Powerful Method Research Articles

Unveiling the Threat: Investigating Distributed and Centralized Backdoor Attacks in Federated Graph Neural Networks

Graph neural networks (GNNs) have gained significant popularity as powerful deep learning methods for processing graph data. However, centralized GNNs face challenges in data-sensitive scenarios due to privacy concerns and regulatory restrictions. Federated learning has emerged as a promising technology that enables collaborative training of a shared global model while preserving privacy. Although federated learning has been applied to train GNNs, no research focuses on the robustness of Federated GNNs against backdoor attacks. This article bridges this research gap by investigating two types of backdoor attacks in Federated GNNs: centralized backdoor attack (CBA) and distributed backdoor attack (DBA). Through extensive experiments, we demonstrate that DBA exhibits a higher success rate than CBA across various scenarios. To further explore the characteristics of these backdoor attacks in Federated GNNs, we evaluate their performance under different scenarios, including varying numbers of clients, trigger sizes, poisoning intensities, and trigger densities. Additionally, we explore the resilience of DBA and CBA against two defense mechanisms. Our findings reveal that both defenses cannot eliminate DBA and CBA without affecting the original task. This highlights the necessity of developing tailored defenses to mitigate the novel threat of backdoor attacks in Federated GNNs.

Fine-mapping causal tissues and genes at disease-associated loci.

Heritable diseases often manifest in a highly tissue-specific manner, with different disease loci mediated by genes in distinct tissues or cell types. We propose Tissue-Gene Fine-Mapping (TGFM), a fine-mapping method that infers the posterior probability (PIP) for each gene-tissue pair to mediate a disease locus by analyzing GWAS summary statistics (and in-sample LD) and leveraging eQTL data from diverse tissues to build cis-predicted expression models; TGFM also assigns PIPs to causal variants that are not mediated by gene expression in assayed genes and tissues. TGFM accounts for both co-regulation across genes and tissues and LD between SNPs (generalizing existing fine-mapping methods), and incorporates genome-wide estimates of each tissue's contribution to disease as tissue-level priors. TGFM was well-calibrated and moderately well-powered in simulations; unlike previous methods, TGFM was able to attain correct calibration by modeling uncertainty in cis-predicted expression models. We applied TGFM to 45 UK Biobank diseases/traits (average N = 316K) using eQTL data from 38 GTEx tissues. TGFM identified an average of 147 PIP > 0.5 causal genetic elements per disease/trait, of which 11% were gene-tissue pairs. Implicated gene-tissue pairs were concentrated in known disease-critical tissues, and causal genes were strongly enriched in disease-relevant gene sets. Causal gene-tissue pairs identified by TGFM recapitulated known biology (e.g., TPO -thyroid for Hypothyroidism), but also included biologically plausible novel findings (e.g., SLC20A2 -artery aorta for Diastolic blood pressure). Further application of TGFM to single-cell eQTL data from 9 cell types in peripheral blood mononuclear cells (PBMC), analyzed jointly with GTEx tissues, identified 30 additional causal gene-PBMC cell type pairs at PIP > 0.5-primarily for autoimmune disease and blood cell traits, including the biologically plausible example of CD52 in classical monocyte cells for Monocyte count. In conclusion, TGFM is a robust and powerful method for fine-mapping causal tissues and genes at disease-associated loci.

Schiff‐Base Aluminum Complexes as Bifunctional Catalysts for the Selective Ring‐Opening Co‐Polymerization of Cyclohexene Oxide and Succinic Anhydride

AbstractThe ring‐opening copolymerization (ROCOP) of epoxides and cyclic anhydrides is emerging as a powerful method for the synthesis of polyesters with innovative structures and properties. In this work, a series of dialkyl‐aluminum complexes stabilized by Schiff‐base ligands with different functional side arms (pyridine, amine, or phosphine) and different substituents on the phenoxy moieties have been synthetized and tested as catalysts in the ROCOP of cyclohexene oxide (CHO) and succinic anhydride (SA). The selectivity of the catalysts in the ROCOP was significatively depending on the nature of the pendant arm and/ or of the substituents of the ligand framework. The complexes designed to have a dynamic coordination sphere in which the side arms simulate the role of a neutral cocatalyst favored the desired selectivity, giving poly(CHO‐alt‐SA) with narrow dispersity and negligible ether linkage. NMR investigations and DFT calculations rationalized the observed behavior.

