AbstractBackground and ObjectiveIn recent years, the association between metabolites and complex human diseases has increasingly been recognized as a major research focus. Traditional wet-lab experiments are considere...
详细信息
AbstractBackground and ObjectiveIn recent years, the association between metabolites and complex human diseases has increasingly been recognized as a major research focus. Traditional wet-lab experiments are considered time-consuming and labor-intensive, while computational methods have been shown to significantly enhance research efficiency. However, existing methods for predicting metabolite-disease associations primarily depend on predefined similarity metrics and static network structures, often failing to capture the complex interactions among node neighborhoods within metabolite and disease networks. This limitation hinders the capture of deeper dynamic relationships between metabolites and diseases, resulting in information loss and noise that deteriorate prediction performance. MethodsAn innovative dynamic adaptive feature learning architecture (DAF-LA) is proposed to predict metabolite-disease associations. This architecture integrates dynamic subgraph construction and an adaptive feature enhancement mechanism, enabling high-precision feature learning and association prediction through progressive optimization from initial to high-order feature representations. ResultsThe architecture was evaluated through five-fold cross-validation, achieving an AUC of 0.9742 and an AUPR of 0.9734. Additionally, the case study demonstrates that DAF-LA accurately predicts metabolites associated with Alzheimer's disease, Type 2 diabetes mellitus and Parkinson's disease. ConclusionsThe results demonstrate that our method effectively uncovers potential associations between metabolites and diseases through dynamic topological modeling and multi-scale collaborative learning. It enables faster identification of likely metabolite-disease relationships, reduces the time and resource costs associated with inefficient large-scale screening in traditional wet-lab experiments, and provides more targeted guidance for subsequent biological validation.
Edge computing is an important field of artificial intelligence and customized manufacturing systems, which can reduce the processing pressure of the system. An autoencoder is a classic data dimensionality reduction t...
详细信息
Edge computing is an important field of artificial intelligence and customized manufacturing systems, which can reduce the processing pressure of the system. An autoencoder is a classic data dimensionality reduction tool in the field of edge computing, and the number of hidden layer neurons in the autoencoder is very important. However, there is currently no effective method to determine this number. To overcome this problem, this article elaborated an adaptive automatic encoder for edge computing data reduction for intelligent customized manufacturing system from the perspective of edge computing model reduction in customized manufacturing system. Specifically, an adaptive method for determining the number of hidden layer neurons was proposed. This method can automatically adjust the number of hidden layer neurons during the training process of the automatic encoder. First, a method was designed to evaluate the importance of hidden layer neurons. Second, based on this method, the factors that affect the output results were identified. Third, based on the preset maximum and minimum thresholds, if the influence of a certain factor is less than the minimum threshold, it is considered that the corresponding neuron has little impact on the final result and the neuron is discarded. If the influence of a factor is greater than the maximum threshold, the neuron has a significant impact on the output result, and ultimately the neuron divides. The experimental results validate the effectiveness of this algorithm in intelligent cognition.
暂无评论