Virulencefactors (VFs) are key molecules that enable pathogens to evade the immune systems of the host. These factors are crucial for revealing the pathogenic processes of microbes and drug discovery. Identification of virulence factors in microbes become an important problem in the field of bioinformatics. To address this problem, this study proposes a novel model DTVF (Deep Transfer Learning for Virulence Factor Prediction), which integrates the ProtT5 protein sequence extraction model with a dual-channel deep learning model. In the dual-channel deep learning model, we innovatively integrate long short-term memory (LSTM) with convolutional neural networks (CNNs), creating a novel integrated architecture. Furthermore, by incorporating the attention mechanism, the accuracy of VF detection was significantly enhanced. We evaluated the DTVF model against other excellent-performing models in the field. DTVF demonstrates superior performance, achieving an accuracy rate of 84.55% and an AUROC of 92.08% on the benchmark dataset. DTVF shows state-of-the-art performance in this field, surpassing the existing models in nearly all metrics. To facilitate the use of biologists, we have also developed an interactive web-based user interface version of DTVF based on Gradio.

Epilepsy is a neurological disorder characterized by recurrent, unprovoked seizures. Currently, the associations among skin microbiota, circulating metabolites, and epilepsy are still not well studied. In this study, we applied univariate and two-step Mendelian randomization analysis using single nucleotide polymorphisms as instrumental variables to analyze the possible associations. Five skin microbiota (asv022, asv008, bacillales, clostridiale, and micrococcaceae) and four circulating metabolites were found to be associated with epilepsy. Our results also showed that leucine level (p-value = 0.0487, OR = 1.067) and glutamine level (p-value = 0.0372, OR = 1.065) show a slight increased association with epilepsy, while phospholipids in medium LDL (p-value = 0.0302, OR = 0.9308) and ratio of saturated fatty acids (p-value = 0.0309, OR = 0.9017) appear to show a slight inverse correlation with epilepsy. In addition, the heterogeneity test and horizontal pleiotropy analysis indicated these results are robust in different situations. By mapping analysis, 54 genes were associated with at least two metabolites. By functional analysis, these genes were enriched in the immune-related pathways, which may have systemic effects on brain function. Our results provide a novel insight for understanding possible mechanisms of some epilepsy associations, which by further study might provide avenues for therapy research.