A Complexity Dichotomy in Spatial Reasoning via Ramsey Theory

Constraint satisfaction problems (CSPs) for first-order reducts of finitely bounded homogeneous structures form a large class of computational problems that might exhibit a complexity dichotomy, P versus NP-complete. A powerful method to obtain polynomial-time tractability results for such CSPs is a certain reduction to polynomial-time tractable finite-domain CSPs defined over k -types, for a sufficiently large k . We give sufficient conditions when this method can be applied and apply these conditions to obtain a new complexity dichotomy for CSPs of first-order expansions of the basic relations of the well-studied spatial reasoning formalism RCC5. We also classify which of these CSPs can be expressed in Datalog. Our method relies on Ramsey theory; we prove that RCC5 has a Ramsey order expansion.

Cobalt/Photoredox Dual-Catalyzed Cross-Radical Coupling of Alkenes via Hydrogen Atom Transfer and Homolytic Substitution.

Cobalt-catalyzed metal hydride hydrogen atom transfer (MHAT) in combination with photoredox catalysis has emerged as a powerful synthetic method, owing to its redox nature and applicability to various radical precursors. Herein, we describe a cross-radical coupling reaction under cobalt/photoredox dual catalysis. MHAT and homolytic substitution (SH2) processes enabled Markovnikov-selective hydrobenzylation of di/trisubstituted alkenes, affording products with a quaternary carbon center in a redox-neutral manner.

Analysis and authentication of avocado oil by low-field benchtop NMR spectroscopy and chemometrics.

Avocado oil is a nutritious, edible oil produced from avocado fruit. It has high commercial value and is increasing in popularity, thus powerful analytical methods are needed to ensure its quality and authenticity. Recent advancements in low-field (LF) NMR spectroscopy allow for collection of high-quality data despite the use of low magnetic fields produced by non-superconductive magnets. Combined with chemometrics, LF NMR opens new opportunities in food analysis using targeted and untargeted approaches. Here, it was used to determine poly-, mono-, and saturated fatty acids in avocado oil. Although direct signal integration of LF NMR spectra was able to determine certain classes of fatty acids, it had several challenges arising from signal overlapping. Thus, we used partial least square regression and developed models with good prediction performance for fatty acid composition, with residual prediction deviation ranging 3.46-5.53 and root mean squared error of prediction CV ranging 0.46-2.48. In addition, LF NMR, combined with unsupervised and supervised methods, enabled the differentiation of avocado oil from other oils, namely, olive oil, soybean oil, canola oil, high oleic (OL) safflower oil, and high OL sunflower oil. This study showed that LF NMR can be used as an efficient alternative for the compositional analysis and authentication of avocado oil. PRACTICAL APPLICATION: Here, we describe the application of LF-NMR for fatty acid analysis and avocado oil authentication. LF-NMR can be an efficient tool for targeted and untargeted analysis, thus becoming an attractive option for companies, regulatory agencies, and quality control laboratories. This tool is especially important for organizations and entities seeking economic, user-friendly, and sustainable analysis solutions.

A Power Method for Computing the Dominant Eigenvalue of a Dual Quaternion Hermitian Matrix

In this paper, we first study the projections onto the set of unit dual quaternions, and the set of dual quaternion vectors with unit norms. Then we propose a power method for computing the dominant eigenvalue of a dual quaternion Hermitian matrix. For a strict dominant eigenvalue, we show the sequence generated by the power method converges to the dominant eigenvalue and its corresponding eigenvector linearly. For a general dominant eigenvalue, we establish linear convergence of the standard part of the dominant eigenvalue. Based upon these, we reformulate the simultaneous localization and mapping problem as a rank-one dual quaternion completion problem. A two-block coordinate descent method is proposed to solve this problem. One block has a closed-form solution and the other block is the best rank-one approximation problem of a dual quaternion Hermitian matrix, which can be computed by the power method. Numerical experiments are presented to show the efficiency of our proposed power method.

Assessment of false discovery rate control in tandem mass spectrometry analysis using entrapment.

A pressing statistical challenge in the field of mass spectrometry proteomics is how to assess whether a given software tool provides accurate error control. Each software tool for searching such data uses its own internally implemented methodology for reporting and controlling the error. Many of these software tools are closed source, with incompletely documented methodology, and the strategies for validating the error are inconsistent across tools. In this work, we identify three different methods for validating false discovery rate (FDR) control in use in the field, one of which is invalid, one of which can only provide a lower bound rather than an upper bound, and one of which is valid but under-powered. The result is that the field has a very poor understanding of how well we are doing with respect to FDR control, particularly for the analysis of data-independent acquisition (DIA) data. We therefore propose a new, more powerful method for evaluating FDR control in this setting, and we then employ that method, along with an existing lower bounding technique, to characterize a variety of popular search tools. We find that the search tools for analysis of data-dependent acquisition (DDA) data generally seem to control the FDR at the peptide level, whereas none of the DIA search tools consistently controls the FDR at the peptide level across all the datasets we investigated. Furthermore, this problem becomes much worse when the latter tools are evaluated at the protein level. These results may have significant implications for various downstream analyses, since proper FDR control has the potential to reduce noise in discovery lists and thereby boost statistical power.

