Navigating the complex landscape of high-dimensional omics data with machine learning models presents a significant challenge. The integration of biological domain knowledge into these models has shown promise in creating more meaningful stratifications of predictor variables, leading to algorithms that are both more accurate and generalizable. However, the wider availability of machine learning tools capable of incorporating such biological knowledge remains limited. Addressing this gap, we introduce BioM2, a novel R package designed for biologically informed multistage machine learning. BioM2 uniquely leverages biological information to effectively stratify and aggregate high-dimensional biological data in the context of machine learning. Demonstrating its utility with genome-wide DNA methylation and transcriptome-wide gene expression data, BioM2 has shown to enhance predictive performance, surpassing traditional machine learning models that operate without the integration of biological knowledge. A key feature of BioM2 is its ability to rank predictor variables within biological categories, specifically Gene Ontology pathways. This functionality not only aids in the interpretability of the results but also enables a subsequent modular network analysis of these variables, shedding light on the intricate systems-level biology underpinning the predictive outcome. We have proposed a biologically informed multistage machine learning framework termed BioM2 for phenotype prediction based on omics data. BioM2 has been incorporated into the BioM2 CRAN package (https://cran.r-project.org/web/packages/BioM2/index.html).

BACKGROUND: Skeletal muscle development plays a crucial role in yield and quality of pork; however, this process is influenced by various factors. In this study, we employed whole-genome bisulfite sequencing (WGBS) and transcriptome sequencing to comprehensively investigate the longissimus dorsi muscle (LDM), aiming to identify key genes that impact the growth and development of Duroc pigs with different average daily gains (ADGs).
RESULTS: Eight pigs were selected and divided into two groups based on ADGs: H (774.89 g) group and L (658.77 g) group. Each pair of the H and L groups were half-siblings. The results of methylation sequencing revealed 2631 differentially methylated genes (DMGs) involved in metabolic processes, signalling, insulin secretion, and other biological activities. Furthermore, a joint analysis was conducted on these DMGs and the differentially expressed genes (DEGs) obtained from transcriptome sequencing of the same individual. This analysis identified 316 differentially methylated and differentially expressed genes (DMEGs), including 18 DMEGs in promoter regions and 294 DMEGs in gene body regions. Finally, LPAR1 and MEF2C were selected as candidate genes associated with muscle development. Bisulfite sequencing PCR (BSP) and quantitative real-time PCR (qRT-PCR) revealed that the promoter region of LPAR1 exhibited significantly lower methylation levels (P < 0.05) and greater expression levels (P < 0.05) in the H group than in the L group. Additionally, hypermethylation was observed in the gene body region of MEF2C, as was a low expression level, in the H group (P < 0.05).
CONCLUSIONS: These results suggest that the differences in the ADGs of Duroc pigs fed the same diet may be influenced by the methylation levels and expression levels of genes related to skeletal muscle development.

The hyperplasia-carcinoma sequence is a stepwise tumourigenic programme towards endometrial cancer in which normal endometrial epithelium becomes neoplastic through non-atypical endometrial hyperplasia (NAEH) and atypical endometrial hyperplasia (AEH), under the influence of unopposed oestrogen. NAEH and AEH are known to exhibit polyclonal and monoclonal cell growth, respectively; yet, aside from focal PTEN protein loss, the genetic and epigenetic alterations that occur during the cellular transition remain largely unknown. We sought to explore the potential molecular mechanisms that promote the NAEH-AEH transition and identify molecular markers that could help to differentiate between these two states. We conducted target-panel sequencing on the coding exons of 596 genes, including 96 endometrial cancer driver genes, and DNA methylome microarrays for 48 NAEH and 44 AEH lesions that were separately collected via macro- or micro-dissection from the endometrial tissues of 30 cases. Sequencing analyses revealed acquisition of the PTEN mutation and the clonal expansion of tumour cells in AEH samples. Further, across the transition, alterations to the DNA methylome were characterised by hypermethylation of promoter/enhancer regions and CpG islands, as well as hypo- and hyper-methylation of DNA-binding regions for transcription factors relevant to endometrial cell differentiation and/or tumourigenesis, including FOXA2, SOX17, and HAND2. The identified DNA methylation signature distinguishing NAEH and AEH lesions was reproducible in a validation cohort with modest discriminative capability. These findings not only support the concept that the transition from NAEH to AEH is an essential step within neoplastic cell transformation of endometrial epithelium but also provide deep insight into the molecular mechanism of the tumourigenic programme. © 2024 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Mdivi-1, Mitochondrial DIVIsion inhibitor 1, has been widely employed in research under the assumption that it exclusively influences mitochondrial fusion, but effects other than mitochondrial dynamics have been underinvestigated. This paper provides transcriptome and DNA methylome-wide analysis for Mdivi-1 treated SH-SY5Y human neuroblastoma cells using RNA sequencing (RNA-seq) and methyl capture sequencing (MC-seq) methods. Gene ontology analysis of RNA sequences revealed that p53 transcriptional gene network and DNA replication initiation-related genes were significantly up and down-regulated, respectively, showing the correlation with the arrest cell cycle in the G1 phase. MC-seq, a powerful sequencing method for capturing DNA methylation status in CpG sites, revealed that although Mdivi-1 does not induce dramatic DNA methylation change, the subtle alterations were concentrated within the CpG island. Integrative analysis of both sequencing data disclosed that the p53 transcriptional network was activated while the Parkinson\'s disease pathway was halted. Next, we investigated several changes in mitochondria in response to Mdivi-1. Copy number and transcription of mitochondrial DNA were suppressed. ROS levels increased, and elevated ROS triggered mitochondrial retrograde signaling rather than inducing direct DNA damage. In this study, we could better understand the molecular network of Mdivi-1 by analyzing DNA methylation and mRNA transcription in the nucleus and further investigating various changes in mitochondria, providing inspiration for studying nuclear-mitochondrial communications.

