Search Results (17)

Gradient Boosted Decision Trees (GBDTs) are popular for their strong predictive performance. However, in domains like finance and healthcare, data are often distributed across organizations, making collaborative model training challenging due to privacy concerns. Vertical federated learning (VFL) enables such collaboration when data are split by features, but many existing methods focus on protecting raw data while exposing sensitive model information, such as gradients and Hessians—especially to the label-owning party. Techniques like Homomorphic Encryption and Secret Sharing help, but often rely on trusted or privileged parties and may still leak intermediate statistics. To address this, we propose MPC-XGB , a privacy-preserving framework for training XGBoost under VFL with an honest-but-curious threat model. It uses secure three-party computation with Replicated Secret Sharing, distributing data across non-colluding servers and performing all computations on shares. This ensures that raw data, labels, and model statistics remain hidden, while supporting both secure training and prediction. Experiments show that MPC-XGB achieves strong performance (0.93 accuracy, 0.82 AUC), comparable to that of existing methods, with improved privacy guarantees. Full article

This paper presents Sillcom, a high-performance secure indoor localization scheme designed to minimize both communication and computational costs while preserving participants’ privacy. Unlike existing privacy-preserving indoor localization techniques, which suffer from high computational overhead and excessive communication, Sillcom integrates replicated secret sharing and function secret sharing in an outsourcing model to achieve significantly lower online communication overhead. A multi-branch tree structure and multi-thread parallelism further optimize both the offline and online phases. Experimental results demonstrate that Silcom outperforms the state-of-the-art online-efficient scheme FAPRIL, reducing online communication by a factor of 15 and end-to-end query time by 75%. Full article

This study investigates the interplay between cellular senescence and inflammation in human umbilical vein endothelial cells (HUVECs). We employed RNA sequencing to analyze gene expression changes in HUVECs subjected to replicative- or radiation-stress-induced senescence, and we compared these profiles with those of cells under acute or chronic TNFα-mediated inflammation. Our findings reveal that both senescence types exhibited significant upregulation of genes associated with epithelial- (or endothelial) mesenchymal transition (EMT) and inflammatory pathways, indicating a shared molecular response. Notably, chronic inflammation led to a pronounced EMT signature, while acute inflammation primarily activated classical inflammatory responses. Experimental validation confirmed reduced proliferation and increased secretion of pro-inflammatory cytokines (IL-6 and IL-8) in senescent and chronically inflamed cells and substantiated the upregulation of EMT marker genes. Additionally, we observed impaired wound healing capacity in senescent and chronically inflamed cells, highlighting the functional consequences of these cellular states. Our study underscores the critical role of inflammation in exacerbating senescence-related changes, contributing to the understanding of age-related cardiovascular pathologies. These insights may inform future therapeutic strategies aimed at mitigating the effects of aging and inflammation on endothelial function and cardiovascular health. Full article

Many leading technology companies currently offer Machine Learning as a Service Platform, enabling developers and organizations to access the inference capabilities of pre-trained models via API calls. However, due to concerns over user data privacy, inter-enterprise competition, and legal and regulatory constraints, directly utilizing pre-trained models in the cloud for inference faces security challenges. In this paper, we propose communication-efficient secure three-party protocols for recurrent neural network (RNN) inference. First, we design novel three-party secret-sharing protocols for digit decomposition, B2A conversion, enabling efficient transformation of secret shares between Boolean and arithmetic rings. Then, we propose the lookup table-based secure three-party protocol. Unlike the intuitive way of directly looking up tables to obtain results, we compute the results by utilizing the inherent mathematical properties of binary lookup tables, and the communication complexity of the lookup table protocol is only related to the output bit width. We also design secure three-party protocols for key functions in the RNN, including matrix multiplication, sigmoid function, and Tanh function. Our protocol divides the computation into online and offline phase, and places most of the computations locally. The theoretical analysis shows that the communication round of our work was reduced from four rounds to one round. The experiment results show that compared with the current SOTA-SIRNN, the online communication overhead of sigmoid and tanh functions decreased by 80.39% and 79.94%, respectively. Full article