A random active set method for strictly convex quadratic problem with simple bounds

The active set method aims at finding the correct active set of the optimal solution and it is a powerful method for solving strictly convex quadratic problems with bound constraints. To guarantee the finite step convergence, existing active set methods all need strict conditions or some additional strategies, which can significantly impact the efficiency of the algorithm. In this paper, we propose a random active set method that introduces randomness in the active set’s update process. We prove that the algorithm can converge in a finite number of iterations with probability one, without any extra conditions on the problem or any supplementary strategies. At last, numerical experiments show that the algorithm can obtain the correct active set within a few iterations, and it has better efficiency and robustness than the existing methods.

BnglViz: online visualization of rule-based models.

Rule-based modeling is a powerful method to describe and simulate interactions among multi-site molecules and multi-molecular species, accounting for the internal connectivity of molecules in chemical species. This modeling technique is implemented in BioNetGen software that is used by various tools and software frameworks, such as BioNetGen stand-alone software, NFSim simulation engine, Virtual Cell simulation and modeling framework, SmolDyn and PySB software tools. These tools exchange models using BioNetGen scripting language (BNGL). Until now, there was no online visualization of such rule-based models. Modelers and researchers reading the manuscripts describing rule-based models had to learn BNGL scripting or master one of these tools to understand the models. Here, we introduce bnglViz, an online platform for visualizing BNGL files as graphical cartoons, empowering researchers to grasp the nuances of rule-based models swiftly and efficiently, and making the exploration of complex biological systems more accessible than ever before. The produced visualizations can be used as supplemental figures in publications or as a way to annotate BNGL models on web repositories. Available at https://bnglviz.github.io/.

Integrated electronic system for FET biosensor assessment based on current-voltage curve tracing

Field-effect transistor (FET) biosensors are pivotal in diverse applications, from environmental monitoring to healthcare diagnostics. Current-voltage (I-V) curve tracing is a powerful method for evaluating FET biosensor behavior, enabling comprehensive analysis of their FET biosensor characteristics. Traditional I-V curve tracing methods often require complex and expensive equipment, limiting their accessibility and practicality for routine sensor assessment. This study aims to develop and demonstrate an integrated electronic system for assessing FET biosensors using I-V curve tracing. The integrated electronic system uses readily available components, including microcontrollers, analog circuitry, and user-friendly software. We developed a compact, low-cost device that generates I-V curves for the FET biosensor. The integrated electronic system successfully generated I-V curves for various FET biosensors. The system demonstrated consistent, reliable performance, portability, and ease of use, making it a practical solution for routine sensor assessment. The average error in measurements using bipolar junction transistors (BJT) and metal-oxide-semiconductor field-effect transistors (MOSFETs) results in 2.62%, and measurements at different pH levels have a sensitivity of 21.6 mV/pH and a linearity of 0.9892. This innovation contributes to the advancement of FET biosensor technology. In the future, the developments should focus on ensuring their accuracy and reliability in various sensor fields.

A 37-Year-Old Man With Intellectual Disability Discovered to Have Aspartylglucosaminuria: Implications for the Diagnosis of Genetic Causes.

The causes of intellectual disability (ID) are varied, with as many as 1,400 causative genes. We attempted to identify the causative gene in a patient with long-standing undiagnosed ID. Although this was an isolated case with no family history, we searched for the causative gene using trio-based whole-exome sequencing (trio-WES), because severe ID is often caused by genetic variations, and inherited metabolic disorders (IMDs) are assumed to be the cause when regression and epilepsy occur. We identified homozygous donor splice-site variants in the AGA gene (aspartylglucosaminidase; NM_000027.4) Chr4(GRCh38):g. 177436275C>A, c.698+1G>T. This gene is implicated in aspartylglucosaminuria (AGU; OMIM #208400) and originated from both of the patient's parents. We confirmed the pathogenicity of the variant by detecting the splicing defect in cDNA from the patient's blood and accumulation of aberrant metabolites in the patient's urine. We discuss how to more readily achieve an accurate diagnosis for patients with undiagnosed intellectual disabilities. Medical practitioners' awareness of the characteristics of the disease leading to clinical suspicion in patients with matching presentations, and the performance of newborn screening when possible, is important for the diagnosis of ID. In addition, the characteristic symptoms and course of the disease give rise to suspicion of IMDs. Given our results, we consider trio-WES to be a powerful method for identifying the causative genes in cases of ID with genetic causes.

