Adverse drug-drug interactions (DDIs) are crucial for drug research and mainly cause morbidity mortality. Thus, the identification of potential DDIs is essential doctors, patients society. Existing traditional machine learning models rely heavily on handcraft features lack generalization. Recently, deep approaches that can automatically learn from molecular graph or drug-related network have improved ability computational to predict unknown DDIs. However, previous works utilized large...

Abstract The biomedical literature is growing rapidly, and the extraction of meaningful information from large amount increasingly important. Biomedical named entity (BioNE) identification one critical fundamental tasks in text mining. Accurate entities facilitates performance other tasks. Given that an end-to-end neural network can automatically extract features, several deep learning-based methods have been proposed for BioNE recognition (BioNER), yielding state-of-the-art performance. In...

Abstract Spatial structures of proteins are closely related to protein functions. Integrating improves the performance protein–protein interaction (PPI) prediction. However, limited quantity known restricts application structure-based prediction methods. Utilizing predicted structure information is a promising method improve sequence-based We propose novel end-to-end framework, TAGPPI, predict PPIs using sequence alone. TAGPPI extracts multi-dimensional features by employing 1D convolution...

Prediction of the drug–target affinity (DTA) plays an important role in drug discovery. Existing deep learning methods for DTA prediction typically leverage a single modality, namely simplified molecular input line entry specification (SMILES) or amino acid sequence to learn representations. SMILES sequences can be encoded into different modalities. Multimodality data provide kinds information, with complementary roles prediction. We propose Modality-DTA, novel method that leverages...

Tissue P systems with promoters provide nondeterministic parallel bioinspired devices that evolve by the interchange of objects between regions, determined existence some special called promoters. However, in cellular biology, movement molecules across a membrane is transported from high to low concentration. Inspired this biological fact, article, an interesting type tissue systems, monodirectional promoters, where communication happens two regions only one direction, considered. Results...

Abstract Single-cell RNA sequencing (scRNA-seq) is a revolutionary breakthrough that determines the precise gene expressions on individual cells and deciphers cell heterogeneity subpopulations. However, scRNA-seq data are much noisier than traditional high-throughput RNA-seq because of technical limitations, leading to many studies about dimensionality reduction visualization remaining at basic data-stacking stage. In this study, we propose an improved variational autoencoder model (termed...

Spiking neural P systems (SN systems, for short) are a class of distributed parallel computing devices inspired from the way neurons communicate by means spikes, where each neuron can have several spiking rules and forgetting work in sense that fire should at computation step. In this work, we consider SN with restrictions: 1) simple (resp. almost simple) has only one rule except neuron); 2) step neuron(s) maximum number spikes among spike will fire. These restrictions correspond to or...

Molecular interaction prediction is essential in various applications including drug discovery and material science. The problem becomes quite challenging when the represented by unmapped relationships molecular networks, namely interaction, because it easily suffers from (i) insufficient labeled data with many false-positive samples, (ii) ignoring a large number of biological entities rich information knowledge graph. Most existing methods cannot properly exploit graph molecule...

Monodirectional tissue P systems with promoters are natural inspired parallel computing paradigms, where only symport rules permitted, and the restriction of "monodirectionality", objects for two given regions transferred in one direction. In this article, a novel kind systems, monodirectional evolutional (MESTP systems) is raised, may be revised during movement between regions. The computational theory MESTP that employed flat maximally pattern investigated. We prove finite number sets...

Abstract Drug–gene interaction prediction occupies a crucial position in various areas of drug discovery, such as repurposing, lead discovery and off-target detection. Previous studies show good performance, but they are limited to exploring the binding interactions ignoring other relationships. Graph neural networks have emerged promising approaches owing their powerful capability modeling correlations under drug–gene bipartite graphs. Despite widespread adoption graph network-based...

Adverse drug-drug interactions (DDIs) pose potential risks in polypharmacy due to unknown physicochemical incompatibilities between co-administered drugs. Recent studies have utilized multi-layer graph neural network architectures model hierarchical molecular substructures of drugs, achieving excellent DDI prediction performance. While extant substructural frameworks effectively encode from atom-level features, they overlook valuable chemical bond representations within graphs. More...

Molecular design inherently involves the optimization of multiple conflicting objectives, such as enhancing bio-activity and ensuring synthesizability. Evaluating these objectives often requires resource-intensive computations or physical experiments. Current molecular methodologies typically approximate Pareto set using a limited number molecules. In this paper, we present an innovative approach, called Multi-Objective Design through Learning Latent Set (MLPS). MLPS initially utilizes...

Inductive Knowledge Graph Completion (KGC) aims to infer missing facts between newly emerged entities within knowledge graphs (KGs), posing a significant challenge. While recent studies have shown promising results in inferring such through subgraph reasoning, they suffer from (i) the semantic inconsistencies of similar relations, and (ii) noisy interactions inherent KGs due presence unconvincing for emerging entities. To address these challenges, we propose Semantic Structure-aware...

Tissue P systems are a class of bio-inspired computing models motivated by biochemical interactions between cells in tissue-like arrangement. with cell division offer theoretical device to generate an exponentially growing structure order solve computationally hard problems efficiently the assumption that there exists global clock mark time for system, execution each rule is completed exactly one unit. Actually, different reactions depends on many uncertain factors. In this work, biological...