Integrative Analysis Of Multi-omics Data Research Articles

Integrated analysis of multi-omics data for the discovery of biomarkers and therapeutic targets for juvenile idiopathic arthritis

BackgroundJuvenile idiopathic arthritis (JIA) is a prevalent chronic rheumatic disease affecting children. Current medications merely alleviate symptoms rather than curing the disease. Hence, the identification and development of novel drug targets and biomarkers for JIA are imperative for enhancing treatment efficacy. MethodsWe employed two-sample Mendelian randomization (MR) analysis to investigate the causal effects of plasma proteins on JIA. Additionally, colocalization, bulk RNA-seq, and single-cell RNA-seq analyses were conducted to further investigate and validate the potential of candidate proteins as drug targets. ResultsThrough MR analysis, we successfully identified five plasma proteins that are causally linked to JIA. Genetically inferred lower levels of AIF1, TNF, and TNFSF11 were associated with an elevated risk of JIA, while higher levels of AGER and GP1BA proteins were positively correlated with JIA risk. Colocalization analysis further supported our findings on GP1BA (OR = 9.26, 95% CI: 2.30-37.20) and TNFSF11 (OR = 0.18, 95% CI: 0.07-0.45). Based on this evidence, we classified these five proteins into two tiers. Finally, we conducted a systematic evaluation of the druggability and current drug development progress for these identified candidate proteins. ConclusionsThis study employed MR analysis to reveal causal relationships between plasma proteins and JIA, identifying five potential candidate proteins as promising drug targets for JIA, particularly focusing on GP1BA and TNFSF11.

Integrative analysis of multi-omics data reveals importance of collagen and the PI3K AKT signalling pathway in CAKUT

Congenital Anomalies of the Kidney and Urinary Tract (CAKUT) is the leading cause of childhood chronic kidney failure and a significant cause of chronic kidney disease in adults. Genetic and environmental factors are known to influence CAKUT development, but the currently known disease mechanism remains incomplete. Our goal is to identify affected pathways and networks in CAKUT, and thereby aid in getting a better understanding of its pathophysiology. With this goal, the miRNome, peptidome, and proteome of over 30 amniotic fluid samples of patients with non-severe CAKUT was compared to patients with severe CAKUT. These omics data sets were made findable, accessible, interoperable, and reusable (FAIR) to facilitate their integration with external data resources. Furthermore, we analysed and integrated the omics data sets using three different bioinformatics strategies: integrative analysis with mixOmics, joint dimensionality reduction and pathway analysis. The three bioinformatics analyses provided complementary features, but all pointed towards an important role for collagen in CAKUT development and the PI3K-AKT signalling pathway. Additionally, several key genes (CSF1, IGF2, ITGB1, and RAC1) and microRNAs were identified. We published the three analysis strategies as containerized workflows. These workflows can be applied to other FAIR data sets and help gaining knowledge on other rare diseases.

Abstract P215: The reninness score: integrative analysis of multi-omic data to define renin cell identity.