Mycobacterium marinum, a slow-growing Actinobacterium, typically induces tuberculosis-like disease in fish. Here, we report a new reference sequence for M. marinum ATCC 927T, along with its DNA methylome. This aims to maximize the research potential of this type strain and facilitates investigations into the pathomechanisms of human tuberculosis.

The definitive diagnosis and classification of individual cancers are crucial for patient care and cancer research. To achieve a robust diagnosis of central nervous system (CNS) tumors, a genotype-phenotype integrated diagnostic approach was introduced in recent versions of the World Health Organization classification, followed by the incorporation of a genome-wide DNA methylome-based classification. Microarray-based platforms are widely used to obtain DNA methylome data, and the German Cancer Research Center (Deutsches Krebsforschungszentrum [DKFZ]) has a webtool for a DNA methylation-based classifier (DKFZ classifier). Integration of DNA methylome will further enhance the precision of CNS tumor classification, especially in diagnostically challenging cases. However, in the clinical application of DNA methylome-based classification, challenges related to data interpretation persist, in addition to technical caveats, regulations, and limited accessibility. Dimensionality reduction (DMR) can complement integrated diagnosis by visualizing a profile and comparing it with other known samples. Therefore, DNA methylome-based classification is a highly useful research tool for auxiliary analysis in challenging diagnostic and rare disease cases, and for establishing novel tumor concepts. Decoding the DNA methylome, especially by DMR in addition to DKFZ classifier, emphasizes the capability of grasping the fundamental biological principles that provide new perspectives on CNS tumors.

Increasing evidence shows that an adverse environment during the early fetal development can affect the epigenetic modifications on a wide range of diabetes-related genes, leading to an increased diabetic susceptibility in adulthood or even in subsequent generations. p,p\'-Dichlorodiphenoxydichloroethylene (p,p\'-DDE) is a break-down product of the pesticide dichlorodiphenyltrichloroethane (DDT). p,p\'-DDE has been associated with various health concerns, such as diabetogenic effect. However, the precise molecular mechanism remains unclear. In this study, p,p\'-DDE was given by gavage to pregnant rat dams from gestational day (GD) 8 to GD15 to generate male germline to investiagate the transgenerational effects. We found that early-life p,p\'-DDE exposure increased the transgenerational diabetic susceptibility through male germline inheritance. In utero exposure to p,p\'-DDE altered the sperm DNA methylome in F1 progeny, and a significant number of those differentially methylated genes could be inherited by F2 progeny. Furthermore, early-life p,p\'-DDE exposure altered DNA methylation in glucose metabolic genes Gck and G6pc in sperm and the methylation modification were also found in liver of the next generation. Our study demonstrate that DNA methylation plays a critical role in mediating transgenerational diabetogenic effect induced by early-life p,p\'-DDE exposure.