Extrachromosomal circular DNAs (eccDNAs) has been found to be widespread and functional in various organisms. However, comparative analyses of pre- and post-infection of virus are rarely known. Herein, we investigated the changes in expression patterns of eccDNA following infection with Bombyx mori cytoplasmic polyhedrosis virus (BmCPV) and explore the role of eccDNA in viral infection. Circle-seq was used to analyze eccDNAs in the midgut of BmCPV-infected and BmCPV-uninfected silkworms. A total of 5508 eccDNAs were identified, with sizes varying from 72 bp to 17 kb. Most of eccDNAs are between 100 to 1000 bp in size. EccDNA abundance in BmCPV-infected silkworms was significantly higher than in BmCPV-uninfected silkworms. GO and KEGG analysis of genes carried by eccDNAs reveals that most are involved in microtubule motor activity, phosphatidic acid binding, cAMP signaling pathway, and pancreatic secretion signaling pathways. Several eccDNAs contain sequences of the transcription factor SOX6, sem-2, sp8b, or Foxa2. Association analysis of eccDNA-mRNA/miRNA/circRNA revealed that some highly expressed genes are transcribed from relevant sequences of eccDNA and the transcription of protein coding genes influenced the frequency of eccDNA. BmCPV infection resulted in changes in the expression levels of six miRNAs, but no known miRNAs with altered expression levels due to changes in eccDNA abundance were identified. Moreover, it was found that 1287 and 924 sequences representing back-spliced junctions of circRNAs were shared by the junctions of eccDNAs in the BmCPV-infected and uninfected silkworms, respectively, and some eccDNAs loci were shared by circRNAs on Chromosomes 2, 7, 11, 14, and 24, suggesting some eccDNAs may exert its function by being transcribed into circRNAs. These findings suggest that BmCPV infection alter the expression pattern of eccDNAs, leading to changes in RNA transcription levels, which may play roles in regulating BmCPV replication. In the future, further experiments are needed to verify the association between eccDNA-mRNA/miRNA/circRNA and its function in BmCPV infection. Full article

HBeAg is a non-structural, secreted protein of hepatitis B virus (HBV). Its p25 precursor is post-translationally modified in the endoplasmic reticulum. The G1862T precore mutation leads to the accumulation of P25 in the endoplasmic reticulum and activation of unfolded protein response. Using mass spectrometry, comparative proteome profiling of Huh-7 cells transfected with wildtype (WT) or G1862T revealed significantly differentially expressed proteins resulting in 12 dysregulated pathways unique to WT-transfected cells and 7 shared between cells transfected with either WT or G1862T. Except for the p38 MAPK signalling pathway, WT showed a higher number of DEPs than G1862T-transfected cells in all remaining six shared pathways. Two signalling pathways: oxidative stress and cell cycle signalling were differentially expressed only in cells transfected with G1862T. Fifteen pathways were dysregulated in G1862T-transfected cells compared to WT. The 15 dysregulated pathways were involved in the following processes: MAPK signalling, DNA synthesis and methylation, and extracellular matrix organization. Moreover, proteins involved in DNA synthesis signalling (replication protein A (RPA) and DNA primase (PRIM2)) were significantly upregulated in G1862T compared to WT. This upregulation was confirmed by mRNA quantification of both genes and immunofluorescent confocal microscopy for RPA only. The dysregulation of the pathways involved in these processes may lead to immune evasion, persistence, and uncontrolled proliferation, which are hallmarks of cancer. Full article

Proteus mirabilis is a leading cause of urinary tract infections and a common commensal of the gastrointestinal tract. Our recent study (JB) showed that P. mirabilis strain BL95 employs a novel contact-dependent killing system against enteric bacteria in the mouse gut and in vitro. To uncover the genetic determinants of this system, we performed whole-genome sequencing of BL95 and compared it with 98 complete genomes of P. mirabilis. BL95 carries 56 coding sequences (CDSs) not found in other P. mirabilis. Over half of these unique genes are located on a novel integrative conjugative element (ICE) named ICEPm2, inserted in tRNA-Phe and exclusive to BL95. ICEPm2 has integration, conjugation, and DNA replication modules nearly identical to ICEPm1 (common in P. mirabilis), but ICEPm2 of BL95 carries two unique operons for P. mirabilis—a phenazine biosynthesis and a contact-dependent growth inhibition (CDI) system. ICEPm2 is absent in the P. mirabilis (AR_0156) closest to BL95 and it is present in the genomes of several Escherichia coli from mouse intestines, indicating its recent horizontal mobilization. BL95 shares over 100 genes of five different secretion systems with other P. mirabilis, mostly poorly studied, making a large pool of candidate genes for the contact-dependent growth inhibition. Full article

