ZetaDesign: an end-to-end deep learning method for protein sequence design and side-chain packing.

Computational protein design has been demonstrated to be the most powerful tool in the last few years among protein designing and repacking tasks. In practice, these two tasks are strongly related but often treated separately. Besides, state-of-the-art deep-learning-based methods cannot provide interpretability from an energy perspective, affecting the accuracy of the design. Here we propose a new systematic approach, including both a posterior probability and a joint probability parts, to solve the two essential questions once for all. This approach takes the physicochemical property of amino acids into consideration and uses the joint probability model to ensure the convergence between structure and amino acid type. Our results demonstrated that this method could generate feasible, high-confidence sequences with low-energy side conformations. The designed sequences can fold into target structures with high confidence and maintain relatively stable biochemical properties. The side chain conformation has a significantly lower energy landscape without delegating to a rotamer library or performing the expensive conformational searches. Overall, we propose an end-to-end method that combines the advantages of both deep learning and energy-based methods. The design results of this model demonstrate high efficiency, and precision, as well as a low energy state and good interpretability.

Single-cell dissection of the multicellular ecosystem and molecular features underlying microvascular invasion in HCC.

BACKGROUND AND AIMS: Microvascular invasion (MVI) is a crucial pathological hallmark of HCC that is closely associated with poor outcomes, early recurrence, and intrahepatic metastasis following surgical resection and transplantation. However, the intricate tumor microenvironment and transcriptional programs underlying MVI in HCC remain poorly understood. APPROACH AND RESULTS: We performed single-cell RNA sequencing of 46,789 individual cells from 10 samples of MVI+ (MVI present) and MVI- (MVI absent) patients with HCC. We conducted comprehensive and comparative analyses to characterize cellular and molecular features associated with MVI and validated key findings using external bulk, single-cell, and spatial transcriptomic datasets coupled with multiplex immunofluorescence assays. The comparison identified specific subtypes of immune and stromal cells critical to the formation of the immunosuppressive and pro-metastatic microenvironment in MVI+ tumors, including cycling T cells, lysosomal associated membrane protein 3+ dendritic cells, triggering receptor expressed on myeloid cells 2+ macrophages, myofibroblasts, and arterial i endothelial cells. MVI+ malignant cells are characterized by high proliferation rates, whereas MVI- malignant cells exhibit an inflammatory milieu. Additionally, we identified the midkine-dominated interaction between triggering receptor expressed on myeloid cells 2+ macrophages and malignant cells as a contributor to MVI formation and tumor progression. Notably, we unveiled a spatially co-located multicellular community exerting a dominant role in shaping the immunosuppressive microenvironment of MVI and correlating with unfavorable prognosis. CONCLUSIONS: This study provides a comprehensive single-cell atlas of MVI in HCC, shedding light on the complex multicellular ecosystem and molecular features associated with MVI. These findings deepen our understanding of the underlying mechanisms driving MVI and provide valuable insights for improving clinical diagnosis and developing more effective treatment strategies.

Effect of remediation reagents on bacterial composition and ecological function in black-odorous water sediments.

Black-odorous urban water bodies and sediments pose a serious environmental problem. In this study, we conducted microcosm batch experiments to investigate the effect of remediation reagents (magnesium hydroxide and calcium nitrate) on native bacterial communities and their ecological functions in the black-odorous sediment of urban water. The dominant phyla (Proteobacteria, Actinobacteria, Chloroflexi, and Planctomycetes) and classes (Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria, Actinobacteria, Anaerolineae, and Planctomycetia) were determined under calcium nitrate and magnesium hydroxide treatments. Functional groups related to aerobic metabolism, including aerobic chemoheterotrophy, dark sulfide oxidation, and correlated dominant genera (Thiobacillus, Lysobacter, Gp16, and Gaiella) became more abundant under calcium nitrate treatment, whereas functional genes potentially involved in dissimilatory sulfate reduction became less abundant. The relative abundance of chloroplasts, fermentation, and correlated genera (Desulfomonile and unclassified Cyanobacteria) decreased under magnesium hydroxide treatment. Overall, these results indicated that calcium nitrate addition improved hypoxia-related reducing conditions in the sediment and promoted aerobic chemoheterotrophy.

Quantitative influence and performance analysis of virtual reality laparoscopic surgical training system.

