Maize is susceptible to various diseases throughout its growth cycle, which can significantly reduce yields. The accurate identification of maize diseases with similar symptomatic manifestations is particularly challenging under field conditions due to heterogeneous lighting and variable weather conditions. This paper proposes a novel detection model named SCFM-DETR, which is based on an improved Real-Time DEtection TRansformer (RT-DETR) to achieve robust identification of maize diseases in complex environments. SimAM-StarNet is employed as the backbone for feature extraction in this model, reducing the number of parameters and improving multiscale feature fusion, thereby diminishing the impact of background noise. Furthermore, the original RepC3 module is replaced with a newly designed CGLU-FasterBlock-MANet (CFM) module, which enhances adaptive feature fusion for finer discriminative capability. The experimental results demonstrate that the SCFM-DETR model achieves an average precision of 96.7% and a recall of 95.8% on a maize disease dataset, exceeding the corresponding metrics of the baseline RT-DETR-R18 model by 3.1% and 6.0%. Additionally, the model reduces the number of parameters and computational load by 47% and 49%, respectively, making it highly suitable for deployment in computationally limited agricultural settings. This work offers a high-accuracy, lightweight framework that facilitates intelligent crop disease monitoring and supports the advancement of smart agriculture.

Accurate localization of strawberries is essential for improving the performance of strawberry-harvesting robots. To address the needs of intelligent strawberry picking in terms of target detection and picking point positioning, this study proposes YOLOv8-SP (YOLOv8-Strawberry and Picking Points), an improved model capable of simultaneously identifying strawberries and detecting their picking points. The model introduces a novel lightweight module named C2f-Star-ELA (CSE), which replaces the Bottleneck in the original C2f module with StarNet blocks and incorporates an ELA attention mechanism. This enhancement improves accuracy while reducing computational cost. Furthermore, the original PAN structure is replaced with a Cross-Scale Feature Fusion Module (CCFM) to strengthen feature integration. Experimental results demonstrate that YOLOv8-SP achieves a strawberry detection accuracy of 94.1

Protein-small-molecule interactions are fundamental to cellular regulation and represent critical targets for therapeutic intervention. Accurate identification of binding residues is essential for elucidating molecular recognition mechanisms and guiding the rational drug design. Experimental approaches, however, are often costly, time-consuming, and limited in scalability, while existing computational methods that rely on handcrafted features or single Protein Language Model (PLM) embeddings fail to capture comprehensive residue-level representations and overlook the potential synergistic effects among diverse PLMs. Here, we present SOPE-MsL, a synergy-optimized approach that integrates PLM embedding fusion with multiscale learning for binding-site prediction. Through a systematic evaluation of representative state-of-the-art PLMs from the ProtTrans, ESM, and Ankh families, we identified ProstT5 and Ankh as the most effective embedding pair. The fused embeddings are then processed by a network that combines multiscale convolutional operations with attention mechanisms, enabling the concurrent modeling of intricate local patterns and long-range dependencies. To address the pronounced class imbalance between binding and nonbinding residues, a weighted focal loss is employed. Beyond predictive performance, t-SNE and SHAP analyses further confirm the advantages of synergistic embedding fusion over single-model representations, providing residue-level interpretability. Extensive experiments across multiple benchmark data sets demonstrate that SOPE-MsL achieves competitive performance and provides a robust and interpretable tool for structure-aware sequence analysis and the identification of protein-small-molecule interaction sites.

Accurate identification of protein-DNA binding residues plays a vital role in elucidating molecular recognition and facilitating drug discovery. Nonetheless, conventional experimental techniques tend to be expensive, require substantial time, and demand considerable manual effort. The reliance of existing methods on handcrafted features often limits their ability to produce a high-quality initial representation. Although Protein Language Model (PLM) embeddings have proven effective, most methods depend on a single embedding source, limiting their generalizability and prediction accuracy. Here, we present IPDLPre, a novel model that combines an integrating pretrained PLM, a CNN-attention network, and contrastive learning for protein-DNA binding site prediction. Specifically, IPDLPre leverages three PLMs combined with a customized CNN-attention architecture to decode evolutionary features and generate residue-level binding confidence scores. To mitigate class imbalance, we employ a hybrid loss function combining triplet center loss and focal loss, enhancing feature representation and predictive performance. The experimental findings reveal that IPDLPre outperforms existing sequence-based techniques and delivers results comparable to those achieved by structure-based methods. Furthermore, IPDLPre demonstrates effective generalization on RNA-binding datasets, highlighting its broad applicability in predicting protein-nucleic acid interactions. In conclusion, IPDLPre provides a valuable strategy that holds great potential for advancing protein engineering and aiding drug development.