Fog computing (FC) is a distributed architecture in which computing resources and services are placed on edge devices closer to data sources. This enables more efficient data processing, shorter latency times, and better performance. Fog computing was shown to be a promising solution for addressing the new computing requirements. However, there are still many challenges to overcome to utilize this new computing paradigm, in particular, reliability and security. Following this need, a systematic literature review was conducted to create a list of requirements. As a result, the following four key requirements were formulated: (1) low latency and response times; (2) scalability and resource management; (3) fault tolerance and redundancy; and (4) privacy and security. Low delay and response can be achieved through edge caching, edge real-time analyses and decision making, and mobile edge computing. Scalability and resource management can be enabled by edge federation, virtualization and containerization, and edge resource discovery and orchestration. Fault tolerance and redundancy can be enabled by backup and recovery mechanisms, data replication strategies, and disaster recovery plans, with a residual number system (RNS) being a promising solution. Data security and data privacy are manifested in strong authentication and authorization mechanisms, access control and authorization management, with fully homomorphic encryption (FHE) and the secret sharing system (SSS) being of particular interest. Full article

TRPM3 belongs to the melastatin sub-family of transient receptor potential (TRPM) cation channels and has been shown to function as a steroid-activated, heat-sensitive calcium ion (Ca²⁺) channel. A missense substitution (p.I65M) in the TRPM3 gene of humans (TRPM3) and mice (Trpm3) has been shown to underlie an inherited form of early-onset, progressive cataract. Here, we model the pathogenetic effects of this cataract-causing mutation using ‘knock-in’ mutant mice and human cell lines. Trpm3 and its intron-hosted micro-RNA gene (Mir204) were strongly co-expressed in the lens epithelium and other non-pigmented and pigmented ocular epithelia. Homozygous Trpm3-mutant lenses displayed elevated cytosolic Ca²⁺ levels and an imbalance of sodium (Na⁺) and potassium (K⁺) ions coupled with increased water content. Homozygous TRPM3-mutant human lens epithelial (HLE-B3) cell lines and Trpm3-mutant lenses exhibited increased levels of phosphorylated mitogen-activated protein kinase 1/extracellular signal-regulated kinase 2 (MAPK1/ERK2/p42) and MAPK3/ERK1/p44. Mutant TRPM3-M65 channels displayed an increased sensitivity to external Ca²⁺ concentration and an altered dose response to pregnenolone sulfate (PS) activation. Trpm3-mutant lenses shared the downregulation of genes involved in insulin/peptide secretion and the upregulation of genes involved in Ca²⁺ dynamics. By contrast, Trpm3-deficient lenses did not replicate the pathophysiological changes observed in Trpm3-mutant lenses. Collectively, our data suggest that a cataract-causing substitution in the TRPM3 cation channel elicits a deleterious gain-of-function rather than a loss-of-function mechanism in the lens. Full article

Cryptosystems based on supersingular isogeny are a novel tool in post-quantum cryptography. One compelling characteristic is their concise keys and ciphertexts. However, the performance of supersingular isogeny computation is currently worse than that of other schemes. This is primarily due to the following factors. Firstly, the underlying field is a quadratic extension of the finite field, resulting in higher computational complexity. Secondly, the strategy for large-degree isogeny evaluation is complex and dependent on the elementary arithmetic units employed. Thirdly, adapting the same hardware to different parameters is challenging. Considering the evolution of similar curve-based cryptosystems, we believe proper algorithm optimization and hardware acceleration will reduce its speed overhead. This paper describes a high-performance and flexible hardware architecture that accelerates isogeny computation. Specifically, we optimize the design by creating a dedicated quadratic Montgomery multiplier and an efficient scheduling strategy that are suitable for supersingular isogeny. The multiplier operates on ${\mathbb{F}}_{p^2}$ under projective coordinate formulas, and the scheduling is tailored to it. By exploiting additional parallelism through replicated multipliers and concurrent isogeny subroutines, our 65 nm SMIC technology cryptographic accelerator can generate ephemeral public keys in 2.40 ms for Alice and 2.79 ms for Bob with a 751-bit prime setting. Sharing the secret key costs another 2.04 ms and 2.35 ms, respectively. Full article

Neural networks have a wide range of promise for image prediction, but in the current setting of neural networks as a service, the data privacy of the parties involved in prediction raises concerns. In this paper, we design and implement a privacy-preserving neural network prediction model in the three-party secure computation framework over secret sharing of private data. Secret sharing allows the original data to be split, with each share held by a different party. The parties cannot know the shares owned by the remaining collaborators, and thus the original data can be kept secure. The three parties refer to the client, the service provider and the third server that assist in the computation, which is different from the previous work. Thus, under the definition of semi-honest and malicious security, we design new computation protocols for the building blocks of the neural network based on replicated secret sharing. Experimenting with MNIST dataset on different neural network architectures, our scheme improves 1.3×/1.5× and 7.4×/47.6× in terms of computation time as well as communication cost compared to the Falcon framework under the semi-honest/malicious security, respectively. Full article

