The burgeoning accumulation of large-scale biomedical data in oncology, alongside significant strides in deep learning (DL) technologies, has established multimodal DL (MDL) as a cornerstone of precision oncology. This review provides an overview of MDL applications in this field, based on an extensive literature survey. In total, 651 articles published before September 2024 are included. We first outline publicly available multimodal datasets that support cancer research. Then, we discuss key DL training methods, data representation techniques, and fusion strategies for integrating multimodal data. The review also examines MDL applications in tumor segmentation, detection, diagnosis, prognosis, treatment selection, and therapy response monitoring. Finally, we critically assess the limitations of current approaches and propose directions for future research. By synthesizing current progress and identifying challenges, this review aims to guide future efforts in leveraging MDL to advance precision oncology.

Microbes play important roles in human health and disease. The interaction between microbes and hosts is a reciprocal relationship, which remains largely under-explored. Current computational resources lack manually and consistently curated data to connect metagenomic data to pathogenic microbes, microbial core genes, and disease phenotypes. We developed the MicroPhenoDB database by manually curating and consistently integrating microbe-disease association data. MicroPhenoDB provides 5677 non-redundant associations between 1781 microbes and 542 human disease phenotypes across more than 22 human body sites. MicroPhenoDB also provides 696,934 relationships between 27,277 unique clade-specific core genes and 685 microbes. Disease phenotypes are classified and described using the Experimental Factor Ontology (EFO). A refined score model was developed to prioritize the associations based on evidential metrics. The sequence search option in MicroPhenoDB enables rapid identification of existing pathogenic microbes in samples without running the usual metagenomic data processing and assembly. MicroPhenoDB offers data browsing, searching, and visualization through user-friendly web interfaces and web service application programming interfaces. MicroPhenoDB is the first database platform to detail the relationships between pathogenic microbes, core genes, and disease phenotypes. It will accelerate metagenomic data analysis and assist studies in decoding microbes related to human diseases. MicroPhenoDB is available through http://www.liwzlab.cn/microphenodb and http://lilab2.sysu.edu.cn/microphenodb.

Noncoding RNAs (ncRNAs) play critical roles in many critical biological processes and have become a novel class of potential targets and bio-markers for disease diagnosis, therapy, and prognosis. Annotating and analyzing ncRNA-disease association data are essential but challenging. Current computational resources lack comprehensive database platforms to consistently interpret and prioritize ncRNA-disease association data for biomedical investigation and application. Here, we present the ncRPheno database platform (http://lilab2.sysu.edu.cn/ncrpheno), which comprehensively integrates and annotates ncRNA-disease association data and provides novel searches, visualizations, and utilities for association identification and validation. ncRPheno contains 482,751 non-redundant associations between 14,494 ncRNAs and 3,210 disease phenotypes across 11 species with supporting evidence in the literature. A scoring model was refined to prioritize the associations based on evidential metrics. Moreover, ncRPheno provides user-friendly web interfaces, novel visualizations, and programmatic access to enable easy exploration, analysis, and utilization of the association data. A case study through ncRPheno demonstrated a comprehensive landscape of ncRNAs dysregulation associated with 22 cancers and uncovered 821 cancer-associated common ncRNAs. As a unique database platform, ncRPheno outperforms the existing similar databases in terms of data coverage and utilities, and it will assist studies in encoding ncRNAs associated with phenotypes ranging from genetic disorders to complex diseases.

