Clothing change person re-identification (CC-ReID) aims to match different images of the same person, even when the clothing varies across images. To reduce manual labeling costs, existing unsupervised CC-ReID methods employ clustering algorithms to generate pseudo-labels. However, they often fail to assign the same pseudo-label to two images with the same identity but different clothing—referred to as a clothing change positive pair—thus hindering clothing-invariant feature learning. To address this issue, we propose the identity-clothing similarity modeling (ICSM) framework. To effectively connect clothing change positive pairs, ICSM first performs clothing-aware learning to leverage all discriminative information, including clothing, to obtain compact clusters. It then extracts cluster-level identity and clothing features and performs inter-cluster similarity estimation to identify clothing change positive clusters, reliable negative clusters, and hard negative clusters for each compact cluster. During optimization, we design an adaptive version of existing optimization methods to enhance similarities of clothing change positive pairs, while also introducing text semantics as a supervisory signal to further promote clothing invariance. Extensive experimental results across multiple datasets validate the effectiveness of the proposed framework, demonstrating its superiority over existing unsupervised methods and its competitiveness with some supervised approaches.

Visible-infrared person re-identification (VI-ReID) aims to match visible and infrared images of the same individual. Supervised VI-ReID (SVI-ReID) methods have achieved promising performance under the guidance of manually annotated identity labels. However, the substantial annotation cost severely limits their scalability in real-world applications. As a result, unsupervised VI-ReID (UVI-ReID) methods have attracted increasing attention. These methods typically rely on pseudo-labels generated by clustering and matching algorithms to replace manual annotations. Nevertheless, the quality of pseudo-labels is often difficult to guarantee, and low-quality pseudo-labels can significantly hinder model performance improvements. To address these challenges, we explore the use of attribute arrays extracted by a large vision-language model (LVLM) to enhance VI-ReID, and propose a novel LVLM-driven attribute-aware modeling (LVLM-AAM) approach. Specifically, we first design an attribute-aware reliable labeling strategy, which refines intra-modality clustering results based on image-level attributes and improves inter-modality matching by grouping clusters according to cluster-level attributes. Next, we develop an explicit-implicit attribute fusion module, which integrates explicit and implicit attributes to obtain more fine-grained identity-related text features. Finally, we introduce an attribute-aware contrastive learning module, which jointly leverages static and dynamic text features to promote modality-invariant feature learning. Extensive experiments conducted on VI-ReID datasets validate the effectiveness of the proposed LVLM-AAM and its individual components. LVLM-AAM not only significantly outperforms existing unsupervised methods but also surpasses several supervised methods.

Drug combination is a promising therapeutic strategy for complex diseases. However, only a small fraction of potential drug combinations exhibit true synergistic effects, making the prediction of drug synergy a critical yet challenging task. In this study, we propose BridgeSyn, a novel bridge fusion framework for drug synergy prediction. BridgeSyn leverages the knowledge from pretrained biological language models to enrich both drug compound and cell line representations. We introduce a bridging fusion mechanism that employs a set of shared latent tokens derived from global features, serving as a semantic interface to effectively fuse the representations of drug pairs and cell lines. By combining biological prior knowledge with this fusion strategy, BridgeSyn can capture complex biological interactions and achieve superior prediction results. Extensive experiments on two public datasets demonstrate that BridgeSyn consistently outperforms existing computation methods.

Protein-protein interactions (PPIs) are crucial for various cellular activities and disease development, and modulating PPIs using small molecule inhibitors (PPIIs) has gradually become a promising therapeutic strategy. Recently, researchers have proposed several machine learning methods to screen PPIIs, but most of the works focused on unimodal representations of molecules or combining multimodal features in a naive splicing manner. Meanwhile, current research progress is being slowed by the lack of large-scale PPII datasets. To address these issues, we propose MCLPPII, a unified multimodal contrastive learning framework for PPII prediction. MCLPPII extracts comprehensive molecular information from four modalities and effectively combines them through an adaptive feature fusion method. Furthermore, we propose a three-stage training strategy to enhance the PPII prediction capability of MCLPPII by leveraging self-supervised pre-training on a large unlabeled dataset. We evaluate MCLPPII on nine PPI targets and two downstream tasks, including the PPI inhibitor identification task and potency prediction task. Experimental results show that MCLPPII achieves competitive performance. The source code and datasets are freely available at https://github.com/1zzt/MCLPPII.

Abstract Accurately predicting the Remaining Useful Life (RUL) of lithium-ion batteries is crucial for battery management systems. Deep learning-based methods have been shown to be effective in predicting RUL by leveraging battery capacity time series data. However, the representation learning of features such as long-distance sequence dependencies and mutations in capacity time series still needs to be improved. To address this challenge, this paper proposes a novel deep learning model, the MLP-Mixer and Mixture of Expert (MMMe) model, for RUL prediction. The MMMe model leverages the Gated Recurrent Unit and Multi-Head Attention mechanism to encode the sequential data of battery capacity to capture the temporal features and a re-zero MLP-Mixer model to capture the high-level features. Additionally, we devise an ensemble predictor based on a Mixture-of-Experts (MoE) architecture to generate reliable RUL predictions. The experimental results on public datasets demonstrate that our proposed model significantly outperforms other existing methods, providing more reliable and precise RUL predictions while also accurately tracking the capacity degradation process. Our code and dataset are available at the website of github.