Viruses share many attributes in common with extracellular vesicles (EVs). The cellular machinery that is used for EV production, packaging of substrates and secretion is also commonly manipulated by viruses for replication, assembly and egress. Viruses can increase EV production or manipulate EVs to spread their own genetic material or proteins, while EVs can play a key role in regulating viral infections by transporting immunomodulatory molecules and viral antigens to initiate antiviral immune responses. Ultimately, the interactions between EVs and viruses are highly interconnected, which has led to interesting discoveries in their associated roles in the progression of different diseases, as well as the new promise of combinational therapeutics. In this review, we summarize the relationships between viruses and EVs and discuss major developments from the past five years in the engineering of virus-EV therapies. Full article

Early- and late-phase radiation-induced lung injuries, namely pneumonitis and lung fibrosis (RILF), severely constrain the maximum dose and irradiated volume in thoracic radiotherapy. As the most radiosensitive targets, epithelial cells respond to radiation either by undergoing apoptosis or switching to a senescent phenotype that triggers the immune system and damages surrounding healthy cells. Unresolved inflammation stimulates mesenchymal cells’ proliferation and extracellular matrix (ECM) secretion, which irreversibly stiffens the alveolar walls and leads to respiratory failure. Although a thorough understanding is lacking, RILF and idiopathic pulmonary fibrosis share multiple pathways and would mutually benefit from further insights into disease progression. Furthermore, current normal tissue complication probability (NTCP) models rely on clinical experience to set tolerance doses for organs at risk and leave aside mechanistic interpretations of the undergoing processes. To these aims, we implemented a 3D agent-based model (ABM) of an alveolar duct that simulates cell dynamics and substance diffusion following radiation injury. Emphasis was placed on cell repopulation, senescent clearance, and intra/inter-alveolar bystander senescence while tracking ECM deposition. Our ABM successfully replicates early and late fibrotic response patterns reported in the literature along with the ECM sigmoidal dose-response curve. Moreover, surrogate measures of RILF severity via a custom indicator show qualitative agreement with published fibrosis indices. Finally, our ABM provides a fully mechanistic alveolar survival curve highlighting the need to include bystander damage in lung NTCP models. Full article

Privacy-preserving machine learning has become an important study at present due to privacy policies. However, the efficiency gap between the plain-text algorithm and its privacy-preserving version still exists. In this paper, we focus on designing a novel secret-sharing-based K-means clustering algorithm. Particularly, we present an efficient privacy-preserving K-means clustering algorithm based on replicated secret sharing with honest-majority in the semi-honest model. More concretely, the clustering task is outsourced to three semi-honest computing servers. Theoretically, the proposed privacy-preserving scheme can be proven with full data privacy. Furthermore, the experimental results demonstrate that our proposed privacy version reaches the same accuracy as the plain-text one. Compared to the existing privacy-preserving scheme, our proposed protocol can achieve about 16.5×–25.2× faster computation and 63.8×–68.0× lower communication. Consequently, the proposed privacy-preserving scheme is suitable for secret-sharing-based secure outsourced computation. Full article

Natural hepatitis C virus (HCV) infection is restricted to humans, whereas other primates such as rhesus macaques are non-permissive for infection. To identify human and rhesus macaque genes that differ or share the ability to inhibit HCV replication, we conducted a medium-throughput screen of lentivirus-expressed host genes that disrupt replication of HCV subgenomic replicon RNA expressing secreted Gaussia luciferase. A combined total of >800 interferon-stimulated genes (ISGs) were screened. Our findings confirmed established anti-HCV ISGs, such as IRF1, PKR and DDX60. Novel species–specific inhibitors were also identified and independently validated. Using a cell-based system that recapitulates productive HCV infection, we identified that over-expression of the ‘Rho Guanine Nucleotide Exchange Factor 3’ gene (ARHGEF3) from both species inhibits full-length virus replication. Additionally, replication of two mosquito-borne flaviviruses, yellow fever virus (YFV) and Zika virus (ZIKV), were also reduced in cell lines over-expressing ARHGEF3 compared to controls. In conclusion, we ascribe novel antiviral activity to the cellular gene ARHGEF3 that inhibits replication of HCV and other important human viral pathogens belonging to the Flaviviridae, and which is conserved between humans and rhesus macaques. Full article