Protein-protein interactions (PPIs) play an important role in most of the biological processes. How to correctly and efficiently detect protein interaction is a problem that worth studying. Although high-throughput technologies provide possibility large-scale PPIs, these cannot be used whole unreliable data may generated. To solve this problem, study, novel computational method was proposed effectively predict PPIs using information sequence. The present adopts Zernike moments extract...

Many life activities and key functions in organisms are maintained by different types of protein⁻protein interactions (PPIs). In order to accelerate the discovery PPIs for species, many computational methods have been developed. Unfortunately, even though constantly evolving, efficient predicting from protein sequence information not found years due limiting factors including both methodology technology. Inspired similarity biological sequences languages, developing a language processing...

Predicting drug-target interactions (DTIs) is crucial in innovative drug discovery, repositioning and other fields. However, there are many shortcomings for predicting DTIs using traditional biological experimental methods, such as the high-cost, time-consumption, low efficiency, so on, which make these methods difficult to widely apply. As a supplement, silico method can provide helpful information predictions of timely manner. In this work, deep walk embedding developed from...

Protein–protein interactions (PPIs) are essential for most living organisms’ process. Thus, detecting PPIs is extremely important to understand the molecular mechanisms of biological systems. Although many data have been generated by high-throughput technologies a variety organisms, whole interatom still far from complete. In addition, has some unavoidable defects, including time consumption, high cost, and error rate. recent years, with development machine learning, computational methods...

Abstract Effectively representing Medical Subject Headings (MeSH) headings (terms) such as disease and drug discriminative vectors could greatly improve the performance of downstream computational prediction models. However, these terms are often abstract difficult to quantify. In this paper, we converted MeSH tree structure into a relationship network applied several graph embedding algorithms on it represent terms. Specifically, consisting nodes (MeSH headings) edges (relationships), which...

Phishing attacks have always been a security issue that has attracted great attention in the cyber community. Recently, famous pre-trained models is being used as an anti-phishing solution. However, existing studies either simply transfer on text to phishing detection task, or pre-train using only extremely small samples. In this paper, we propose PhishBERT, veritable pretrained deep transformer network model for URL detection. Using tailor pre-training objective, PhishBERT obtained general...

The advent of blockchain technology has facilitated the widespread adoption smart contracts in financial sector. However, current fraud detection methodologies exhibit limitations capturing both global structural patterns within transaction networks and local semantic relationships embedded data. Most existing models focus on either information or features individually, leading to suboptimal performance detecting complex patterns.In this paper, we propose a dynamic feature fusion model that...

It is significant for biological cells to predict self-interacting proteins (SIPs) in the field of bioinformatics. SIPs mean that two or more identical can interact with each other by one gene expression. This plays a major role evolution protein‒protein interactions (PPIs) and cellular functions. Owing limitation experimental identification proteins, it develop useful tool prediction from protein sequence information. Therefore, we propose novel model called RP-FFT merges Random Projection...

Protein–protein interactions (PPI) are key to protein functions and regulations within the cell cycle, DNA replication, cellular signaling. Therefore, detecting whether a pair of proteins interact is great importance for study molecular biology. As researchers have become aware computational methods in predicting PPIs, many techniques been developed performing this task computationally. However, there few technologies that really meet needs their users. In paper, we develop novel efficient...

Various biochemical functions of organisms are performed by protein-protein interactions (PPIs). Therefore, recognition is very important for understanding most life activities, such as DNA replication and transcription, protein synthesis secretion, signal transduction metabolism. Although high-throughput technology makes it possible to generate large-scale PPIs data, requires expensive cost both time labor, leave a risk high false positive rate. In order formulate more ingenious solution,...

Protein-protein interactions (PPIs) play a very large part in most cellular processes. Although great deal of research has been devoted to detecting PPIs through high-throughput technologies, these methods are clearly expensive and cumbersome. Compared with the traditional experimental methods, computational have attracted much attention because their good performance PPIs. In our work, novel method named as PCVM-LM is proposed which combines probabilistic classification vector machine...

Abstract Abundant life activities are maintained by various biomolecule relationships in human cells. However, many previous computational models only focus on isolated objects, without considering that cell is a complete entity with ample functions. Inspired holism, we constructed Molecular Associations Network (MAN) including 9 kinds of among 5 types biomolecules, and prediction model called MAN-GF. More specifically, biomolecules can be represented as vectors the algorithm biomarker2vec...

Abstract The advent of single-cell sequencing technologies has revolutionized cell biology studies. However, integrative analyses diverse data face serious challenges, including technological noise, sample heterogeneity, and different modalities species. To address these problems, we propose scCorrector, a variational autoencoder-based model that can integrate from studies map them into common space. Specifically, designed Study Specific Adaptive Normalization for each study in decoder to...

Advanced driver assistance systems (ADAS) have matured over the past few decades with dedication to enhance user experience and gain a wider market penetration. However, personalization components, as means make current technologies more acceptable trustworthy for users, has only recently been gaining momentum. In this work we develop an algorithm learning personalized longitudinal driving behaviors via Gaussian Process (GP) model. The proposed method learns from individual driver's...

Abstract Background Protein subcellular localization plays a crucial role in understanding cell function. Proteins need to be the right place at time, and combine with corresponding molecules fulfill their functions. Furthermore, prediction of protein location not only should guiding drug design development due potential molecular targets but also an essential genome annotation. Taking current status image-based as example, there are three common drawbacks, i.e., obsolete datasets without...

Self-interacting Proteins (SIPs) plays a critical role in series of life function most living cells. Researches on SIPs are important part molecular biology. Although numerous data be provided, traditional experimental methods labor-intensive, time-consuming and costly can only yield limited results real-world needs. Hence,it's urgent to develop an efficient computational prediction method fill the gap. Deep learning technologies have proven produce subversive performance improvements many...

This paper introduces a toolbox for generating virtual trajectories, which we call VT-tools v1.0, to address challenges in analyzing large but imperfect trajectory datasets. v1.0 is able generate trajectories from raw datasets that are typically challenging process due their size. We also provide set of these resulting I–24 MOTION INCEPTION data and demonstrate the practical utility as-sessing speed variability travel times across different lanes within dataset. The opens future research on...

Most of the electroencephalogram (EEG) emotion recognitions are conducted in linear Euclidean space. However, it is difficult to accurately describe nonlinear characteris-tics multivariate EEG signals. Comparatively, Riemannian manifold a space which features can be analyzed more thoroughly. Therefore, inspired by geographical knowledge, an recognition methodology based on geomorphological (GFRM) proposed. Firstly, terms Wasserstein scalar curvature, automatic search strategy developed...

URLs play a crucial role in understanding and categorizing web content, particularly tasks related to security control online recommendations. While pre-trained models are currently dominating various fields, the domain of URL analysis still lacks specialized models. To address this gap, paper introduces URLBERT, first representation learning model applied variety classification or detection tasks. We train tokenizer on corpus billions data tokenization. Additionally, we propose two novel...