Accurately identifying protein-RNA binding residues is crucial for deciphering molecular recognition mechanisms and advancing drug design. While Protein Language Models (PLMs) have shown promise in residue-level feature extraction, existing methods often overlook the complementary benefits of integrating multiple feature modalities, leaving room for improved predictive performance. In this study, we present MFEPre, a novel multi-feature fusion framework that synergistically combines sequence-based PLMs embeddings, graph-based structural representations, and conventional handcrafted features to enhance the prediction of protein-RNA binding residues. Specifically, MFEPre leverages ProtBert embeddings to capture evolutionary and contextual sequence patterns, employs Graph Attention Networks (GATs) to model residue-level topological interactions in protein structures, and integrates handcrafted features. These features are processed through a three-channel convolutional neural network and performs feature fusion in a fully connected layer to predict binding sites. The results showed that the area under ROC curve values of the MFEPre on the test datasets reached 0.827, indicating superior performance compared to other existing models. Ablation studies confirm that three categories of features are complementary, highlighting the importance of multi-feature fusion. Our work offers new perspectives on protein-RNA binding site prediction by unifying sequence, structure, and biochemical insights, offering a robust tool for biological research and drug design.

Greenhouses are a critical component of modern agriculture, facilitating crop growth and development, and accurate predictions of temperature and humidity are essential for mitigating crop diseases and optimizing the growth environment. However, short- and medium-term forecasts of temperature and humidity are challenging because of the complexity of greenhouse microclimates. This paper presents a hybrid model that integrates a frequency-enhanced channel attention mechanism optimized with a temporal convolutional network (TCN-FECAM) and an iTransformer. The model employs a cross-attention mechanism incorporating the advantages of the two models, and a 48-sequence sliding window strategy is used to ensure accurate multistep predictions of temperature and humidity over spans of 3 h to 24 h. The experimental results demonstrate that the TCN-FECAM-iTransformer model outperforms other models across diverse time scales, including GRU, LSTM, Informer, Autoformer, Crossformer, FAM-LSTM, and TPA-LSTM. Specifically, in temperature prediction, the model achieves R2 coefficients of 0.979, 0.973, 0.968, and 0.953 and RMSE values of 0.657, 0.806, 0.923, and 1.126, for 3 h, 6 h, 12 h, and 24 h intervals, respectively. In humidity prediction, the model obtains R2 coefficients of 0.976, 0.961, 0.947, and 0.939 and RMSE values of 1.805, 2.567, 3.132, and 3.451 for 3 h, 6 h, 12 h, and 24 h intervals, respectively. The model therefore exhibits reliable performance in predicting temperature and humidity in greenhouse environments, offering robust support for monitoring and early warnings in crop growth environments.

Strawberry harvesting represents a labor-intensive agricultural operation where existing end-effector pose control algorithms frequently exhibit insufficient precision in fruit grasping, often resulting in unintended damage to target fruits. Concurrently, deep learning-based pose control algorithms suffer from inherent training instability, slow convergence rates, and inefficient learning processes in complex environments characterized by high-density fruit clusters and occluded picking scenarios. To address these challenges, this paper proposes an enhanced reinforcement learning framework DRFW-TQC that integrates Dynamic L2 Regularization for adaptive model stabilization and a Group-Wise Feature Weighting Network for discriminative feature representation. The methodology further incorporates a picking posture traction mechanism to optimize end-effector orientation control. The experimental results demonstrate the superior performance of DRFW-TQC compared to the baseline. The proposed approach achieves a 16.0% higher picking success rate and a 20.3% reduction in angular error with four target strawberries. Most notably, the framework’s transfer strategy effectively addresses the efficiency challenge in complex environments, maintaining an 89.1% success rate in eight-strawberry while reducing the timeout count by 60.2% compared to non-adaptive methods. These results confirm that DRFW-TQC successfully resolves the tripartite challenge of operational precision, training stability, and environmental adaptability in robotic fruit harvesting systems.

