Diverse Targets Research Articles

Accurate protein-ligand binding free energy estimation using QM/MM on multi-conformers predicted from classical mining minima

Accurate prediction of binding free energy is crucial for the rational design of drug candidates and understanding protein-ligand interactions. To address this, we have developed four protocols that combine QM/MM calculations and the mining minima (M2) method, tested on 9 targets and 203 ligands. Our protocols carry out free energy processing with or without conformational search on the selected conformers obtained from M2 calculations, where their force field atomic charge parameters are substituted with those obtained from a QM/MM calculation. The method achieved a high Pearson’s correlation coefficient (0.81) with experimental binding free energies across diverse targets, demonstrating its generality. Using a differential evolution algorithm with a universal scaling factor of 0.2, we achieved a low mean absolute error of 0.60 kcal mol-1. This performance surpasses many existing methods and is comparable to popular relative binding free energy techniques but at significantly lower computational cost.

Reception of health data using the Australian Burden of Disease study as an example

Abstract Background Health data, especially on Burden of Disease (BoD), are vital for evidence-based health promotion and effective dissemination is crucial due to resource intensiveness. Germany’s national BoD study is currently transitioning to permanence. Analysing the Australian Institute of Health and Welfare’s (AIHW) well-established collaboration with stakeholders and their reception of health data at national and regional level offers insights for improved knowledge translation. Methods To follow the path of stakeholder engagement, this 3-month project involved scientific exchanges with the AIHW and Department of Health, Western Australia. It compiled communication strategy examples and documents stating examples of BoD data use. A quantitative online survey accompanied by a series of 18 qualitative 45-minute interviews was conducted. The survey implemented via Swift Digital has been running for 20+ days. Interview partners originated from the fields of health research, politics, planning and advocacy for patients or population groups. Results were analysed, illustrated digitally as a booklet and refined in a participatory workshop. Results Communication strategies such as embargoed press and stakeholder releases as well as regular online formats were identified. The survey and interviews displayed the required technical and content-detail level from key facts to raw data and visualisation needs. Noteworthy are the demand to link data to economic aspects like disease expenditure, trends and future scenarios. It elucidated how diverse target groups utilize data to place disease significance and preventive public health measures, aiding dissemination. Alongside, it shows examples of knowledge translation by emphasizing impact over indicators when targeting communities. Conclusions Prioritizing target group identification and employing tailored communication strategies regarding data format and detail level are essential when utilizing new data. Key messages • Analyzing a well-established collaboration with stakeholders offers insights for improved knowledge translation and reveals leveraging engagement strategies for others. • Understanding stakeholder needs is essential for promoting broader health data use in policy and practice, fostering evidence-based decision-making, and enhancing public health outcomes.

YOLO-DroneMS: Multi-Scale Object Detection Network for Unmanned Aerial Vehicle (UAV) Images

In recent years, research on Unmanned Aerial Vehicles (UAVs) has developed rapidly. Compared to traditional remote-sensing images, UAV images exhibit complex backgrounds, high resolution, and large differences in object scales. Therefore, UAV object detection is an essential yet challenging task. This paper proposes a multi-scale object detection network, namely YOLO-DroneMS (You Only Look Once for Drone Multi-Scale Object), for UAV images. Targeting the pivotal connection between the backbone and neck, the Large Separable Kernel Attention (LSKA) mechanism is adopted with the Spatial Pyramid Pooling Factor (SPPF), where weighted processing of multi-scale feature maps is performed to focus more on features. And Attentional Scale Sequence Fusion DySample (ASF-DySample) is introduced to perform attention scale sequence fusion and dynamic upsampling to conserve resources. Then, the faster cross-stage partial network bottleneck with two convolutions (named C2f) in the backbone is optimized using the Inverted Residual Mobile Block and Dilated Reparam Block (iRMB-DRB), which balances the advantages of dynamic global modeling and static local information fusion. This optimization effectively increases the model’s receptive field, enhancing its capability for downstream tasks. By replacing the original CIoU with WIoUv3, the model prioritizes anchoring boxes of superior quality, dynamically adjusting weights to enhance detection performance for small objects. Experimental findings on the VisDrone2019 dataset demonstrate that at an Intersection over Union (IoU) of 0.5, YOLO-DroneMS achieves a 3.6% increase in mAP@50 compared to the YOLOv8n model. Moreover, YOLO-DroneMS exhibits improved detection speed, increasing the number of frames per second (FPS) from 78.7 to 83.3. The enhanced model supports diverse target scales and achieves high recognition rates, making it well-suited for drone-based object detection tasks, particularly in scenarios involving multiple object clusters.