Accurate prediction of Drug-Target binding Affinity (DTA) is a daunting yet pivotal task in the sphere of drug discovery. Over the years, a plethora of deep learning-based DTA models have emerged, rendering promising results in predicting the binding affinities between drugs and their target proteins. However, in contrast to the conventional approach of modeling binding affinity in vector spaces, we propose a more nuanced modeling process in a continuous space to account for the diversity of input samples. Initially, the drug is encoded using the Simplified Molecular Input Line Entry System (SMILES), while the target sequences are characterized via a pretrained language model. Subsequently, highly correlative information is extracted utilizing residual gated convolutional neural networks. In a departure from existing deep learning-based models, our model learns the hidden representations of the drugs and targets jointly. Instead of employing two vectors, our hidden representations consist of two Gaussian distributions. To validate the effectiveness of our proposal, we conducted evaluations on commonly utilized benchmark datasets. The experimental outcomes corroborated that our method surpasses the state-of-the-art vectorial representation methods in terms of performance. This approach, therefore, offers potential enhancements in the precision of DTA predictions, potentially contributing to more efficient drug discovery processes.

Accurate prediction of small molecule modulators targeting protein-protein interactions (PPIMs) remains a significant challenge in drug discovery. Existing machine learning-based models rely on manual feature engineering, which is tedious and task-specific. Recently, deep learning models based on graph neural networks have made remarkable progress in molecular representation learning. However, many graph-based approaches ignore molecular hierarchical structure modeling guided by domain knowledge. In chemistry, the functional groups of a molecule determine its interaction with specific targets. Therefore, we propose a hierarchical graph neural network framework (called HiGPPIM) for predicting PPIMs by integrating atom-level and functional group-level features of molecules. HiGPPIM constructs atom-level and functional group-level graphs based on chemical knowledge and learns graph representations using graph attention networks. Furthermore, a hypergraph attention network is designed in HiGPPIM to aggregate and transform two-level graph information. We evaluate the performance of HiGPPIM on eight PPI families and two prediction tasks, namely PPIM identification and potency prediction. Experimental results demonstrate that HiGPPIM achieves state-of-the-art performance on both tasks and that using functional group information to guide PPIM prediction is effective.

Identifying synergistic drug combinations is paramount significance in addressing complex diseases while reducing the risks of toxicities and other adverse effects. Although a plethora of computational methods have been proposed in this domain, most of them are underpinned only by physicochemical or biological features. Recently, Chemical Language Models (CLMs) are shown to be capable of learning better representations that hold utility across diverse tasks, from molecular property prediction, de novo drug design, drug-target interaction, and more. In this study, we proposed CLMSyn, a continuous prompt for CLM aided synergistic drug combinations prediction. Unlike existing works employ CLMs for downstream tasks, we adopt the prompt learning to fine-tune CLM, that is, only train small-scale prompt while keeping CLM fixed. Furthermore, we harness the the multi-head attention mechanism to fuse the learned vector from the CLM, chemical descriptors and gene expression of cell line. A comprehensive array of experiments conducted on a benchmark dataset, encompassing four distinct synergy types, substantiates the superior performance of CLMSyn when contrasted against existing state-of-the-art methods. These empirical findings provide compelling evidence attesting to the efficacy of CLMSyn as a potent instrumentality in expediting the identification of pioneering combination therapies.

MOTIVATION: Accurate prediction of drug-target binding affinity (DTA) is crucial for drug discovery. The increase in the publication of large-scale DTA datasets enables the development of various computational methods for DTA prediction. Numerous deep learning-based methods have been proposed to predict affinities, some of which only utilize original sequence information or complex structures, but the effective combination of various information and protein-binding pockets have not been fully mined. Therefore, a new method that integrates available key information is urgently needed to predict DTA and accelerate the drug discovery process. RESULTS: In this study, we propose a novel deep learning-based predictor termed DataDTA to estimate the affinities of drug-target pairs. DataDTA utilizes descriptors of predicted pockets and sequences of proteins, as well as low-dimensional molecular features and SMILES strings of compounds as inputs. Specifically, the pockets were predicted from the three-dimensional structure of proteins and their descriptors were extracted as the partial input features for DTA prediction. The molecular representation of compounds based on algebraic graph features was collected to supplement the input information of targets. Furthermore, to ensure effective learning of multiscale interaction features, a dual-interaction aggregation neural network strategy was developed. DataDTA was compared with state-of-the-art methods on different datasets, and the results showed that DataDTA is a reliable prediction tool for affinities estimation. Specifically, the concordance index (CI) of DataDTA is 0.806 and the Pearson correlation coefficient (R) value is 0.814 on the test dataset, which is higher than other methods. AVAILABILITY AND IMPLEMENTATION: The codes and datasets of DataDTA are available at https://github.com/YanZhu06/DataDTA.

