Deeply integrating latent consistent representations in high-noise multi-omics data for cancer subtyping.

Cancer is a complex and high-mortality disease regulated by multiple factors. Accurate cancer subtyping is crucial for formulating personalized treatment plans and improving patient survival rates. The underlying mechanisms that drive cancer progression can be comprehensively understood by analyzing multi-omics data. However, the high noise levels in omics data often pose challenges in capturing consistent representations and adequately integrating their information. This paper proposed a novel variational autoencoder-based deep learning model, named Deeply Integrating Latent Consistent Representations (DILCR). Firstly, multiple independent variational autoencoders and contrastive loss functions were designed to separate noise from omics data and capture latent consistent representations. Subsequently, an Attention Deep Integration Network was proposed to integrate consistent representations across different omics levels effectively. Additionally, we introduced the Improved Deep Embedded Clustering algorithm to make integrated variable clustering friendly. The effectiveness of DILCR was evaluated using 10 typical cancer datasets from The Cancer Genome Atlas and compared with 14 state-of-the-art integration methods. The results demonstrated that DILCR effectively captures the consistent representations in omics data and outperforms other integration methods in cancer subtyping. In the Kidney Renal Clear Cell Carcinoma case study, cancer subtypes were identified by DILCR with significant biological significance and interpretability.

Efficient screening of adsorbed receptors for Salmonella phage LP31 and identification of receptor-binding protein.

The adsorption process is the first step in the lifecycle of phages and plays a decisive role in the entire infection process. Identifying the adsorption mechanism of phages not only makes phage therapy more precise and efficient but also enables the exploration of other potential applications and modifications of phages. Phage LP31 can lyse multiple Salmonella serotypes, efficiently clearing biofilms formed by Salmonella enterica serovar Enteritidis (S. Enteritidis) and significantly reducing the concentration of S. Enteritidis in chicken feces. Therefore, LP31 has great potential for many practical applications. In this study, we established an efficient screening method for phage infection-related genes and identified a total of 10 genes related to the adsorption process of phage LP31. After the construction of strain C50041ΔrfaL 58-358, it was found that the knockout strain had a rough phenotype as an O-antigen-deficient strain. Adsorption rate and transmission electron microscopy experiments showed that the receptor for phage LP31 was the O9 antigen of S. Enteritidis. Homology comparison and adsorption experiments confirmed that the tail fiber protein Lp35 of phage LP31 participated in the adsorption process as a receptor-binding protein. IMPORTANCE A full understanding of the interaction between phages and their receptors can help with the development of phage-related products. Phages like LP31 with the tail fiber protein Lp35, or a closely related protein, have been reported to effectively recognize and infect multiple Salmonella serotypes. However, the role of these proteins in phage infection has not been previously described. In this study, we established an efficient screening method to detect phage adsorption to host receptors. We found that phage LP31 can utilize its tail fiber protein Lp35 to adsorb to the O9 antigen of S. Enteritidis, initiating the infection process. This study provides a great model system for further studies of how a phage-encoded receptor-binding protein (RBP) interacts with its host's RBP binding target, and this new model offers opportunities for further theoretical and experimental studies to understand the infection mechanism of phages.

Isolation and Characterisation of Bacteriophages with Activity against Invasive Non-Typhoidal Salmonella Causing Bloodstream Infection in Malawi.

In recent years, novel lineages of invasive non-typhoidal Salmonella (iNTS) serovars Typhimurium and Enteritidis have been identified in patients with bloodstream infection in Sub-Saharan Africa. Here, we isolated and characterised 32 phages capable of infecting S. Typhimurium and S. Enteritidis, from water sources in Malawi and the UK. The phages were classified in three major phylogenetic clusters that were geographically distributed. In terms of host range, Cluster 1 phages were able to infect all bacterial hosts tested, whereas Clusters 2 and 3 had a more restricted profile. Cluster 3 contained two sub-clusters, and 3.b contained the most novel isolates. This study represents the first exploration of the potential for phages to target the lineages of Salmonella that are responsible for bloodstream infections in Sub-Saharan Africa.