Protein-DNA complex interactivity plays a crucial role in biological activities such as gene expression, modification, replication and transcription. Understanding the physiological significance of protein-DNA binding interfacial hot spots, as well as the development of computational biology, depends on the precise identification of these regions. In this paper, a hot spot prediction method called EC-PDH is proposed. First, we extracted features of these hot spots’ solid solvent-accessible surface area (ASA) and secondary structure, and then the mean, variance, energy and autocorrelation function values of the first three intrinsic modal components (IMFs) of these conventional features were extracted as new features via the empirical modal decomposition algorithm (EMD). A total of 218 dimensional features were obtained. For feature selection, we used the maximum correlation minimum redundancy sequence forward selection method (mRMR-SFS) to obtain an optimal 11-dimensional-feature subset. To address the issue of data imbalance, we used the SMOTE-Tomek algorithm to balance positive and negative samples and finally used cat gradient boosting (CatBoost) to construct our hot spot prediction model for protein-DNA binding interfaces. Our method performs well on the test set, with AUC, MCC and F1 score values of 0.847, 0.543 and 0.772, respectively. After a comparative evaluation, EC-PDH outperforms the existing state-of-the-art methods in identifying hot spots.

Strawberry maturity detection plays an essential role in modern strawberry yield estimation and robot-assisted picking and sorting. Due to the small size and complex growth environment of strawberries, there are still problems with existing recognition systems’ accuracy and maturity classifications. This article proposes a strawberry maturity recognition algorithm based on an improved YOLOv5s model named YOLOv5s-BiCE. This algorithm model is a replacement of the upsampling algorithm with a CARAFE module structure. It is an improvement on the previous model in terms of its content-aware processing; it also widens the field of vision and maintains a high level of efficiency, resulting in improved object detection capabilities. This article also introduces a double attention mechanism named Biformed for small-target detection, optimizing computing allocation, and enhancing content perception flexibility. Via multi-scale feature fusion, we utilized double attention mechanisms to reduce the number of redundant computations. Additionally, the Focal_EIOU optimization method was introduced to improve its accuracy and address issues related to uneven sample classification in the loss function. The YOLOv5s-BiCE algorithm was better at recognizing strawberry maturity compared to the original YOLOv5s model. It achieved a 2.8% increase in the mean average precision and a 7.4% increase in accuracy for the strawberry maturity dataset. The improved algorithm outperformed other networks, like YOLOv4-tiny, YOLOv4-lite-e, YOLOv4-lite-s, YOLOv7, and Fast RCNN, with recognition accuracy improvements of 3.3%, 4.7%, 4.2%, 1.5%, and 2.2%, respectively. In addition, we developed a corresponding detection app and combined the algorithm with DeepSort to apply it to patrol robots. It was found that the detection algorithm exhibits a fast real-time detection speed, can support intelligent estimations of strawberry yield, and can assist picking robots.

Abstract To control water quality and seawater desalination dosage, modeling the coagulation process of saltwater is crucial. With a focus on the features of seawater coagulation with a long lag, a machine-learning sequence-based modeling approach is suggested. The link between influent and effluent turbidities, flow rates, flocculant and coagulant dosages, and other parameters is modeled using structured units such as a gate recurrent unit encoder and a linear network decoder. The model’s validity is confirmed by numerical experiments based on real operating data, which also offer a solid foundation for managing flocculant and coagulant assistance reduction.

Background Rice disease can significantly reduce yields, so monitoring and identifying the diseases during the growing season is crucial. Some current studies are based on images with simple backgrounds, while realistic scene settings are full of background noise, making this task challenging. Traditional artificial prevention and control methods not only have heavy workload, low efficiency, but are also haphazard, unable to achieve real-time monitoring, which seriously limits the development of modern agriculture. Therefore, using target detection algorithm to identify rice diseases is an important research direction in the agricultural field. Methods In this article a total of 7,220 pictures of rice diseases taken in Jinzhai County, Lu’an City, Anhui Province were chosen as the research object, including rice leaf blast, bacterial blight and flax leaf spot. We propose a rice disease identification method based on the improved YOLOV5s, which reduces the computation of the backbone network, reduces the weight file of the model to 3.2MB, which is about 1/4 of the original model, and accelerates the prediction speed by three times. Results Compared with other mainstream methods, our method achieves better performance with low computational cost. It solves the problem of slow recognition speed due to the large weight file and calculation amount of model when the model is deployed in mobile terminal.

