Search

Abstract The current study aimed to investigate the status of genes with prognostic DNA methylation sites in bladder cancer (BLCA). We obtained bulk transcriptome sequencing data, methylation data, and single-cell sequencing data of BLCA from public databases. Initially, Cox survival analysis was conducted for each methylation site, and genes with more than 10 methylation sites demonstrating prognostic significance were identified to form the BLCA prognostic methylation gene set. Subsequently, the intersection of marker genes associated with epithelial cells in single-cell sequencing analysis was obtained to acquire epithelial cell prognostic methylation genes. Utilizing ten machine learning algorithms for multiple combinations, we selected key genes (METRNL, SYT8, COL18A1, TAP1, MEST, AHNAK, RPP21, AKAP13, RNH1) based on the C-index from multiple validation sets. Single-factor and multi-factor Cox analyses were conducted incorporating clinical characteristics and model genes to identify independent prognostic factors (AHNAK, RNH1, TAP1, Age, and Stage) for constructing a Nomogram model, which was validated for its good diagnostic efficacy, prognostic prediction ability, and clinical decision-making benefits. Expression patterns of model genes varied among different clinical features. Seven immune cell infiltration prediction algorithms were used to assess the correlation between immune cell scores and Nomogram scores. Finally, drug sensitivity analysis of Nomogram model genes was conducted based on the CMap database, followed by molecular docking experiments. Our research offers a reference and theoretical basis for prognostic evaluation, drug selection, and understanding the impact of DNA methylation changes on the prognosis of BLCA.

Abstract Optimizing resource allocation and routing to satisfy service needs is paramount in large-scale networks. Software-defined networking (SDN) is a new network paradigm that decouples forwarding and control, enabling dynamic management and configuration through programming, which provides the possibility for deploying intelligent control algorithms (such as deep reinforcement learning algorithms) to solve network routing optimization problems in the network. Although these intelligent-based network routing optimization schemes can capture network state characteristics, they are prone to falling into local optima, resulting in poor convergence performance. In order to address this issue, this paper proposes an African Vulture Routing Optimization (AVRO) algorithm for achieving SDN routing optimization. AVRO is based on the African Vulture Optimization Algorithm (AVOA), a population-based metaheuristic intelligent optimization algorithm with global optimization ability and fast convergence speed advantages. First, we improve the population initialization method of the AVOA algorithm according to the characteristics of the network routing problem to enhance the algorithm’s perception capability towards network topology. Subsequently, we add an optimization phase to strengthen the development of the AVOA algorithm and achieve stable convergence effects. Finally, we model the network environment, define the network optimization objective, and perform comparative experiments with the baseline algorithms. The experimental results demonstrate that the routing algorithm has better network awareness, with a performance improvement of 16.9% compared to deep reinforcement learning algorithms and 71.8% compared to traditional routing schemes.