiment was performed in an intracranial orthotopic xenograft model by knockdown of LGR5 or by using the Wnt/β-catenin pathway inhibitor Wnt-C59. LGR5 expression was determined in 268 glioma specimens by immunohistochemistry.
LGR5
cells possessed stronger stemness properties compared to LGR5
cells. The expression of SOX2, Nanog, CD133, CD44, CD24 and EpCAM was modulated by LGR5. Both LGR5 knockdown and Wnt-C59 reduced tumor invasion and migration and blocked EMT by inhibiting the Wnt/β-catenin pathway in vitro and suppressed the intracranial orthotopic xenograft growth and prolonged the survival of xenograft mice in vivo. Moreover, LGR5 was positively correlated with Ki67, N-cadherin and WHO grade and negatively correlated with IDH1. Glioma patients with high expression of LGR5 showed significantly poorer prognosis.
LGR5 is a new functional GSC marker and prognostic indicator that can promote EMT by activating the Wnt/β-catenin pathway and would thus be a novel therapeutic target for glioma.
Considering the correlation among the conditioning factors, this study explores the impact of the orthogonal transformation for factors on model performance, with the LR model. Landslide inventory ...and factor theme maps were first constructed and used for factor analysis. Subsequently, effective factors with high correlation were applied for orthogonal transformation, to obtain the reconstructed factors. Meanwhile, reconstructed factors and the remaining effective factors were jointly involved in the model construction. Finally, differences in model performance before and after the transformation were discussed, in terms of the predictive ability and computational ability. Results indicate that the modified model maintains high predictive ability compared with the initial model, i.e., high values of AUC (0.939), specificity (0.956), sensitivity (0.977), and accuracy (0.968) in the training phase, and AUC (0.928), specificity (0.791), sensitivity (0.866), and accuracy (0.849) in the test phase. Besides, the modified model not only exhibits high operation speed but also smaller standard errors of the regression coefficients compared with the initial model, overall. The high predictive and computational performance of the modified susceptibility model provides a window for optimization of the landslide susceptibility in the modeling workflow.
The Perception Neuron Studio (PNS) is a cost-effective and widely used inertial motion capture system. However, a comprehensive analysis of its upper-body motion capture accuracy is still lacking, ...before it is being applied to biomechanical research. Therefore, this study first evaluated the validity and reliability of this system in upper-body capturing and then quantified the system’s accuracy for different task complexities and movement speeds. Seven participants performed simple (eight single-DOF upper-body movements) and complex tasks (lifting a 2.5 kg box over the shoulder) at fast and slow speeds with the PNS and OptiTrack (gold-standard optical system) collecting kinematics data simultaneously. Statistical metrics such as CMC, RMSE, Pearson’s r, R2, and Bland–Altman analysis were utilized to assess the similarity between the two systems. Test–retest reliability included intra- and intersession relations, which were assessed by the intraclass correlation coefficient (ICC) as well as CMC. All upper-body kinematics were highly consistent between the two systems, with CMC values 0.73–0.99, RMSE 1.9–12.5°, Pearson’s r 0.84–0.99, R2 0.75–0.99, and Bland–Altman analysis demonstrating a bias of 0.2–27.8° as well as all the points within 95% limits of agreement (LOA). The relative reliability of intra- and intersessions was good to excellent (i.e., ICC and CMC were 0.77–0.99 and 0.75–0.98, respectively). The paired t-test revealed that faster speeds resulted in greater bias, while more complex tasks led to lower consistencies. Our results showed that the PNS could provide accurate enough upper-body kinematics for further biomechanical performance analysis.
Machine learned force fields typically require manual construction of training sets consisting of thousands of first principles calculations, which can result in low training efficiency and ...unpredictable errors when applied to structures not represented in the training set of the model. This severely limits the practical application of these models in systems with dynamics governed by important rare events, such as chemical reactions and diffusion. We present an adaptive Bayesian inference method for automating the training of interpretable, low-dimensional, and multi-element interatomic force fields using structures drawn on the fly from molecular dynamics simulations. Within an active learning framework, the internal uncertainty of a Gaussian process regression model is used to decide whether to accept the model prediction or to perform a first principles calculation to augment the training set of the model. The method is applied to a range of single- and multi-element systems and shown to achieve a favorable balance of accuracy and computational efficiency, while requiring a minimal amount of ab initio training data. We provide a fully open-source implementation of our method, as well as a procedure to map trained models to computationally efficient tabulated force fields.
Virtual reality (VR) technology offers a great opportunity to explore stress disorder therapies. We created a VR stress training system, which incorporates three highly interactive stressful scenes ...to elicit stress, and demonstrate the concurrent variations between physiological data (heart rate, electrodermal activity and eye-blink rate) and self-reported stress ratings through a self-designed customized perceived stress questionnaire (SSAI) and wearable devices. Several supervised learning models were rigorously applied to automate stress recognition. Our findings include the evaluations of the VR system by computing Cronbach's alpha (<inline-formula> <tex-math notation="LaTeX">\alpha =0.72 </tex-math></inline-formula>) and Kaiser-Meyer-Olkin (KMO) coefficient (<inline-formula> <tex-math notation="LaTeX">\eta =0.78 </tex-math></inline-formula>) through a retrospective survey, which were subsequently confirmed as reliable on four aspects (sense of presence, sense of space, sense of immersion and sense of reality) via factor analysis. Additionally, we demonstrate the effectiveness of physiology-based stress level classification (no stress, low stress and high stress) and continuous SSAI score prediction, with accuracy reaching 0.742 by bagging ensemble learning model and goodness-of-fit reaching 0.44 via multivariate stepwise regression. This study provides detailed insight regarding the effect of objective physiological measures on the validation of subjective self-ratings under a novel complex VR stress training system, which stimulates the further investigations of stress disorder recognition and treatment.
Bile duct obstruction is a potent stimulus for cholangiocyte proliferation, especially for large cholangiocytes. Our previous studies reported that conjugated bile acids (CBAs) activate the protein ...kinase B (AKT) and extracellular signal‐regulated kinase 1 and 2 (ERK1/2) signaling pathways through sphingosine 1‐phosphate receptor (S1PR) 2 in hepatocytes and cholangiocarcinoma cells. It also has been reported that taurocholate (TCA) promotes large cholangiocyte proliferation and protects cholangiocytes from bile duct ligation (BDL)‐induced apoptosis. However, the role of S1PR2 in bile‐acid–mediated cholangiocyte proliferation and cholestatic liver injury has not been elucidated. Here, we report that S1PR2 is the predominant S1PR expressed in cholangiocytes. Both TCA‐ and sphingosine‐1‐phosphate (S1P)‐induced activation of ERK1/2 and AKT were inhibited by JTE‐013, a specific antagonist of S1PR2, in cholangiocytes. In addition, TCA‐ and S1P‐induced cell proliferation and migration were inhibited by JTE‐013 and a specific short hairpin RNA of S1PR2, as well as chemical inhibitors of ERK1/2 and AKT in mouse cholangiocytes. In BDL mice, expression of S1PR2 was up‐regulated in whole liver and cholangiocytes. S1PR2 deficiency significantly reduced BDL‐induced cholangiocyte proliferation and cholestatic injury, as indicated by significant reductions in inflammation and liver fibrosis in S1PR2 knockout mice. Treatment of BDL mice with JTE‐013 significantly reduced total bile acid levels in serum and cholestatic liver injury. Conclusion: This study suggests that CBA‐induced activation of S1PR2‐mediated signaling pathways plays a critical role in obstructive cholestasis and may represent a novel therapeutic target for cholestatic liver diseases. (Hepatology 2017;65:2005‐2018).