Ensemble Docking as a Tool for the Rational Design of Peptidomimetic Staphylococcus aureus Sortase A Inhibitors.

Sortase A (SrtA) of Staphylococcus aureus has long been shown to be a relevant molecular target for antibacterial development. Moreover, the designed SrtA inhibitors act via the antivirulence mechanism, potentially causing less evolutional pressure and reduced antimicrobial resistance. However, no marketed drugs or even drug candidates have been reported until recently, despite numerous efforts in the field. SrtA has been shown to be a tough target for rational structure-based drug design (SBDD), which hampers the regular development of small-molecule inhibitors using the available arsenal of drug discovery tools. Recently, several oligopeptides resembling the sorting sequence LPxTG (Leu-Pro-Any-Thr-Gly) of the native substrates of SrtA were reported to be active in the micromolar range. Despite the good experimental design of those works, their molecular modeling parts are still not convincing enough to be used as a basis for a rational modification of peptidic inhibitors. In this work, we propose to use the ensemble docking approach, in which the relevant SrtA conformations are extracted from the molecular dynamics simulation of the LPRDA (Leu-Pro-Arg-Asp-Ala)-SrtA complex, to effectively represent the most significant and diverse target conformations. The developed protocol is shown to describe the known experimental data well and then is applied to a series of new peptidomimetic molecules resembling the active oligopeptide structures reported previously in order to prioritize structures from this work for further synthesis and activity testing. The proposed approach is compared to existing alternatives, and further directions for its development are outlined.

Bis(2-butoxyethyl) Ether-Promoted O2-Mediated Oxidation of Alkyl Aromatics to Ketones under Clean Conditions.

Conventional oxidation processes for alkyl aromatics to ketones employ oxidants that tend to generate harmful byproducts and cause severe equipment corrosion, ultimately creating critical environmental problems. Thus, in this study, a practical, efficient, and green method was developed for the synthesis of aromatic ketones by applying a bis(2-butoxyethyl) ether/O2 system under external catalyst-, additive-, and base-free conditions. This O2-mediated oxidation system can tolerate various functional groups and is suitable for large-scale synthesis. Diverse target ketones were prepared under clean conditions in moderate-to-high yields. The late-stage functionalization of drug derivatives with the corresponding ketones and one-pot sequential chemical conversions to ketone downstream products further broaden the application prospects of this approach.

A comprehensive investigation of multimodal deep learning fusion strategies for breast cancer classification

In breast cancer research, diverse data types and formats, such as radiological images, clinical records, histological data, and expression analysis, are employed. Given the intricate nature of natural phenomena, relying on the features of a single modality is seldom sufficient for comprehensive analysis. Therefore, it is possible to guarantee medical relevance and achieve improved clinical outcomes by combining several modalities. The presen study carefully maps and reviews 47 primary articles from six well-known digital libraries that were published between 2018 and 2023 for breast cancer classification based on multimodal deep learning fusion (MDLF) techniques. This systematic literature review encompasses various aspects, including the medical modalities combined, the datasets utilized in these studies, the techniques, models, and architectures used in MDLF and it also discusses the advantages and limitations of each approach. The analysis of selected papers has revealed a compelling trend: the emergence of new modalities and combinations that were previously unexplored in the context of breast cancer classification. This exploration has not only expanded the scope of predictive models but also introduced fresh perspectives for addressing diverse targets, ranging from screening to diagnosis and prognosis. The practical advantages of MDLF are evident in its ability to enhance the predictive capabilities of machine learning models, resulting in improved accuracy across diverse applications. The prevalence of deep learning models underscores their success in autonomously discerning complex patterns, offering a substantial departure from traditional machine learning approaches. Furthermore, the paper explores the challenges and future directions in this field, including the need for larger datasets, the use of ensemble learning methods, and the interpretation of multimodal models.

Recognition of phylogenetically diverse pathogens through enzymatically amplified recruitment of RNF213.