Multistep Forecasting Method for Offshore Wind Turbine Power Based on Multi-Timescale Input and Improved Transformer

Wind energy is highly volatile, and large-scale wind power grid integration significantly impacts grid stability. Accurate forecasting of wind turbine power can improve wind power consumption and ensure the economy of the power grid. This paper proposes a multistep forecasting method for offshore wind turbine power based on a multi-timescale input and an improved transformer. First, the wind speed sequence is decomposed by the VMD method to extract adequate timing information and remove the noise, after which the decomposition signals are merged with the rest of the timing features, and the dataset is split according to different timescales. A GRU receives the short-timescale inputs, and the Improved Transformer captures the timing relationship of the long-timescale inputs. Finally, a CNN is used to extract the information of each time point at the output of each branch, and the fully connected layer outputs multistep forecasting results. Experiments were conducted on operation data from four wind turbines located within the offshore wind farm but not near the edge. The results show that the proposed method achieved average errors of 0.0522 in MAE, 0.0084 in MSE, and 0.0907 in RMSE on a four-step forecast. This outperformed comparison methods LSTM, CNN-LSTM, LSTM-Attention, and Informer. The proposed method demonstrates superior forecasting performance and accuracy for multistep offshore wind turbine power forecasting.

Strained diamond for quantum sensing applications

Apart from being an extraordinary optical and electronic material, diamond has also found applications in quantum mechanics especially in quantum sensing with the discovery and research development of various color centers. Elastic strain engineering (ESE), as a powerful modulation method, can tune the quantum properties and improve the performance of diamond quantum sensors. In recent years, deep ESE (DESE, when >5% elastic strain, or >σ ideal/2 is achieved) has been realized in micro/nano-fabricated diamond and shows a great potential for tuning the quantum mechanical properties of diamond substantially. In this perspective, we briefly review the quantum properties of diamond and some of the corresponding sensing applications carried out with ESE, and look at how DESE could be applied for further tuning the quantum sensing properties of diamond with desired applications and what the critical challenges are.

Efficient small fragment sequencing of human, cow, and bison miRNA, small RNA or csRNA-seq libraries using AVITI.

Next-Generation Sequencing (NGS) catalyzed breakthroughs across various scientific domains. Illumina's sequencing by synthesis method has long been essential for NGS but emerging technologies like Element Biosciences' sequencing by avidity (AVITI) represent a novel approach. It has been reported that AVITI offers improved signal-to-noise ratios and cost reductions. However, the method relies on rolling circle amplification which can be impacted by polymer size, raising questions about its efficacy sequencing small RNAs (sRNA) molecules including microRNAs (miRNAs), piwi-interacting RNAs (piRNAs), and others that are crucial regulators of gene expression and involved in various biological processes. In addition, capturing capped small RNAs (csRNA-seq) has emerged as a powerful method to map active or "nascent" RNA polymerase II transcription initiation in tissues and clinical samples. Here, we report a new protocol for seamlessly sequencing short DNA fragments on the AVITI and demonstrate that AVITI and Illumina sequencing technologies equivalently capture human, cattle (Bos taurus) and the bison (Bison bison) sRNA or csRNA sequencing libraries, augmenting the confidence in both approaches. Additionally, analysis of generated nascent transcription start sites (TSSs) data for cattle and bison revealed inaccuracies in their current genome annotations and highlighted the possibility and need to translate small RNA sequencing methodologies to livestock. Our accelerated and optimized protocol therefore bridges the advantages of AVITI sequencing and critical methods that rely on sequencing short DNA fragments.

Transfer learning reveals sequence determinants of the quantitative response to transcription factor dosage.

Deep learning approaches have made significant advances in predicting cell type-specific chromatin patterns from the identity and arrangement of transcription factor (TF) binding motifs. However, most models have been applied in unperturbed contexts, precluding a predictive understanding of how chromatin state responds to TF perturbation. Here, we used transfer learning to train and interpret deep learning models that use DNA sequence to predict, with accuracy approaching experimental reproducibility, how the concentration of two dosage-sensitive TFs (TWIST1, SOX9) affects regulatory element (RE) chromatin accessibility in facial progenitor cells. High-affinity motifs that allow for heterotypic TF co-binding and are concentrated at the center of REs buffer against quantitative changes in TF dosage and strongly predict unperturbed accessibility. In contrast, motifs with low-affinity or homotypic binding distributed throughout REs lead to sensitive responses with minimal contributions to unperturbed accessibility. Both buffering and sensitizing features show signatures of purifying selection. We validated these predictive sequence features using reporter assays and showed that a biophysical model of TF-nucleosome competition can explain the sensitizing effect of low-affinity motifs. Our approach of combining transfer learning and quantitative measurements of the chromatin response to TF dosage therefore represents a powerful method to reveal additional layers of the cis-regulatory code.