Measuring the semantic similarity between Gene Ontology (GO) terms is a fundamental step in numerous functional bioinformatics applications. To fully exploit the metadata of GO terms, word embedding-based methods have been proposed recently to map GO terms to low-dimensional feature vectors. However, these representation methods commonly overlook the key information hidden in the whole GO structure and the relationship between GO terms. In this paper, we propose a novel representation model for GO terms, named GT2Vec, which jointly considers the GO graph structure obtained by graph contrastive learning and the semantic description of GO terms based on BERT encoders. Our method is evaluated on a protein similarity task on a collection of benchmark datasets. The experimental results demonstrate the effectiveness of using a joint encoding graph structure and textual node descriptors to learn vector representations for GO terms.

MOTIVATION: The capability to predict the potential drug binding affinity against a protein target has always been a fundamental challenge in silico drug discovery. The traditional experiments in vitro and in vivo are costly and time-consuming which need to search over large compound space. Recent years have witnessed significant success on deep learning-based models for drug-target binding affinity prediction task. RESULTS: Following the recent success of the Transformer model, we propose a multigranularity protein-ligand interaction (MGPLI) model, which adopts the Transformer encoders to represent the character-level features and fragment-level features, modeling the possible interaction between residues and atoms or their segments. In addition, we use the convolutional neural network to extract higher-level features based on transformer encoder outputs and a highway layer to fuse the protein and drug features. We evaluate MGPLI on different protein-ligand interaction datasets and show the improvement of prediction performance compared to state-of-the-art baselines. AVAILABILITY AND IMPLEMENTATION: The model scripts are available at https://github.com/IILab-Resource/MGDTA.git.

The compound-protein binding affinity (CPA) pre-diction is vital for drug discovery and drug repurposing. Deep learning methods have been developed to model the complicated relationship between CPA and the sequences or structures of proteins and molecules. This study proposes a novel deep learning method, SeqGO-CPA, integrating protein function knowledge represented by Gene Ontology (GO) annotations in the CPA prediction. To capture the semantic information of GO annotations, a fine-tuned natural language processing model for biomedical domains is utilized to encode the set of GO terms. Meanwhile, based on the observation that CPA often occurs in sub-structures, our method uses the tokenization algorithm to learn sub-structure information of proteins and compounds from a large number of unlabeled sequences. Further, a deep neural network architecture involving the jointly-feature representation and a highway block is developed to enhance the CPA prediction ability. The proposed model was evaluated on two public benchmark datasets in both standard cross-validation and blinding split settings. The experimental results demonstrate our method outperforms the deep learning-based baselines, meanwhile the incorporating of GO information further improves the prediction performance.

Ontologies are essential description tools for biomedical concepts and entities, supporting biomedical fundamental research such as semantic similarity analysis, protein-protein interaction prediction and so on. An increasing amount of ontology-like domain knowledge is published in scientific publications, meanwhile, advanced natural language processing (NLP) techniques have been widespread to extract information from text resources automatically, both of which facilitate the exploration of the semantic representation of biomedical ontologies. We propose a novel distributional semantic representation methodology based on the combination of two pre-trained and domain-specific word embedding tools, the non-contextualized Word2Vec and the context-dependent NCBI-blueBERT, to enhance the encoding ability for biomedical ontologies. Furthermore, we utilize a randomly initialized bidirectional LSTM to project the obtained word vector sequence to a fixed-length sentence vector, facilitating a flexible and uniform way for the computation of downstream tasks. We evaluate our method in two categories of tasks: the similarity access of ontology terms, and the ontology annotation-based protein-protein interaction classification. Experimental results demonstrate that our method provides encouraging results compared to the baselines in all tests. Our approach offers promising opportunities for representing ontologies semantics and in turn characterizing entities including proteins in biomedical research.

Cell tracking automatically in time-lapse image sequences is important for understanding the dynamic pattern of micro-cell. In this paper, we present a novel method for tracking cell with shape feature based on the delta generalized labeled multi-Bernoulli (delta-GLMB) filter which is of great research significance. The delta-GLMB filter with cell shape parameters can improve the tracking accuracy. This approach is evaluated and compared with raw detection using the generalized optimal sub-pattern assignment (GOSPA) metric on real N2DH-SIM cell sequences. Experiment results show that the delta-GLMB filter can provide the shape information as well as the better estimation than raw detection and KTH method.

Research on biological objects requires tracking hundreds of micro-objects from the microscopy video. We propose an automated tracking framework to extract trajectories of micro-objects. This framework uses a particle probability hypothesis density (PF-PHD) tracker to implement a recursive Bayesian state estimation and trajectories association. In the framework, an ellipse target model is presented to describe the micro-objects with shape parameters instead of point-like targets. Furthermore, an orientation and positional constraint model is developed to deal with the data association of crossing trajectories in multitarget tracking. Using this framework, a significantly larger number of tracks are obtained than manual tracking. The experiments on simulated image sequences of microtubule movement are performed in order to evaluate the proposed PF-PHD tracking method.