Innate immunity senses microbial ligands known as pathogen-associated molecular patterns (PAMPs). Except for nucleic acids, PAMPs are exceedingly taxa-specific, thus enabling pattern recognition receptors to detect cognate pathogens while ignoring others. How the E3 ubiquitin ligase RNF213 can respond to phylogenetically distant pathogens, including Gram-negative Salmonella, Gram-positive Listeria, and eukaryotic Toxoplasma, remains unknown. Here we report that the evolutionary history of RNF213 is indicative of repeated adaptation to diverse pathogen target structures, especially in and around its newly identified CBM20 carbohydrate-binding domain, which we have resolved by cryo-EM. We find that RNF213 forms coats on phylogenetically distant pathogens. ATP hydrolysis by RNF213's dynein-like domain is essential for coat formation on all three pathogens studied as is RZ finger-mediated E3 ligase activity for bacteria. Coat formation is not diffusion-limited but instead relies on rate-limiting initiation events and subsequent cooperative incorporation of further RNF213 molecules. We conclude that RNF213 responds to evolutionarily distant pathogens through enzymatically amplified cooperative recruitment.

A network pharmacology approach to elucidate the anti-inflammatory and antioxidant effects of bitter leaf (Vernonia amygdalina Del.)

The therapeutic potential of bitter leaf (Vernonia amygdalina Del.) has been established both empirically and in various scientific investigations. However, the molecular pathways related to its possible anti-inflammatory and antioxidant properties remain unclear. Therefore, the aim of this study was to elucidate the molecular interactions between bitter leaf's bioactive compounds and cellular targets involved in these activities. The compounds in bitter leaf were identified using gas chromatography-mass spectrometry (GC-MS) analysis, and subsequently, a network pharmacology approach was employed together with molecular docking and dynamics simulations. Acetonitrile (4.5%) and dimethylamine (4.972%) were the most prevalent compounds among the 38 identified by the GC-MS analysis of bitter leaf extract. The proto-oncogene tyrosine-protein kinase (SRC) demonstrated significant connectivity within the antioxidant network, highlighting its pivotal role in facilitating inter-protein communication. It also exhibited strategic positioning in anti-inflammatory mechanisms based on closeness centrality (0.385). The enrichment analysis suggested multifaceted mechanisms of bitter leaf compounds, including transcriptional regulation and diverse cellular targeting, indicating broad antioxidant and anti-inflammatory effects. Eicosapentaenoyl ethanolamide (EPEA) displayed strong interactions with multiple proteins, including SRC (-7.17 kcal/mol) and CYP3A4 (-6.88 kcal/mol). Moreover, EPEA demonstrated to form a stable interaction with SRC during a 100 ns simulation. In conclusion, the computational simulations revealed that the hypothetical antioxidant and anti-inflammatory actions of bitter leaf compounds were achieved by specifically targeting SRC. However, confirmation using either in vitro or in vivo techniques is necessary.

A Double-Stranded Aptamer for Highly Sensitive Fluorescent Detection of Glutathione S-Transferases.

Aptamer-based biosensors have been widely constructed and applied to detect diverse targets. Glutathione S-transferase (GST), a pivotal phase II metabolic enzyme, plays a critical role in biotransformation in vivo, and aberrant GST expression is associated with various health risks. Herein, aptamers targeting GST were systematically selected from a randomized single-stranded DNA (ssDNA) library of 79 nucleotides (nt) using a biotinylated GST-immobilized streptavidin agarose (SA) bead SELEX technology. Following rigorous screening across eight rounds, four aptamers with strikingly similar secondary structures emerged. Among these, Seq3 exhibited the highest affinity towards GST and was selected for further optimization. A semi-rational post-SELEX truncation strategy was then employed based on base composition analysis, secondary structure analysis and affinity assessment. This strategy enabled the systematic removal of redundant nucleotides in Seq3 without compromising its affinity, ultimately yielding a truncated aptamer, Seq3-3, which retains its specificity with a compact 39nt length. Building upon Seq3-3, a double-stranded fluorescent aptamer probe was ingeniously designed for the in vitro detection of GST. The detection mechanism hinges on the competitive displacement of the complementary chain from the probe, mediated by the target protein, leading to the separation of the antisense oligonucleotide from the double-stranded complex. This process triggers the restoration of the fluorescence signal, enabling sensitive detection, and the probe exhibits excellent response within a linear range of GST activity ranging from 0 to 1500 U/L. The results show that not only an efficient strategy for screening robust and practicable aptamers but also an ultrahighly sensitive detection platform for GST was established.