BACKGROUND: Virtual reality (VR) surgery training has become a trend in clinical education. Many research papers validate the effectiveness of VR-based surgical simulators in training medical students. However, most existing articles employ subjective methods to study the residents' surgical skills improvement. Few of them investigate how to improve the surgery skills on specific dimensions substantially. METHODS: Our paper resorts to physiological approaches to objectively study the quantitative influence and performance analysis of VR laparoscopic surgical training system for medical students. Fifty-one participants were recruited from a pool of medical students. They conducted four pre and post experiments in the training box. They were trained on VR-based laparoscopic surgery simulators (VRLS) in the middle of pre and post experiments. Their operation and physiological data (heart rate and electroencephalogram) are recorded during the pre and post experiments. The physiological data is used to compute cognitive load and flow experience quantitatively. Senior surgeons graded their performance using newly designed hybrid standards for fundamental tasks and Global operative assessment of laparoscopic skills (GOALS) standards for colon resection tasks. Finally, the participants were required to fill the questionnaires about their cognitive load and flow experience. RESULTS: After training on VRLS, the time of the experimental group to complete the same task could drop sharply (p < 0.01). The performance scores are enhanced significantly (p < 0.01). The performance and cognitive load computed from EEG are negatively correlated (p < 0.05). CONCLUSION: The results show that the VRLS could highly improve medical students' performance and enable the participants to obtain flow experience with a lower cognitive load. Participants' performance is negatively correlated with cognitive load through quantitative physiological analysis. This might provide a new way of assessing skill acquirement.

Deep learning-based dental plaque detection on primary teeth: a comparison with clinical assessments.

BACKGROUND: Dental plaque causes many common oral diseases (e.g., caries, gingivitis, and periodontitis). Therefore, plaque detection and control are extremely important for children's oral health. The objectives of this study were to design a deep learning-based artificial intelligence (AI) model to detect plaque on primary teeth and to evaluate the diagnostic accuracy of the model. METHODS: A conventional neural network (CNN) framework was adopted, and 886 intraoral photos of primary teeth were used for training. To validate clinical feasibility, 98 intraoral photos of primary teeth were assessed by the AI model. Additionally, tooth photos were acquired using a digital camera. One experienced pediatric dentist examined the photos and marked the regions containing plaque. Then, a plaque-disclosing agent was applied, and the areas with plaque were identified. After 1 week, the dentist drew the plaque area on the 98 photos taken by the digital camera again to evaluate the consistency of manual diagnosis. Additionally, 102 intraoral photos of primary teeth were marked to denote the plaque areas obtained by the AI model and the dentist to evaluate the diagnostic capacity of each approach based on lower-resolution photos. The mean intersection-over-union (MIoU) metric was employed to indicate detection accuracy. RESULTS: The MIoU for detecting plaque on the tested tooth photos was 0.726 ± 0.165. The dentist's MIoU was 0.695 ± 0.269 when first diagnosing the 98 photos taken by the digital camera and 0.689 ± 0.253 after 1 week. Compared to the dentist, the AI model demonstrated a higher MIoU (0.736 ± 0.174), and the results did not change after 1 week. When the dentist and the AI model assessed the 102 intraoral photos, the MIoU was 0.652 ± 0.195 for the dentist and 0.724 ± 0.159 for the model. The results of a paired t-test found no significant difference between the AI model and human specialist (P > .05) in diagnosing dental plaque on primary teeth. CONCLUSIONS: The AI model showed clinically acceptable performance in detecting dental plaque on primary teeth compared with an experienced pediatric dentist. This finding illustrates the potential of such AI technology to help improve pediatric oral health.

Strategic management of Public Hospitals' medical services.

Purpose: The quality of medical services provided by competing public hospitals is the primary consideration of the public in determining the selection of a specific hospital for treatment. The main objective of strategic planning is to improve the quality of public hospital medical services. This paper provides an introduction to the history, significance, principles and practices of public hospital medical service strategy, as well as advancing the opinion that public hospital service strategy must not merely aim to produce but actually result in the highest possible level of quality, convenience, efficiency and patient satisfaction.

A novel gemycircularvirus in an unexplained case of child encephalitis.