Inframe insertion/deletion (indel) variants may alter protein sequence and function, which are closely related to an extensive variety of diseases. Although recent researches have paid attention to the associations between inframe indels and diseases, modeling indels in silico and interpreting their pathogenicity remain challenging, mainly due to the lack of experimental information and computational methodologies. In this article, we propose a novel computational method named PredinID (Predictor for inframe InDels) via graph convolutional network (GCN). PredinID leverages k-nearest neighbor algorithm to construct the feature graph for aggregating more informative representation, regarding the pathogenic inframe indel prediction as a node classification task. An edge-based sampling strategy is designed for extracting information from both the potential connections of feature space and the topological structure of subgraphs. Evaluated by 5-fold cross-validations, the PredinID method achieves satisfactory performance and is superior to four classic machine learning algorithms and two GCN methods. Comprehensive experiments show that PredinID has superior performances when compared with the state-of-the-art methods on the independent test set. Moreover, we also implement a web server at http://predinid.bio.aielab.cc/ , to facilitate the use of the model.

Protein-RNA interactions play a key role in various biological cellular processes, and many experimental and computational studies have been initiated to analyze their interactions. However, experimental determination is quite complex and expensive. Therefore, researchers have worked to develop efficient computational tools to detect protein-RNA binding residues. The accuracy of existing methods is limited by the features of the target and the performance of the computational models; there remains room for improvement. To solve the problem of the accurate detection of protein-RNA binding residues, we propose a convolutional network model named PBRPre based on improved MobileNet. First, by extracting the position information of the target complex and the 3-mer amino acid feature data, the position-specific scoring matrix (PSSM) is improved by using spatial neighbor smoothing processing and discrete wavelet transform to fully exploit the spatial structure information of the target and enrich the feature dataset. Second, the deep learning model MobileNet is used to integrate and optimize the potential features in the target complexes; then, by introducing the Vision Transformer (ViT) network classification layer, the deep-level information of the target is mined to enhance the processing ability of the model for global information and to improve the detection accuracy of the classifiers. The results show that the AUC value of the model can reach 0.866 in the independent testing dataset, which shows that PBRPre can effectively realize the detection of protein-RNA binding residues. All datasets and resource codes of PBRPre are available at https://github.com/linglewu/PBRPre for academic use.

Abstract Background Identification of hot spots in protein–DNA binding interfaces is extremely important for understanding the underlying mechanisms of protein–DNA interactions and drug design. Since experimental methods for identifying hot spots are time-consuming and expensive, and most of the existing computational methods are based on traditional protein–DNA features to predict hot spots, unable to make full use of the effective information in the features. Results In this work, a method named WTL-PDH is proposed for hot spots prediction. To deal with the unbalanced dataset, we used the Synthetic Minority Over-sampling Technique to generate minority class samples to achieve the balance of dataset. First, we extracted the solvent accessible surface area features and structural features, and then processed the traditional features using discrete wavelet transform and wavelet packet transform to extract the wavelet energy information and wavelet entropy information, and obtained a total of 175 dimensional features. In order to obtain the best feature subset, we systematically evaluate these features in various feature selection strategies. Finally, light gradient boosting machine (LightGBM) was used to establish the model. Conclusions Our method achieved good results on independent test set with AUC, MCC and F1 scores of 0.838, 0.533 and 0.750, respectively. WTL-PDH can achieve generally better performance in predicting hot spots when compared with state-of-the-art methods. The dataset and source code are available at https://github.com/chase2555/WTL-PDH .

Regenerative medicine and anti-aging research have made great strides at the molecular and cellular levels in dermatology and the medical aesthetic field, targeting potential treatments with skin therapeutic and intervention pathways, which make it possible to develop effective skin regeneration and repair ingredients. With the rapid development of computational biology, bioinformatics as well as artificial intelligence (A.I.), the development of new ingredients for regenerative medicine has been greatly accelerated, and the success rate has been improved. Some application cases have appeared in topical skin regeneration and repair scenarios. This review will briefly introduce the application of bioactive peptides in skin repair and anti-aging as emerging ingredients in cosmeceutics and emphasize how A.I. based computational biology technology may accelerate the development of innovative peptide molecules and ultimately translate them into potential skin regenerative and anti-aging scenarios. Typically, two research routines have been summarized and current limitations as well as directions were discussed for border applications in future research.