Two-stage deep learning framework for occlusal crown depth image generation

The generation of depth images of occlusal dental crowns is complicated by the need for customization in each case. To decrease the workload of skilled dental technicians, various computer vision models have been used to generate realistic occlusal crown depth images with definite crown surface structures that can ultimately be reconstructed to three-dimensional crowns and directly used in patient treatment. However, it has remained difficult to generate images of the structure of dental crowns in a fluid position using computer vision models. In this paper, we propose a two-stage model for generating depth images of occlusal crowns in diverse positions. The model is divided into two parts: segmentation and inpainting to obtain both shape and surface structure accuracy. The segmentation network focuses on the position and size of the crowns, which allows the model to adapt to diverse targets. The inpainting network based on a GAN generates curved structures of the crown surfaces based on the target jaw image and a binary mask made by the segmentation network. The performance of the model is evaluated via quantitative metrics for the area detection and pixel-value metrics. Compared to the baseline model, the proposed method reduced the MSE score from 0.007001 to 0.002618 and increased DICE score from 0.9333 to 0.9648. It indicates that the model showed better performance in terms of the binary mask from the addition of the segmentation network and the internal structure through the use of inpainting networks. Also, the results demonstrated an improved ability of the proposed model to restore realistic details compared to other models.

Are there unique facets of therapeutic alliance for users of digital mental health interventions? An examination with the eHealth Therapeutic Alliance Inventory

Therapeutic alliance (TA) is a well-established predictor of clinical outcomes in traditional psychotherapy. However, its association with outcomes in eHealth interventions has been inconsistent, which might be due to the absence of measurements specifically designed to capture TA in eHealth settings. The eHealth Therapeutic Alliance Inventory (ETAI) incorporates conventional as well as unique eHealth TA subscales, enabling to examine the contribution of new facets of TA beyond traditional concepts. This study investigates the predictive contribution of eHealth TA subscales compared to conventional TA subscales on clinical outcomes and evaluates the concurrent criterion validity of the ETAI. The study was conducted within the framework of a randomized controlled trial involving a 10-week digital parent training program aimed at addressing child disruptive behaviors. Parents were randomly assigned to either an enhanced-quality or a standard-quality program. Parents from 68 families completed the ETAI at five weeks' post-program initiation and at the post-intervention phase. The primary outcome was the improvement in child behavior, measured by the Eyberg Child Behavior Inventory. Positive Pearson correlations were found between all ETAI subscales covering unique eHealth TA aspects, measured at the 5-week time-point, and improvement in child behavior at post-intervention (rs ≥ 0.23, ps < 0.03). The conventional TA subscale showed no significant Pearson correlation with improvement in child behavior. When examining the unique contributions of ETAI-subscales to explain the improvement in child behavior, only ETAI-Perceived Emotional Investment subscale was found to have a unique contribution (β = 0.29, p = 0.019). In addition, scores on most ETAI subscales were significantly higher among parents using the enhanced-quality program compared with the standard program (Cohen's ds > 0.48), reinforcing ETAI's criterion validity. The development of TA scales that incorporates unique eHealth TA subscales show initial promise in predicting outcomes. Further research is needed to better understand how different factors of eHealth TA relate to clinical outcomes across diverse clinical targets and programs.

PROTAC-mediated activation, rather than degradation, of a nuclear receptor reveals complex ligand-receptor interaction network.

Proteolysis-targeting chimeras (PROTACs) are heterobifunctional molecules containing a ligand for a protein of interest linked to an E3 ubiquitin ligase ligand that induce protein degradation through E3 recruitment to the target protein. Small changes in PROTAC linkers can have drastic consequences, including loss of degradation activity, but the structural mechanisms governing such changes are unclear. To study this phenomenon, we screened PROTACs of diverse targeting modalities and identified dTAG-13 as an activator of the xenobiotic-sensing pregnane X receptor (PXR), which promiscuously binds various ligands. Characterization of dTAG-13 analogs and precursors revealed interplay between the PXR-binding moiety, linker, and E3 ligand that altered PXR activity without inducing degradation. A crystal structure of PXR ligand binding domain bound to a precursor ligand showed ligand-induced binding pocket distortions and a linker-punctured tunnel to the protein exterior at a region incompatible with E3 complex formation, highlighting the effects of linker environment on PROTAC activity.

Leader Cells: Invade and Evade-The Frontline of Cancer Progression.