BACKGROUND: Recently, a diverse group of viruses with circular, replication initiator protein(Rep) encoding, single stranded DNA (CRESS-DNA) genomes, were discovered from wide range of eukaryotic organisms ranging from mammals to fungi. Gemycircularvirus belongs to a distinct group of CRESS-DNA genomes and is classified under the genus name of Gemycircularvirus. FINDINGS: Here, a novel gemycircularvirus named GeTz1 from cerebrospinal fluid sample of a child with unexplainable encephalitis was characterized. The novel gemycircularvirus encodes two major proteins, including a capsid protein (Cap) and a replication-associated protein (Rep). Phylogenetic analysis based on the amino acid sequence of Rep indicated that GeTz1 clusters with one gemycircularvirus discovered from bird (KF371633), sharing 46.6 % amino acid sequence identity with each other. CONCLUSION: A novel gemycircularvirus was discovered from cerebrospinal fluid sample of a child with unexplainable encephalitis. Further studies, such as testing human sera for specific antibodies, should be performed to investigate whether gemycircularvirus infects human and is associated with encephalitis.

Antioxidant Responses of Vallisneria asiatica to Eutrophic Sediments in Lake Taihu, China.

Three kinds of representative sediments were obtained from a macrophyte-dominated bay (East Lake Taihu) and two algae-dominated regions (Western Lake Taihu and Meiliang Bay). Physiological responses of Vallisneria asiatica to these sediments were compared. Results from 20 days exposures showed no obvious differences in malondialdehyde (MDA) in roots, while the MDA content in leaves of plants exposed to Western Lake Taihu sediment was significantly (p<0.05) higher than those exposed to the other two sediments. In comparison to the other two sediments, plants exposed to Western Lake Taihu sediment showed significantly lower (p<0.05) superoxide dismutase in roots and leaves on the 10th and 40th day. On the 40th day, root catalase (CAT) activities in V. asiatica from Western Lake Taihu and Meiliang Bay sediments were lower than that from East Lake Taihu sediment, while leaf CAT activity in V. asiatica from Western Lake Taihu sediment was higher than that from East Lake Taihu sediment (p<0.05). Western Lake Taihu sediment caused more serious oxidative stress in V. asiatica than East Lake Taihu sediment. Results indicated eutrophic sediment was a contributing factor in the disappearance of V. asiatica in Western Lake Taihu.

Understanding the mechanisms behind the antibacterial activity of magnesium hydroxide nanoparticles against sulfate-reducing bacteria in sediments.

Nanomaterials, with their small size, surface characteristics, and antibacterial properties, are extensively employed across environmental, energy, biomedical, agricultural, and other industries. This study examined the antibacterial efficacy of magnesium hydroxide (Mg(OH)2) nanoparticles (NPs) against sulfate-reducing bacteria (SRB) within sediments. The inhibitory effects of two types of Mg(OH)2 NPs with distinct particle sizes (20.3 and 29.6 nm) and concentrations (0-10.0 mg/mL) were examined under optimal treatment conditions. The antibacterial mechanisms of Mg(OH)2 NPs through direct contact and dissolution effects were determined. The results revealed a correlation between the concentration, particle size, and inhibitory activity, with the smallest NPs (20.3 nm) at the highest concentration (10.0 mg/mL) substantially reducing SRB counts from 8.77 ± 0.18 to 6.48 ± 0.13 log10 colony forming units/mL after 6 h treatment. Treatment with high concentrations of Mg(OH)2 NPs induced cellular damage, reduced intracellular lactate dehydrogenase activity, and elevated intracellular catalase activity and H2O2 content, suggesting that the contact effect of NPs stimulated SRB. This leads to oxidative stress response and structural damage to the cell membrane, which has emerged as the primary driver of the antibacterial action of Mg(OH)2 NPs. This study presents a novel nanomaterial that can inhibit and control SRB in natural sedimentary environments.

Efficient Photon Beam Diffusion for Directional Subsurface Scattering.

Real-time subsurface scattering techniques are widely used in translucent material rendering. Among advanced methods that rely on the bidirectional scattering-surface reflectance distribution function (BSSRDF), screen space algorithms exhibit limited translucency, while existing large-distance methods are inefficient and yield poor illumination details. To address these limitations for better large-distance scattering, we develop a novel algorithm by extending the photon beam diffusion (PBD) model within the light view and screen space. Unlike surface irradiance in prior methods, we incorporate the refracted beam in the medium into real-time scattering estimation, presenting a new consideration for photon beam utilization. Concretely, we store all photon beam samples in light view textures and utilize an adaptive sampling pattern for beam sample selection in large filtering kernel sizes. This can reduce the sample count based on surface attributes. In screen space, virtual sources are derived from samples to estimate PBD contributions, with an approximation that preserves boundary conditions. To avoid possible overestimation, we implement correction factors that scale contributions, effectively aligning our results with path-tracing references. Through these reformulations, our efficient PBD generates results closest to references among existing methods. The experiments accurately represent better front-face illumination details and backlit translucency effects, while significantly accelerating performance compared to previous large-distance methods.

