Identifying drug-target interactions (DTI) is a fundamental yet costly step in drug discovery, motivating the development of accurate and efficient computational prediction methods. In this paper, we propose FKAN-a, a DTI prediction framework that integrates Fourier-enhanced Kolmogorov-Arnold networks (KAN) with attention mechanisms under a contrastive learning paradigm. Drug molecules and protein sequences are preprocessed and encoded using pretrained representations to capture interaction-relevant features. The resulting embeddings are further transformed through KAN with learnable Fourier bases to model complex nonlinear relationships. A cross-modality attention module is introduced to enhance the modeling of fine-grained drug-protein associations. Experiments conducted on three public benchmark datasets demonstrate that FKAN-a consistently outperforms representative state-of-the-art methods in terms of prediction performance and computational efficiency. These results indicate that the proposed framework provides an effective solution for DTI prediction and practical candidate prioritization in drug discovery.

Arrhythmia is a prevalent cardiovascular disease, which has garnered widespread attention due to its age‐related increases in mortality rates. In the analysis of arrhythmia, the electrocardiogram (ECG) plays an important role. Arrhythmia classification often suffers from a significant data imbalance issue due to the limited availability of data for certain arrhythmia categories. This imbalance problem significantly affects the classification performance of the model. To address this challenge, data augmentation emerges as a viable solution, aiming to neutralize the adverse effects of imbalanced datasets on the model. To this end, this paper proposes a novel Multimodality Data Augmentation Network (MM‐DANet) for arrhythmia classification. The MM‐DANet consists of two modules: the multimodality data matching‐based data augmentation module and the multimodality feature encoding module. In the multimodality data matching‐based data augmentation module, we expand the underrepresented arrhythmia categories to match the size of the largest category. Subsequently, the multimodality feature encoding module employs convolutional neural networks (CNN) to extract the modality‐specific features from both signals and images and concatenate them for efficient and accurate classification. The MM‐DANet was evaluated on the MIT‐BIH Arrhythmia Database and achieving an accuracy of 98.83%, along with an average specificity of 98.87%, average sensitivity of 92.92%, average precision of 91.05%, and average F 1_score of 91.96%. Furthermore, its performance was also assessed on the St. Petersburg INCART arrhythmia database and the MIT‐BIH supraventricular arrhythmia database, yielding AUC values of 81.98% and 90.93%, respectively. These outstanding results not only underscore the effectiveness of MM‐DANet but also indicate its potential for facilitating reliable automated analysis of arrhythmias.

Genome-wide association studies have shown that common genetic variants associated with complex diseases are mostly located in non-coding regions, which may not be causal. In addition, the limited number of validated non-coding functional variants makes it difficult to develop an effective supervised learning model. Therefore, improving the accuracy of predicting non-coding causal variants has become critical. This study aims to build a transfer learning-based machine learning method for predicting regulatory variants to overcome the problem of limited sample size. This paper presents a supervised learning method transfer support vector machine (TSVM) for massively parallel reporter assays (MPRA) validated regulatory variants prediction. First, uses a convolutional neural network to extract features with transfer learning. Second, the extracted features are selected by random forest method. Third, the selected features are used to train support vector machine for classification. We performed scale sensitivity experiments on the MPRA dataset and validated the effectiveness of transfer learning. The model achieves the Mcc of 0.326 and the AUC of 0.720, which are higher than the state-of-the-art method.

The function labeling of enzymes has a wide range of application value in the medical field, industrial biology and other fields. Scientists define enzyme categories by enzyme commission (EC) numbers. At present, although there are some tools for enzyme function prediction, their effects have not reached the application level. To improve the precision of enzyme function prediction, we propose a parallel convolutional contrastive learning (PCCL) method to predict enzyme functions. First, we use the advanced protein language model ESM-2 to preprocess the protein sequences. Second, PCCL combines convolutional neural networks (CNNs) and contrastive learning to improve the prediction precision of multifunctional enzymes. Contrastive learning can make the model better deal with the problem of class imbalance. Finally, the deep learning framework is mainly composed of three parallel CNNs for fully extracting sample features. we compare PCCL with state-of-art enzyme function prediction methods based on three evaluation metrics. The performance of our model improves on both two test sets. Especially on the smaller test set, PCCL improves the AUC by 2.57%.

RNA-binding proteins are important for the process of cell life activities. High-throughput technique experimental method to discover RNA-protein binding sites is time-consuming and expensive. Deep learning is an effective theory for predicting RNA-protein binding sites. Using weighted voting method to integrate multiple basic classifier models can improve model performance. Thus, in our study, we propose a weighted voting deep learning model (WVDL), which uses weighted voting method to combine convolutional neural network (CNN), long short term memory network (LSTM) and residual network (ResNet). First, the final forecast result of WVDL outperforms the basic classifier models and other ensemble strategies. Second, WVDL can extract more effective features by using weighted voting to find the best weighted combination. And, the CNN model also can draw the predicted motif pictures. Third, WVDL gets a competitive experiment result on public RBP-24 datasets comparing with other state-of-the-art methods. The source code of our proposed WVDL can be found in https://github.com/biomg/WVDL.