The individual identification of communication emitters is a process of identifying different emitters based on the radio frequency fingerprint features extracted from the received signals. Due to the inherent non-linearity of the emitter power amplifier, the fingerprints provide distinguishing features for emitter identification. In this study, approximate entropy is introduced into variational mode decomposition, whose features performed in each mode which is decomposed from the reconstructed signal are extracted while the local minimum removal method is used to filter out the noise mode to improve SNR. We proposed a semi-supervised dimensionality reduction method named exponential semi-supervised discriminant analysis in order to reduce the high-dimensional feature vectors of the signals, and LightGBM is applied to build a classifier for communication emitter identification. The experimental results show that the method performs better than the state-of-the-art individual communication emitter identification technology for the steady signal data set of radio stations with the same plant, batch and model.

Plants produce a wide range of bioactive peptides as part of their innate defense mechanisms. With the explosive growth of plant-derived peptides, verifying the therapeutic function using traditional experimental methods are resources and time consuming. Therefore, it is necessary to predict the therapeutic function of plant-derived peptides more effectively and accurately with reduced waste of resources and thus expedite the development of plant peptides. We herein developed a repository of plant peptides predicted to have multiple therapeutic functions, named as MFPPDB (multi-functional plant peptide database). MFPPDB including 1,482,409 single or multiple functional plant origin therapeutic peptides derived from 121 fundamental plant species. The functional categories of these therapeutic peptides include 41 different features such as anti-bacterial, anti-fungal, anti-HIV, anti-viral, and anti-cancer. The detailed physicochemical information of these peptides was presented in functional search and physicochemical property search module, which can help users easily access the peptide information by the plant peptide species, ID, and functions, or by their peptide ID, isoelectric point, peptide sequence, and molecular weight through web-friendly interface. We further matched the predicted peptides to nine state-of-the-art curated functional peptide databases and found that at least 293,408 of the peptides possess functional potentials. Overall, MFPPDB integrated a massive number of plant peptides have single or multiple therapeutic functions, which will facilitate the comprehensive research in plant peptidomics. MFPPDB can be freely accessed through http://124.223.195.214:9188/mfppdb/index.

Specific emitter identification involves extracting the fingerprint features that represent the individual differences of the emitter. This is achieved by processing the radio-frequency signals. Feature extraction and classifier selection are key factors that affect SEI performance. This paper proposes a deep convolutional neural network model based on multisignal feature fusion to identify the emitters. As part of the implementation model, the methods of singular spectrum analysis (SSA), variational mode decomposition (VMD), and intrinsic time-scale decomposition (ITD) are used to extract various signal features of emitter signals. Finally, a multichannel deep learning model is adopted to fuse each signal feature automatically and identify different signal emitters. Experimental results show that the proposed method completely considers the complementarity and independence of varying signal features and excavates hidden deep feature information. Hence, the process is considered reliable and effective.

Maize leaf diseases significantly reduce maize yield; therefore, monitoring and identifying the diseases during the growing season are crucial. Some of the current studies are based on images with simple backgrounds, and the realistic field settings are full of background noise, making this task challenging. We collected low-cost red, green, and blue (RGB) images from our experimental fields and public dataset, and they contain a total of four categories, namely, southern corn leaf blight (SCLB), gray leaf spot (GLS), southern corn rust (SR), and healthy (H). This article proposes a model different from convolutional neural networks (CNNs) based on transformer and self-attention. It represents visual information of local regions of images by tokens, calculates the correlation (called attention) of information between local regions with an attention mechanism, and finally integrates global information to make the classification. The results show that our model achieves the best performance compared to five mainstream CNNs at a meager computational cost, and the attention mechanism plays an extremely important role. The disease lesions information was effectively emphasized, and the background noise was suppressed. The proposed model is more suitable for fine-grained maize leaf disease identification in a complex background, and we demonstrated this idea from three perspectives, namely, theoretical, experimental, and visualization.