TalkingStyle: Personalized Speech-Driven 3D Facial Animation with Style Preservation.

It is a challenging task to create realistic 3D avatars that accurately replicate individuals' speech and unique talking styles for speech-driven facial animation. Existing techniques have made remarkable progress but still struggle to achieve lifelike mimicry. This paper proposes "TalkingStyle", a novel method to generate personalized talking avatars while retaining the talking style of the person. Our approach uses a set of audio and animation samples from an individual to create new facial animations that closely resemble their specific talking style, synchronized with speech. We disentangle the style codes from the motion patterns, allowing our method to associate a distinct identifier with each person. To manage each aspect effectively, we employ three separate encoders for style, speech, and motion, ensuring the preservation of the original style while maintaining consistent motion in our stylized talking avatars. Additionally, we propose a new style-conditioned transformer decoder, offering greater flexibility and control over the facial avatar styles. We comprehensively evaluate TalkingStyle through qualitative and quantitative assessments, as well as user studies demonstrating its superior realism and lip synchronization accuracy compared to current state-of-the-art methods. To promote transparency and further advancements in the field, we also make the source code publicly available at https://github.com/wangxuanx/TalkingStyle.

Fluid Inverse Volumetric Modeling and Applications from Surface Motion.

In this study, we devise a framework for volumetrically reconstructing fluid from observable, measurable free surface motion. Our innovative method amalgamates the benefits of deep learning and conventional simulation to preserve the guiding motion and temporal coherence of the reproduced fluid. We infer surface velocities by encoding and decoding spatiotemporal features of surface sequences, and a 3D CNN is used to generate the volumetric velocity field, which is then combined with 3D labels of obstacles and boundaries. Concurrently, we employ a network to estimate the fluid's physical properties. To progressively evolve the flow field over time, we input the reconstructed velocity field and estimated parameters into the physical simulator as the initial state. Our approach yields promising results for both synthetic fluid generated by different fluid solvers and captured real fluid. The developed framework naturally lends itself to a variety of graphics applications, such as 1) effective reproductions of fluid behaviors visually congruent with the observed surface motion, and 2) physics-guided re-editing of fluid scenes. Extensive experiments affirm that our novel method surpasses state-of-the-art approaches for 3D fluid inverse modeling and animation in graphics.

A Unified Particle-Based Solver for Non-Newtonian Behaviors Simulation.

In this article, we present a unified framework to simulate non-Newtonian behaviors. We combine viscous and elasto-plastic stress into a unified particle solver to achieve various non-Newtonian behaviors ranging from fluid-like to solid-like. Our constitutive model is based on a Generalized Maxwell model, which incorporates viscosity, elasticity and plasticity in one non-linear framework by a unified way. On the one hand, taking advantage of the viscous term, we construct a series of strain-rate dependent models for classical non-Newtonian behaviors such as shear-thickening, shear-thinning, Bingham plastic, etc. On the other hand, benefiting from the elasto-plastic model, we empower our framework with the ability to simulate solid-like non-Newtonian behaviors, i.e., visco-elasticity/plasticity. In addition, we enrich our method with a heat diffusion model to make our method flexible in simulating phase change. Through sufficient experiments, we demonstrate a wide range of non-Newtonian behaviors ranging from viscous fluid to deformable objects. We believe this non-Newtonian model will enhance the realism of physically-based animation, which has great potential for computer graphics.

Expressive 3D Facial Animation Generation Based on Local-to-Global Latent Diffusion.