BACKGROUND: Although mounting evidence supports that aberrant DNA methylation occurs in the hearts of patients with atrial fibrillation (AF), noninvasive epigenetic characterization of AF has not yet been defined.
METHODS: We investigated DNA methylome changes in peripheral blood CD4+ T cells isolated from 10 patients with AF relative to 11 healthy subjects (HS) who were enrolled in the DIANA clinical trial (NCT04371809) via reduced-representation bisulfite sequencing (RRBS).
RESULTS: An atrial-specific PPI network revealed 18 hub differentially methylated genes (DMGs), wherein ROC curve analysis revealed reasonable diagnostic performance of DNA methylation levels found within CDK5R1 (AUC = 0.76; p = 0.049), HSPG2 (AUC = 0.77; p = 0.038), WDFY3 (AUC = 0.78; p = 0.029), USP49 (AUC = 0.76; p = 0.049), GSE1 (AUC = 0.76; p = 0.049), AIFM1 (AUC = 0.76; p = 0.041), CDK5RAP2 (AUC = 0.81; p = 0.017), COL4A1 (AUC = 0.86; p < 0.001), SEPT8 (AUC = 0.90; p < 0.001), PFDN1 (AUC = 0.90; p < 0.01) and ACOT7 (AUC = 0.78; p = 0.032). Transcriptional profiling of the hub DMGs provided a significant overexpression of PSDM6 (p = 0.004), TFRC (p = 0.01), CDK5R1 (p < 0.001), HSPG2 (p = 0.01), WDFY3 (p < 0.001), USP49 (p = 0.004) and GSE1 (p = 0.021) in AF patients vs HS.
CONCLUSIONS: CDK5R1, GSE1, HSPG2 and WDFY3 resulted the best discriminatory genes both at methylation and gene expression level. Our results provide several candidate diagnostic biomarkers with the potential to advance precision medicine in AF.

Convolutional neural networks (CNNs) are becoming increasingly valuable tools for advanced computational histopathology, promoting precision medicine through exceptional visual decoding abilities. Meningiomas, the most prevalent primary intracranial tumors, necessitate accurate grading and classification for informed clinical decision-making. Recently, DNA methylation-based molecular classification of meningiomas has proven to be more effective in predicting tumor recurrence than traditional histopathological methods. However, DNA methylation profiling is expensive, labor-intensive, and not widely accessible. Consequently, a digital histology-based prediction of DNA methylation classes would be advantageous, complementing molecular classification. In this study, we developed and rigorously assessed an attention-based multiple-instance deep neural network for predicting meningioma methylation classes using tumor methylome data from 142 (+51) patients and corresponding hematoxylin-eosin-stained histological sections. Pairwise analysis of sample cohorts from three meningioma methylation classes demonstrated high accuracy in two combinations. The performance of our approach was validated using an independent set of 51 meningioma patient samples. Importantly, attention map visualization revealed that the algorithm primarily focuses on tumor regions deemed significant by neuropathologists, offering insights into the decision-making process of the CNN. Our findings highlight the capacity of CNNs to effectively harness phenotypic information from histological sections through computerized images for precision medicine. Notably, this study is the first demonstration of predicting clinically relevant DNA methylome information using computer vision applied to standard histopathology. The introduced AI framework holds great potential in supporting, augmenting, and expediting meningioma classification in the future.

Increasing evidence indicates that PM2.5 exposure disrupts early embryonic development, but the mechanisms remain unclear. We hypothesized that PM2.5 cause abnormal embryonic development by interfering with DNA methylation and mRNA expression. In this study, we observed that human embryonic stem cells (hESCs) treated with extractable organic matters (EOM) from PM2.5 concentrations above 100 μg/mL exhibited reduced viability. While EOM within non-cytotoxicity concentrations did not affect the expression levels of pluripotency genes, it did enhance cellular proliferation, as indicated by increased Edu incorporation and the upregulation of cell cycle genes (Cdk2, Mdm2). Additionally, EOM significantly influenced the transcriptome patterns in hESCs. Notably, the differentially expressed genes were found to be significantly enriched in processes such as extracellular matrix organization, cell-cell junction organization, chromatin organization, and DNA methylation. Furthermore, we observed whole genomic-wide DNA methylation changes. Through a cross-analysis of changes in DNA methylation and mRNA expression, we identified an enrichment of terms related to the VEGFR signaling pathway and extracellular matrix. The gene signal transduction networks revealed that crucial hubs were implicated in cell growth and division. In conclusion, our findings demonstrate that PM2.5 induce significant alterations in transcriptome and DNA methylome in hESCs, leading to aberrant cell proliferation. This research provides novel insights into the molecular mechanisms underlying the developmental toxicity of PM2.5.