Exosomes, a subtype of extracellular vesicles secreted by mammalian cells with typical size range 30-150 nm, have been implicated in many biological processes as intercellular communication carriers. The isolation exosomes is an essential and challenging step before subsequent analysis functional studies, due to the complexity body fluids, well small low density exosomes. Ultracentrifugation (UC) exclusion chromatography (SEC) are two methods that extensively used for studies recent years....

In our previous work, we have demonstrated an integrated proteome analysis device (iPAD-100) to analyze proteomes from 100 cells. (1) this for the first time, a novel single-cell (iPAD-1) was developed profile proteins in single cell within 1 h. iPAD-1, selected directly sucked into 22 μm i.d. capillary. Then lysis and protein digestion were simultaneously accomplished capillary 2 nL volume, which could prevent loss excessive dilution. Digestion accelerated by using elevated temperature with...

The heterogeneous populations of exosomes with distinct nanosize have impeded our understanding their corresponding function as intercellular communication agents. Profiling signaling proteins packaged in each size-dependent subtype can disclose this heterogeneity exosomes. Herein, new strategy was developed for deconstructing distinct-size urine exosome subpopulations by profiling N-glycoproteomics and phosphoproteomics simultaneously. Two-dimension size exclusion liquid chromatography...

Continuous monitoring of the spatio-temporal dynamic behavior critical infrastructure networks, such as power systems, is a challenging but important task. In particular, accurate and timely prediction (electro-mechanical) transient trajectories grid necessary for early detection any instability prevention catastrophic failures. Existing approaches either rely on availability physical models system, use computationally expensive time-domain simulations, or are applicable only at local...

Recent AI agents, such as ChatGPT and LLaMA, primarily rely on instruction tuning reinforcement learning to calibrate the output of large language models (LLMs) with human intentions, ensuring outputs are harmless helpful. Existing methods heavily depend manual annotation high-quality positive samples, while contending issues noisy labels minimal distinctions between preferred dispreferred response data. However, readily available toxic samples clear safety often filtered out, removing...

Direction-of-Arrival (DOA) estimation capability of sensor arrays is greatly influenced by geometric configuration arrays. We proposed a new type 2-D planar array, improved Archimedean spiral array (IASA), to arrange elements and enhance the performance through curves. The MUSIC algorithm used estimate spatial spectrum signals. analysis complemented with numerical simulations. By comparing root mean square error, IASA has better DOA results than circular rectangular concentric ring array.

Electroencephalography (EEG) eye state classification is important and useful to detect human's cognition state. Previous research has validated the feasibility of machine learning statistical approaches for EEG classification. This paper proposes a novel identification approach based on Incremental Attribute Learning (IAL). Experimental results show that, with proper feature extraction ordering, IAL can not only cope time series problems efficiently, but also exhibit better performance in...

In this paper, we propose an extended deep learning approach that incorporates instance selection and bootstrapping techniques for imbalanced data classification. supervised learning, classification performance often deteriorates when the training set is where at least one of classes has a substantially fewer number instances than others. We to use adaptive synthetic sampling (ADASYN) generate minority class. A pruning process based on multiple correspondence analysis (MCA) then performed...

This paper presents a hybrid Maximum Power Point Tracking (MPPT) method for improving the power –conversion efficiency of Photovoltaic (PV) generators. By detecting output changes caused by environmental reasons, proposed performs variable-step online search process with an accurate estimation (MPP) locus. A PV generator Single Ended Primary Inductance Converter (SEPIC) is developed in PSIM to verify feasibility and suitability method. Simulation results show that it can not only deliver...

Bio-inspired metaheuristic algorithms have been widely proposed to estimate parameters of photovoltaic (PV) models in recent years due its ability handle nonlinear functions regardless the derivatives information. However, these normally utilize multiple agents/particles search process, and it takes much time possible solutions whole domain by sequential computing devices. This paper proposes parallel particle swarm optimization (PPSO) method extract a PV model. The algorithm is implemented...

Abstract In this paper, the direction of arrival (DOA) estimation signals in presence impulsive noise environment is studied. Complex isotropic symmetric alpha-stable ( SαS ) random variables are modeled as noise, then a novel second-order statistic method that correntropy-based covariance matrix (CBCM) defined, based on combination CBCM array sensor outputs with signal subspace technique (e.g., multiple classification (MUSIC)), which can be achieved source localization under environments....

Catastrophes have caused tremendous damages in human history and triggered record high post-disaster relief from the governments. The research of catastrophic modeling can help estimate effects natural disasters like hurricanes, floods, surges, earthquakes. In every Atlantic hurricane season, state Florida United States has potential to suffer economic losses hurricanes. Public Hurricane Loss Model (FPHLM), funded by Office Insurance Regulation, assisted residential insurance industry for...

A number of soft computing approaches such as neural networks, evolutionary algorithms, and fuzzy logic have been widely used for classifier agents to adaptively evolve solutions on classification problems. However, most work in the literature focuses learning ability individual agent. This article explores incremental, collaborative a multiagent environment. We use genetic algorithm (GA) incremental GA (IGA) main techniques rule set apply new class acquisition typical example illustrate...

Backbones refer to critical tree structures that span a set of nodes interests in networks. This paper introduces novel class attributed backbones and detection algorithms richly Unlike conventional models, capture dynamics edge cost model: it specifies affinitive attributes for each edge, the is dynamically determined by selection its associated closeness their values at end nodes. The backbone discovery compute an covers interested with smallest connection selected attributes. While this...

The power-voltage (P-V) characteristics of a photovoltaic (PV) array become more complex with multiple peaks under the partial shading conditions. This may cause failure for conventional perturb and observe (P&O) method in tracking maximum power points. paper introduces new two-stage point (MPPT) framework, known as estimation revision (ER). It intends to combine offline random search using metaheuristic algorithms online MPPT method. feasibility ER is verified an system implemented specific...

Data analytical pipelines routinely involve various domain-specific data science models. Such models require expensive manual or training effort and often incur validation costs (e.g., via scientific simulation analysis). Meanwhile, high-value remain to be ad-hocly created, isolated, underutilized for a broad community. Searching accessing proper analysis is desirable yet challenging users without domain knowledge. This paper introduces ModsNet, novel MODel SelectioN framework that only...

This paper investigates the problem of subgraph query generation with output that satisfies both diversity and fairness constraints. Given a set groups associated cardinality requirements, it is to compute queries diversified meanwhile covers desired cardinality. Such need evident in web social search We formalize as bi-criteria optimization on properties queries, verify its hardness approximability. show Σp2 , remains NP-complete even for single-node queries. Despite hardness, (1) we...

We introduce the concept of a neural network based recursive clustering which creates an ensemble clusters by decomposition data. The work involves hybrid combination global algorithm followed corresponding local algorithm. Evolutionary self organizing maps are used to create clusters. A set core patterns is isolated and separately trained using SOM. process recursively applied remaining partition each recursion integrated with previous recursion. correlation ground truth information (in...