3D Facial animations, crucial to augmented and mixed reality digital media, have evolved from mere aesthetic elements to potent storytelling media. Despite considerable progress in facial animation of neutral emotions, existing methods still struggle to capture the authenticity of emotions. This paper introduces a novel approach to capture fine facial expressions and generate facial animations using audio synchronization. Our method consists of two key components: First, the Local-to-global Latent Diffusion Model (LG-LDM) tailored for authentic facial expressions, which can integrate audio, time step, facial expressions, and other conditions towards possible encoding of emotionally rich yet latent features in response to possibly noisy raw audio signals. The core of LG-LDM is our carefully designed Facial Denoiser Model (FDM) for aligning the local-to-global animation feature with audio. Second, we redesign an Emotion-centric Vector Quantized-Variational AutoEncoder framework (EVQ-VAE) to finely decode the subtle differences under different emotions and reconstruct the final 3D facial geometry. Our work significantly contributes to the key challenges of emotionally realistic 3D facial animation for audio synchronization and enhances the immersive experience and emotional depth in augmented and mixed reality applications. We provide a reproducibility kit including our code, dataset, and detailed instructions for running the experiments. This kit is available at https://github.com/wangxuanx/Face-Diffusion-Model.

Pixel is All You Need: Adversarial Spatio-Temporal Ensemble Active Learning for Salient Object Detection.

Although weakly-supervised techniques can reduce the labeling effort, it is unclear whether a saliency model trained with weakly-supervised data (e.g., point annotation) can achieve the equivalent performance of its fully-supervised version. This paper attempts to answer this unexplored question by proving a hypothesis: there is a point-labeled dataset where saliency models trained on it can achieve equivalent performance when trained on the densely annotated dataset. To prove this conjecture, we proposed a novel yet effective adversarial spatio-temporal ensemble active learning. Our contributions are four- fold: 1) Our proposed adversarial attack triggering uncertainty can conquer the overconfidence of existing active learning methods and accurately locate these uncertain pixels. 2) Our proposed spatio-temporal ensemble strategy not only achieves outstanding performance but significantly reduces the model's computational cost. 3) Our proposed relationship-aware diversity sampling can conquer oversampling while boosting model performance. 4) We provide theoretical proof for the existence of such a point-labeled dataset. Experimental results show that our approach can find such a point-labeled dataset, where a saliency model trained on it obtained 98%-99% performance of its fully-supervised version with only ten annotated points per image. The code is available at https://github.com/wuzhenyubuaa/ASTE-AL.

IG-Net: An Instrument-guided real-time semantic segmentation framework for prostate dissection during surgery for low rectal cancer.

BACKGROUND AND OBJECTIVE: Accurate prostate dissection is crucial in transanal surgery for patients with low rectal cancer. Improper dissection can lead to adverse events such as urethral injury, severely affecting the patient's postoperative recovery. However, unclear boundaries, irregular shape of the prostate, and obstructive factors such as smoke present significant challenges for surgeons. METHODS: Our innovative contribution lies in the introduction of a novel video semantic segmentation framework, IG-Net, which incorporates prior surgical instrument features for real-time and precise prostate segmentation. Specifically, we designed an instrument-guided module that calculates the surgeon's region of attention based on instrument features, performs local segmentation, and integrates it with global segmentation to enhance performance. Additionally, we proposed a keyframe selection module that calculates the temporal correlations between consecutive frames based on instrument features. This module adaptively selects non-keyframe for feature fusion segmentation, reducing noise and optimizing speed. RESULTS: To evaluate the performance of IG-Net, we constructed the most extensive dataset known to date, comprising 106 video clips and 6153 images. The experimental results reveal that this method achieves favorable performance, with 72.70% IoU, 82.02% Dice, and 35 FPS. CONCLUSIONS: For the task of prostate segmentation based on surgical videos, our proposed IG-Net surpasses all previous methods across multiple metrics. IG-Net balances segmentation accuracy and speed, demonstrating strong robustness against adverse factors.

Occurrence and distribution of microcystins in Lake Taihu, China.

The occurrence and distribution of microcystins were investigated in Lake Taihu, the third largest lake in China. An extensive survey, larger and broader in scale than previous studies, was conducted in summer 2010. The highest microcystin concentration was found at southern part of Taihu, which was newly included in this survey. In northern coastal areas, total cellular concentrations of 20 to 44 µg/L were observed. In northern offshore waters, levels were up to 4.8 µg/L. Microcystin occurrence was highly correlated with chemical oxygen demand, turbidity, and chlorophyll-a. Extracellular/total cellular microcystin (E/T) ratios were calculated and compared to other water quality parameters. A higher correlation was found using E/T ratios than original microcystin values. These results show that algal blooms are having a severe impact on Lake Taihu, and further and extensive monitoring and research are required to suppress blooms effectively.

FineStyle: Semantic-Aware Fine-Grained Motion Style Transfer with Dual Interactive-Flow Fusion.