Renin cell (RC) descendants regain the endocrine renin phenotype to overcome homeostatic threats, a process known as recruitment. The determinants that control the identity, fate, and recruitment of RCs are not well understood. We aim to 1) define the chromatin pattern that determines RC identity; and 2) develop a computational tool to score samples by similarity to that pattern. We compared open chromatin regions (ATAC-seq) between non-renin-expressing (n = 184) and renin-expressing samples (n = 19) from three different sources and stimulation states: 1) primary RCs in basal state (WT); 2) primary RCs chronically stimulated to produce renin (Recruited); and 3) constitutively renin-expressing tumoral cells. Differential analyses between the three types of RCs and the non-RCs revealed 1,525 unique open chromatin regions shared in at least two RC groups, including regions associated with RC-specific genes: Ren1 and Akr1b7. Increased accessibility regions in RC groups are cell type specific and enriched in renin-related GO terms ( e.g., cAMP and Notch signaling pathway). The bZip family ( e.g. , Atf3, p < 10 -2147 ) was the most enriched motifs in tumoral cells, and MEF2 family ( e.g. , MEF2a, p < 10 -91 ) in primary cells. We did not find a unique set of TFs specific to the Recruited group, indicating that renin recruitment may not be regulated by a recruitment-specific factor family, but could instead be driven by post-translational modification, differential co-factors, or epigenetic modifications. Next, we used a machine-learning approach to calculate a "reninness" score by training a genomic region model with renin and non-renin bulkATAC-seq data. We identified a cluster of regions (n = 112,992) around the trained renin label, which included 89% (21,820 of 24,612) of the regions with increased accessibility from the differential analyses. We calculated the reninness score based on the distance between the trained label and the predicted bulkATA sample embeddings using the trained model. Homogeneous WT RCs had the highest score, followed by renin-expressing tumoral and primary cells. Non-RCs had the lowest scores. We also adapted bulkATAC region-embedding models to pseudo-bulk ATAC- and scATAC-seq data. The observed reninness scores show that the model can also capture differences in renin recruitment potential. In conclusion, we identified the chromatin configurations defining the identity and recruitment of RCs and developed a computational score to quantify it.

Maternal immune cell gene expression associates with maternal gut microbiome, milk composition and infant gut microbiome

Background– Pre-pregnancy overweight and obesity promote deleterious health impacts on both mothers during pregnancy and the offspring. Significant changes in the maternal peripheral blood mononuclear cells (PBMCs) gene expression due to obesity are well-known. However, the impact of pre-pregnancy overweight on immune cell gene expression during pregnancy and its association with maternal and infant outcomes is not well explored. Methods– Blood samples were collected from healthy normal weight (NW, pre-pregnancy BMI 18.5-24.9) or overweight (OW, pre-pregnancy BMI 25-29.9) 2nd parity pregnant women at 12, 24 and 36 weeks of pregnancy. PBMCs were isolated from the blood and subjected to mRNA sequencing. Maternal and infant microbiota were analyzed by 16S rRNA gene sequencing. Integrative multi-omics data analysis was performed to evaluate the association of gene expression with maternal diet, gut microbiota, milk composition, and infant gut microbiota. ResultsGene expression analysis revealed that 453 genes were differentially expressed in the OW women compared to NW women at 12 weeks of pregnancy, out of which 354 were upregulated and 99 were downregulated. Several up-regulated genes in the OW group were enriched in inflammatory, chemokine-mediated signaling and regulation of interleukin-8 production-related pathways. At 36 weeks of pregnancy healthy eating index score was positively associated with several genes that include, DTD1, ELOC, GALNT8, ITGA6-AS1, KRT17P2, NPW, POT1-AS1 and RPL26. In addition, at 36 weeks of pregnancy, genes involved in adipocyte functions, such as NG2 and SMTNL1, were negatively correlated to human milk 2’FL and total fucosylated oligosaccharides content collected at 1 month postnatally. Furthermore, infant Akkermansia was positively associated with maternal PBMC anti-inflammatory genes that include CPS1 and RAB7B, at 12 and 36 weeks of pregnancy. Conclusions– These findings suggest that prepregnancy overweight impacts the immune cell gene expression profile, particularly at 12 weeks of pregnancy. Furthermore, deciphering the complex association of PBMC’s gene expression levels with maternal gut microbiome and milk composition and infant gut microbiome may aid in developing strategies to mitigate obesity-mediated effects.

MosGraphGen: a novel tool to generate multi-omics signaling graphs to facilitate integrative and interpretable graph AI model development.