Metastasis is the leading cause of cancer-related mortality; however, a complete understanding of the molecular programs driving the metastatic cascade is lacking. Metastasis is dependent on collective invasion-a developmental process exploited by many epithelial cancers to establish secondary tumours and promote widespread disease. The key drivers of collective invasion are "Leader Cells", a functionally distinct subpopulation of cells that direct migration, cellular contractility, and lead trailing or follower cells. While a significant body of research has focused on leader cell biology in the traditional context of collective invasion, the influence of metastasis-promoting leader cells is an emerging area of study. This review provides insights into the expanded role of leader cells, detailing emerging evidence on the hybrid epithelial-mesenchymal transition (EMT) state and the phenotypical plasticity exhibited by leader cells. Additionally, we explore the role of leader cells in chemotherapeutic resistance and immune evasion, highlighting their potential as effective and diverse targets for novel cancer therapies.

Enhancing landfill efficiency to drive greenhouse gas reduction: A comprehensive study on best practices and policy recommendations.

This article investigates the pivotal role of non-hazardous waste landfills in achieving greenhouse gas (GHG) reduction objectives within the European Union (EU).1 This study leverages the experience of key stakeholders in the European landfilling, assesses the efficacy of 'best-in-class' landfill installations, evaluates their potential impact on GHG reduction, and offers concrete recommendations for operators and policymakers. 'Best-in-class' landfills exceed the commonly accepted best practices by implementing all the following practices: (1) an anticipated capture system during the operating phase, (2) prompt installation of the final cover and capture system, with use of an impermeable cover, (3) operated as bioreactor, keeping optimal humidity, (4) adequate maintenance and reporting, (5) recovery of captured gas and (6) treatment of residual methane emissions throughout the waste decomposition process. The main finding is that switching from the actual mix of practices to 'best in class' practices would reduce by ~21 MtCO2eq (-36%) the emissions due to the degradation of waste landfilled between 2024 and 2035, compared to the 'business-as-usual scenario', while also providing a renewable energy source, bringing potential avoided emissions and energy sovereignty. The findings underscore that in addition to implementing the organics diversion and waste reduction targets of the EU, adopting 'best-in class' landfill practices has the potential to bolster energy recovery, mitigate emissions and stimulate biomethane production, thereby advancing the EU environmental goals.

Effect of Rest Interval Length on Oral Swallowing Pressure During Effortful Swallowing Exercises in Healthy Adults: A Crossover Design.

Evidence supporting the prescription of effortful swallowing (ES) as a rehabilitation exercise remains lacking. This study aimed to evaluate the effect of rest interval length between sets on oral swallowing pressure during ES exercises in healthy adults. This study was a randomized trial of participants using a crossover design. Forty-three healthy adults (25.0 ± 3.7 years; 32 women and 11 men) without swallowing difficulties were recruited and participated in all four conditions (rest intervals between sets: 0-, 1-, 3-, and 5-min intervals) in a random order. The exercise constituted five sets, each comprising 10 ES repetitions. ES was performed with saliva swallows. The rest interval between each repetition was fixed at 10s, and the oral swallowing pressure during ES was measured using a tongue pressure measuring device. Oral swallowing pressures were significantly lower in the fifth set than in the first set at the 0-min rest condition. In the comparison by condition, oral swallowing pressures were significantly higher in the 5-min than in the 0-min rest conditions in the fourth set and in the 3- and 5-min rest conditions than in the 0-min rest condition in the fifth set. A rest interval of appropriate length between sets during ES exercises may consistently maintain the targeted high swallowing pressures during the exercises. Further studies using more diverse equipment and targeting older patients and those with dysphagia are required to determine the effect of the rest interval length between sets on the degree of exercise intensity to improve the swallowing-related muscle strength.

Multi-band compatible camouflage enabled by phase transition modulation of flexible GST films

With the rapid development of detection technology, there is an urgent need for multi-band compatible camouflage technology, particularly in the visible-infrared band. However, the conflicting requirements between camouflage across different bands and surface thermal management pose significant challenges in achieving synchronized control for multiple bands in diverse application environments and targets. Herein, we propose a spectrally selective emitter leveraging phase-change modulation of Ge2Sb2Te5, capable of achieving effective camouflage in both the visible band and two atmospheric windows over a wide temperature range. The designed emitters exhibit colorful appearances, compatible with various background environments, thereby ensuring the visible camouflage. Meanwhile, the emitters demonstrate low emissivity values of 0.1 in the MWIR band and 0.08 in the LWIR band, effectively reducing the infrared emission through atmospheric windows. The radiative temperatures of the optimized emitters in the MWIR and LWIR bands are reduced by 105°C and 140°C relative to the target surface (200°C), respectively, providing excellent radiative suppression compared to similar studies. Remarkably, the designed emitter still demonstrates robust flexibility, making it suitable for a variety of targets with complex curved surfaces. This work paves the way for designing the flexible visible-infrared compatible camouflage coatings with integrated thermal management functionality.