We present FineStyle, a novel framework for motion style transfer that generates expressive human animations with specific styles for virtual reality and vision fields. It incorporates semantic awareness, which improves motion representation and allows for precise and stylish animation generation. Existing methods for motion style transfer have all failed to consider the semantic meaning behind the motion, resulting in limited controls over the generated human animations. To improve, FineStyle introduces a new cross-modality fusion module called Dual Interactive-Flow Fusion (DIFF). As the first attempt, DIFF integrates motion style features and semantic flows, producing semantic-aware style codes for fine-grained motion style transfer. FineStyle uses an innovative two-stage semantic guidance approach that leverages semantic clues to enhance the discriminative power of both semantic and style features. At an early stage, a semantic-guided encoder introduces distinct semantic clues into the style flow. Then, at a fine stage, both flows are further fused interactively, selecting the matched and critical clues from both flows. Extensive experiments demonstrate that FineStyle outperforms state-of-the-art methods in visual quality and controllability. By considering the semantic meaning behind motion style patterns, FineStyle allows for more precise control over motion styles. Source code and model are available on https://github.com/XingliangJin/Fine-Style.git.

Graph Diffusion Convolutional Network for Skeleton Based Semantic Recognition of Two-Person Actions.

Graph Convolutional Networks (GCNs) have successfully boosted skeleton-based human action recognition. However, existing GCN-based methods mostly cast the problem as separated person's action recognition while ignoring the interaction between the action initiator and the action responder, especially for the fundamental two-person interactive action recognition. It is still challenging to effectively take into account the intrinsic local-global clues of the two-person activity. Additionally, message passing in GCN depends on adjacency matrix, but skeleton-based human action recognition methods tend to calculate the adjacency matrix with the fixed natural skeleton connectivity. It means that messages can only travel along a fixed path at different layers of the network or in different actions, which greatly reduces the flexibility of the network. To this end, we propose a novel graph diffusion convolutional network for skeleton based semantic recognition of two-person actions by embedding the graph diffusion into GCNs. At technical fronts, we dynamically construct the adjacency matrix based on practical action information, so that we can guide the message propagation in a more meaningful way. Simultaneously, we introduce the frame importance calculation module to conduct dynamic convolution, so that we can avoid the negative effect caused by the traditional convolution, wherein the shared weights may fail to capture key frames or be affected by noisy frames. Besides, we comprehensively leverage the multidimensional features related to joints' local visual appearances, global spatial relationship and temporal coherency, and for different features, different metrics are designed to measure the similarity underlying the corresponding real physical law of the motions. Moreover, extensive experiments and comprehensive evaluations on four public large-scale datasets (NTU-RGB+D 60, NTU-RGB+D 120, Kinetics-Skeleton 400, and SBU-Interaction) demonstrate that our method outperforms the state-of-the-art methods.

MPMNet: A Data-Driven MPM Framework for Dynamic Fluid-Solid Interaction.

High-accuracy, high-efficiency physics-based fluid-solid interaction is essential for reality modeling and computer animation in online games or real-time Virtual Reality (VR) systems. However, the large-scale simulation of incompressible fluid and its interaction with the surrounding solid environment is either time-consuming or suffering from the reduced time/space resolution due to the complicated iterative nature pertinent to numerical computations of involved Partial Differential Equations (PDEs). In recent years, we have witnessed significant growth in exploring a different, alternative data-driven approach to addressing some of the existing technical challenges in conventional model-centric graphics and animation methods. This paper showcases some of our exploratory efforts in this direction. One technical concern of our research is to address the central key challenge of how to best construct the numerical solver effectively and how to best integrate spatiotemporal/dimensional neural networks with the available MPM's pressure solvers. In particular, we devise the MPMNet, a hybrid data-driven framework supporting the popular and powerful Material Point Method (MPM), to combine the comprehensive properties of MPM in numerically handling physical behaviors ranging from fluid to deformable solids and the high efficiency of data-driven models. At the architectural level, our MPMNet comprises three primary components: A data processing module to describe the physical properties by way of the input fields; A deep neural network group to learn the spatiotemporal features; And an iterative refinement process to continue to reduce possible numerical errors. The goal of these special technical developments is to aim at involved numerical acceleration while preserving physical accuracy, realizing efficient and accurate fluid-solid interactions in a data-driven fashion. The extensive experimental results verify that our MPMNet can tremendously speed up the computation compared with the popular numerical methods as the complexity of interaction scenes increases while better retaining the numerical accuracy.