Computational prediction of the RBP bound sites using features learned from existing annotation knowledge is an effective method because high-throughput experiments are complex, expensive and time-consuming. Many methods have been proposed to predict RNA-protein binding sites. However, the partial information of RNA sequence is not fully used. In this study, we propose multiple convolutional neural networks (MCNN) method, which predicts RNA-protein binding sites by integrating multiple convolutional neural networks constructed by RNA sequence information extracted from windows with different lengths. First, MCNN trains multiple CNNs base on RNA sequences extracted by different window lengths. Second, MCNN can extract more binding patterns of RBPs by combining these trained multiple CNNs previously. Third, MCNN only uses RNA base sequence information for RNA-protein binding sites prediction, which extracts sequence binding features and predicts the result with same architecture. This avoids the information loss of feature extraction step. Our proposed MCNN demonstrates a competitive performance comparing with other methods on a large-scale dataset derived from CLIP-seq, which is an effective method for RNA-protein binding sites prediction. The source code of our proposed MCNN method can be found in https://github.com/biomg/MCNN.

Several online Chinese handwriting databases have been proposed recently. Though they have been introduced in detail, to date, no one has ever evaluated these databases with experimental comparison. To help the researchers use the corresponding database properly for algorithm evaluation and real application, we compare the property of the handwriting characters in these databases, and evaluate them with the same experimental setup and handwriting recognizer. Moreover, we analyze the connection between the property and the corresponding recognition accuracy for the handwriting characters in different databases. These empirical evaluation results can help the researchers choose the right database for different algorithms and applications.

Automatic analysis of sentiments expressed in large scale online reviews is very important for intelligent business applications. Sentiment classification is the most popular task of sentiment analysis, which is more challenging than traditional topic-based text classification. Basic features, such as vocabulary words, are not enough to classify sentiments well. Deep Belief Network (DBN) is introduced to discover more abstract features of sentiments. To capture full information of the features, large-size network can be constructed, but at the same time, large-size network tends to over fit the training data and even noise, which will reduce the generalization ability of the network. In this paper, L2-norm Deep Belief Network (L2DBN) is proposed, which uses L2-norm regularization to optimize the network parameters of DBN. L2DBN is first initialized by an unsupervised layer-wise training algorithm, and then fine-tuned by a supervised procedure. Network parameters are optimized using both classification loss and network complexity. Experimental results show that the proposed L2DBN outperforms the state-of-the-art method and the basic DBN on golden, noisy and heterogeneous datasets.

Recently, we have proposed a handwriting Chinese character database HIT-OR3C. Though it has been introduced in detail, to date, it has not been evaluated by any handwriting recognition method. To help the researchers use this database for algorithm evaluation, we propose the structure of HIT-OR3C database. Moreover, we evaluate the OR3C database with a series of experiments using state-of-the-art handwriting recognizer. These experiment results on the different subsets can be a benchmark for the researchers who will use the database. The low average recognition rate confirms that the HIT-OR3C database is challenging.

This paper proposes an opening recognition corpus, HIT-OR3C, and its construction toolkit to facilitate the unconstrained online Chinese handwriting text recognition. The characters of HIT-OR3C are collected through handwriting pad and are recorded and labeled automatically via the proposed handwriting document collection software OR3C Toolkit. HIT-OR3C consists of 5 subsets, namely GB1, GB2, Letter, Digit and Document. The first 4 corpora contain 6,825 categories produced by 122 persons and 832,650 samples in total. The document corpus is corresponding to 10 news articles that contain 2,442 categories produced by 20 persons and 77,168 samples in total. HIT-OR3C can be used for training and evaluation of character recognition algorithms. The OR3C Toolkit provides an efficient, device-independent, and unconstrained platform for the building of large scale handwriting corpus.

This paper presents a novel semi-supervised learning algorithm called Discriminative Deep Belief Networks (DDBN), to address the image classification problem with limited labeled data. We first construct a new deep architecture for classification using a set of Restricted Boltzmann Machines (RBM). The parameter space of the deep architecture is initially determined using labeled data together with abundant of unlabeled data, by greedy layer-wise unsupervised learning. Then, we fine-tune the whole deep networks using an exponential loss function to maximize the separability of the labeled data, by gradient-descent based supervised learning. Experiments on the artificial dataset and real image datasets show that DDBN outperforms most semi-supervised algorithm and deep learning techniques, especially for the hard classification tasks.

This paper introduces a new type of deep learning method named Deep Quantum Network (DQN) for classification. DQN inherits the capability of modeling the structure of a feature space by fuzzy sets. At first, we propose the architecture of DQN, which consists of quantum neuron and sigmoid neuron and can guide the embedding of samples divisible in new Euclidean space. The parameter of DQN is initialized through greedy layer-wise unsupervised learning. Then, the parameter space of the deep architecture and quantum representation are refined by supervised learning based on the global gradient-descent procedure. An exponential loss function is introduced in this paper to guide the supervised learning procedure. Experiments conducted on standard datasets show that DQN outperforms other feed forward neural networks and neuro-fuzzy classifiers.