The Art of the Possible: The Constraints Associated with Writing Local TESOL Textbooks

Abstract Drawing upon interviews with eight textbook authors working for a Middle Eastern Ministry of Education, this article explores the multifaceted constraints associated with the production of local TESOL textbooks. Significant challenges are at play stemming from tight production deadlines of just 2 months per textbook, thereby precluding any piloting of materials. Furthermore, severe MoE strictures regarding the form textbooks can take are in operation with regard to unit titles, moral values to be transmitted, skills and language content, and space allocated to each segment of the book. Additional constraints include the challenge of writing for a diverse target audience, the proscription of manifold taboo topics, and the lack of textbook development training the authors receive. These constraints reportedly lead to compromises in textbook quality and limit the authors’ ability to produce pedagogically effective and culturally sensitive materials. The findings illuminate the complex context-specific dynamics mediating textbook production, highlighting the need for a more nuanced understanding of the local textbook production process and the constraints and challenges writers face.

Insights from Traditional Chinese Medicine for Restoring Skin Barrier Functions.

The skin barrier is essential for maintaining the body's internal homeostasis, protecting against harmful external substances, and regulating water and electrolyte balance. Traditional Chinese Medicine (TCM) offers notable advantages in restoring skin barrier function due to its diverse components, targets, and pathways. Recent studies have demonstrated that active ingredients in TCM can safely and effectively repair damaged skin barriers, reinstating their proper functions. This review article provides a comprehensive overview of the mechanisms underlying skin barrier damage and explores how the bioactive constituents of TCM contribute to skin barrier repair, thereby offering a theoretical framework to inform clinical practices.

GMS-YOLO: An Algorithm for Multi-Scale Object Detection in Complex Environments in Confined Compartments.

Many compartments are prone to pose safety hazards such as loose fasteners or object intrusion due to their confined space, making manual inspection challenging. To address the challenges of complex inspection environments, diverse target categories, and variable scales in confined compartments, this paper proposes a novel GMS-YOLO network, based on the improved YOLOv8 framework. In addition to the lightweight design, this network accurately detects targets by leveraging more precise high-level and low-level feature representations obtained from GhostHGNetv2, which enhances feature-extraction capabilities. To handle the issue of complex environments, the backbone employs GhostHGNetv2 to capture more accurate high-level and low-level feature representations, facilitating better distinction between background and targets. In addition, this network significantly reduces both network parameter size and computational complexity. To address the issue of varying target scales, the first layer of the feature fusion module introduces Multi-Scale Convolutional Attention (MSCA) to capture multi-scale contextual information and guide the feature fusion process. A new lightweight detection head, Shared Convolutional Detection Head (SCDH), is designed to enable the model to achieve higher accuracy while being lighter. To evaluate the performance of this algorithm, a dataset for object detection in this scenario was constructed. The experiment results indicate that compared to the original model, the parameter number of the improved model decreased by 37.8%, the GFLOPs decreased by 27.7%, and the average accuracy increased from 82.7% to 85.0%. This validates the accuracy and applicability of the proposed GMS-YOLO network.

Metasurface Enabled Multi‐Target and Multi‐Wavelength Diffraction Neural Networks

AbstractBenefiting from low power consumption and high processing speed, there is a growing interest in diffraction neural networks (DNNs), which are typically showcased with 3D printing devices, leading to large volumes, high costs, and low levels of integration. Metasurfaces can desirably manipulate wavefronts of electromagnetic waves, providing a compact platform for mimicking DNNs with novel functions. Although multi‐wavelength and multi‐target recognition provides a richer and more detailed understanding of complex environments, existing architectures are primarily trained to classify a single target at a specific wavelength. A metasurface approach is proposed to design multiplexed DNNs that can classify multiple targets and spatial sequences across various wavelengths in multiple channels. To realize multi‐task processing, the dielectric metasurface is designed based on phase and wavelength multiplexing, which can integrate multi‐target DNNs with different tasks such as operating at distinct wavelengths and classifying diverse targets. The efficacy of this method is exemplified through the numerical simulation and experimental demonstration of recognizing a single target with two wavelengths, two targets at a single wavelength, and two targets at dual wavelengths. This compact metasurface approach enables the design of multi‐target and multi‐wavelength DNNs, opening a new window to develop massively parallel processing and versatile artificial intelligence systems.