Background: It is now understood that the effectiveness of checkpoint immunotherapy can be impaired by immunosuppressive tumor-associated macrophages (TAMs). Nonetheless, the impact of different TAM subpopulations on the antitumor immune response remains unclear, mainly due to their heterogeneity. Herein, we identified a novel TAM subpopulation in esophageal squamous cell carcinoma (ESCC) that might contribute to poor clinical outcomes and immunotherapy modulation.Methods and results: We analyzed two single-cell RNA sequencing (scRNA-seq) datasets (GSE145370 and GSE160269) of esophageal squamous cell carcinoma to identify a novel TREM2-positive TAM subpopulation characterized by upregulation of TREM2, C1QC, C1QB, C1QA, SPP1, and APOE. Quantitative real-time PCR (qRT-PCR) and enzyme-linked immunosorbent assay (ELISA) demonstrated that these genes were significantly overexpressed in ESCC. Multiplex immunofluorescence validated the infiltration of TREM2⁺ TAMs in ESCC tissues, which correlated with poorer overall survival (OS). The scRNA-seq analysis in dataset GSE120575 indicated significant enrichment of TREM2⁺ TAMs in melanoma patients (n=48) with poor immunotherapy response, which had an identical gene signature with TREM2⁺ TAMs from ESCC. Analysis of 29 bulk-RNA melanoma samples from dataset GSE78220 revealed that a gene signature of 40 genes associated with TREM2⁺ TAMs was upregulated in the transcriptome of melanomas that did not respond to anti-PD1 therapy. Validation in the TCGA ESCC cohort (n=80) showed that a high enrichment score of the TREM2⁺ TAM was associated with poor prognosis. In addition, 10 ESCC patients treated with anti-PD1 therapy suggested that patients who are not sensitive to immunotherapy have higher density of TREM2+TAMs infiltration.Conclusion: Overall, TREM2⁺ TAM infiltration in ESCC is associated with poor prognosis and may serve as a biomarker for predicting outcomes and immunotherapy modulation in this patient population. modulation; single-cell RNA sequencing

Microbes play important roles in human health and disease. The interaction between microbes and hosts is a reciprocal relationship, which remains largely under-explored. Current computational resources lack manually and consistently curated data to connect metagenomic data to pathogenic microbes, microbial core genes, and disease phenotypes. We developed the MicroPhenoDB database by manually curating and consistently integrating microbe-disease association data. MicroPhenoDB provides 5677 non-redundant associations between 1781 microbes and 542 human disease phenotypes across more than 22 human body sites. MicroPhenoDB also provides 696,934 relationships between 27,277 unique clade-specific core genes and 685 microbes. Disease phenotypes are classified and described using the Experimental Factor Ontology (EFO). A refined score model was developed to prioritize the associations based on evidential metrics. The sequence search option in MicroPhenoDB enables rapid identification of existing pathogenic microbes in samples without running the usual metagenomic data processing and assembly. MicroPhenoDB offers data browsing, searching and visualization through user-friendly web interfaces and web service application programming interfaces. MicroPhenoDB is the first database platform to detail the relationships between pathogenic microbes, core genes, and disease phenotypes. It will accelerate metagenomic data analysis and assist studies in decoding microbes related to human diseases. MicroPhenoDB is available through http://www.liwzlab.cn/microphenodb and http://lilab2.sysu.edu.cn/microphenodb .

Background: Immune checkpoint blockade (ICB)-based immunotherapy has significantly improved survival in advanced melanoma. However, many patients exhibit resistance to these therapies. This study examines the impact of BTLA promoter methylation on its expression, immune cell infiltration, and clinical outcomes, evaluating its potential as a prognostic and predictive biomarker for immunotherapy response. Methods: We analyzed methylation and gene expression data from public datasets (The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO)) and an in-house cohort of melanoma patients treated with ICB therapy at the First Affiliated Hospital of Zhengzhou University. We developed a quantitative methylation-specific PCR (qMSP) assay to measure methylation levels of the cg24157392 and cg03995631 CpG sites, and a targeted bisulfite sequencing assay was used to validate the accuracy of qMSP. We measured BTLA protein expression using multiplex immunofluorescence and immunohistochemical staining methods. Pearson correlation, survival analysis, and immune cell infiltration estimation were conducted to explore the associations between BTLA promoter methylation, mRNA and protein expression, clinical outcomes, and immune characteristics. Results: Hypomethylation at CpG sites cg24157392 and cg03995631 in the BTLA promoter were significantly associated with higher BTLA mRNA and protein expression. In the TCGA dataset, low methylation at these sites predicted longer overall survival and was validated in an independent cohort of 50 stage III/IV melanoma patients, with an area under the curve of 0.94 for predicting 5-year survival. Furthermore, BTLA promoter hypomethylation correlated with higher infiltration of immune cells, such as CD8+T cells, CD4+T cells, B cells, and macrophages. Additionally, low methylation at cg24157392 and cg03995631, as quantified by the qMSP assay, was significantly associated with better progression-free survival in patients treated with immune checkpoint inhibitors. These findings were further validated using GEO datasets. Conclusions: BTLA promoter hypomethylation serves as a significant biomarker for favorable prognosis and enhanced response to ICB therapy in melanoma. The developed qMSP assays for cg24157392 and cg03995631 accurately quantified methylation levels and demonstrated their potential for clinical application in patient stratification and personalized immunotherapy.