Historical Pigments and Paint Layers: Raman Spectral Library with 852 nm Excitation Laser

Raman spectroscopy (RS), for its robust analytical capabilities under constant development, is a powerful method for the identification of various materials, in particular pigments in cultural heritage. Characterization of the artist’s palette is of fundamental importance for the correct formulation of restoration intervention as well as for preventive conservation of artworks. Here we examine the number and variability of research studies exploiting Bravo handheld Raman spectrophotometer relying on the excitation of Raman signal with temperature-shifted diode lasers emitting at 852 and 785 nm. To this end, we explore the spectral features of common historical pigments examined as powders and in the paint layer. We show that some materials may exhibit slightly different spectra as concerns especially the relative intensity of Raman lines with 852 nm laser excitation wavelength as compared to the standard 785 nm. The aim is to provide the research community with a reference spectral database that facilitates the identification of unknown pigments using the 852 nm excitation source.

Quasi-periodic oscillation analysis for a sample of blazars at the optical band

Quasi-periodic behavior in the light curves of blazars can help us understand the physics. The identification of the supermassive black hole binary (SMBHB) is an active topic, and the periodicity analysis of light curves is a powerful method for searching for these sources. In this work, we aim to identify quasi-periodic oscillations (QPOs) in the optical band light curves in our sample and discuss the possible physical origins behind these targets. In this study, we collected 155 optical band light curves from three different monitoring programs. We searched for QPOs in our sample using the generalised Lomb-Scargle (GLS) and weighted wavelet Z-transform (WWZ) methods. We simulated $10^ $ artificial light curves and evaluated the significance of the results using the Monte Carlo method. Our work reveals that 18 targets show QPOs with timescales ranging from 200 days to 1400 days. These QPOs could be explained by three scenarios, including the SMBHB, instability of thick disks and jet precession. Since the frequencies corresponding to QPOs are in the nHz regime, our work provides candidates of SMBHBs for further verification.

A real-time hydrogen consumption estimation method for fuel cell vehicles

ABSTRACT As an economic evaluation indicator for fuel cell vehicles (FCVs), real-time hydrogen consumption’s accuracy and immediateness are important. Real-time hydrogen consumption obtained through the fuel cell system (FC Power method) has strong immediateness but poor accuracy, while through the hydrogen mass (Mass method) has good accuracy but poor immediateness. Given the difficulty in obtaining sufficient training data on actual vehicles, this paper establishes a hydrogen storage system model whose input and output can be used to obtain data. Then, a new estimation method based on the prediction of the mass’s difference (PMD method) has been constructed based on the Neural basis expansion analysis with exogenous variables (NBEATSx) algorithm. After the NBEATSx algorithm is trained, the prediction accuracy of the PMD method is verified under urban and high-speed conditions. The result shows that the new method proposed in this article improves Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) by nearly 45% compared to the FC Power method, and the hydrogen consumption’s sampling point is advanced compared to the Mass method. The trained model can be input into Hydrogen Management System (HMS) which is installed on the hydrogen storage system and output data in real-time.

The SPECTRAL Perfusion Arm Clamping dAtaset (SPECTRALPACA) for video-rate functional imaging of the skin

Spectral imaging has the potential to become a key technique in interventional medicine as it unveils much richer optical information compared to conventional RBG (red, green, and blue)-based imaging. Thus allowing for high-resolution functional tissue analysis in real time. Its higher information density particularly shows promise for the development of powerful perfusion monitoring methods for clinical use. However, even though in vivo validation of such methods is crucial for their clinical translation, the biomedical field suffers from a lack of publicly available datasets for this purpose. Closing this gap, we generated the SPECTRAL Perfusion Arm Clamping dAtaset (SPECTRALPACA). It comprises ten spectral videos (∼20 Hz, approx. 20,000 frames each) systematically recorded of the hands of ten healthy human participants in different functional states. We paired each spectral video with concisely tracked regions of interest, and corresponding diffuse reflectance measurements recorded with a spectrometer. Providing the first openly accessible in human spectral video dataset for perfusion monitoring, our work facilitates the development and validation of new functional imaging methods.