Multi-omics data, i.e., genomics, epigenomics, transcriptomics, proteomics, characterize cellular complex signaling systems from multi-level and multi-view and provide a holistic view of complex cellular signaling pathways. However, it remains challenging to integrate and interpret multi-omics data for mining key disease targets and signaling pathways. Graph AI models have been widely used to analyze graph-structure datasets, and are ideal for integrative multi-omics data analysis because they can naturally integrate and represent multi-omics data as a biologically meaningful multi-level signaling graph and interpret multi-omics data via graph node and edge ranking analysis. However, it is non-trivial for graph-AI model developers to pre-analyze multi-omics data and convert the data into biologically meaningful graphs, which can be directly fed into graph-AI models. To resolve this challenge, we developed mosGraphGen (multi-omics signaling graph generator), generating Multi-omics Signaling graphs (mos-graph) of individual samples by mapping multi-omics data onto a biologically meaningful multi-level background signaling network with data normalization by aggregating measurements and aligning to the reference genome. With mosGraphGen, AI model developers can directly apply and evaluate their models using these mos-graphs. In the results, mosGraphGen was used and illustrated using two widely used multi-omics datasets of TCGA and Alzheimer's disease (AD) samples. The code of mosGraphGen is open-source and publicly available via GitHub: https://github.com/FuhaiLiAiLab/mosGraphGen.

Integrative Analysis of Multi-Omics Data to Identify Deregulated Molecular Pathways and Druggable Targets in Chronic Lymphocytic Leukemia.

Chronic Lymphocytic Leukemia (CLL) is the most common B-cell malignancy in the Western world, characterized by frequent relapses despite temporary remissions. Our study integrated publicly available proteomic, transcriptomic, and patient survival datasets to identify key differences between healthy and CLL samples. We exposed approximately 1000 proteins that differentiate healthy from cancerous cells, with 608 upregulated and 415 downregulated in CLL cases. Notable upregulated proteins include YEATS2 (an epigenetic regulator), PIGR (Polymeric immunoglobulin receptor), and SNRPA (a splicing factor), which may serve as prognostic biomarkers for this disease. Key pathways implicated in CLL progression involve RNA processing, stress resistance, and immune response deficits. Furthermore, we identified three existing drugs-Bosutinib, Vorinostat, and Panobinostat-for potential further investigation in drug repurposing in CLL. We also found limited correlation between transcriptomic and proteomic data, emphasizing the importance of proteomics in understanding gene expression regulation mechanisms. This generally known disparity highlights once again that mRNA levels do not accurately predict protein abundance due to many regulatory factors, such as protein degradation, post-transcriptional modifications, and differing rates of translation. These results demonstrate the value of integrating omics data to uncover deregulated proteins and pathways in cancer and suggest new therapeutic avenues for CLL.

NetMIM: network-based multi-omics integration with block missingness for biomarker selection and disease outcome prediction.

Compared with analyzing omics data from a single platform, an integrative analysis of multi-omics data provides a more comprehensive understanding of the regulatory relationships among biological features associated with complex diseases. However, most existing frameworks for integrative analysis overlook two crucial aspects of multi-omics data. Firstly, they neglect the known dependencies among biological features that exist in highly credible biological databases. Secondly, most existing integrative frameworks just simply remove the subjects without full omics data to handle block missingness, resulting in decreasing statistical power. To overcome these issues, we propose a network-based integrative Bayesian framework for biomarker selection and disease outcome prediction based on multi-omics data. Our framework utilizes Dirac spike-and-slab variable selection prior to identifying a small subset of biomarkers. The incorporation of gene pathway information improves the interpretability of feature selection. Furthermore, with the strategy in the FBM (stand for "full Bayesian model with missingness") model where missing omics data are augmented via a mechanistic model, our framework handles block missingness in multi-omics data via a data augmentation approach. The real application illustrates that our approach, which incorporates existing gene pathway information and includes subjects without DNA methylation data, results in more interpretable feature selection results and more accurate predictions.

Proteogenomic insights into the biology and treatment of pan-melanoma

Melanoma is one of the most prevalent skin cancers, with high metastatic rates and poor prognosis. Understanding its molecular pathogenesis is crucial for improving its diagnosis and treatment. Integrated analysis of multi-omics data from 207 treatment-naïve melanomas (primary-cutaneous-melanomas (CM, n = 28), primary-acral-melanomas (AM, n = 81), primary-mucosal-melanomas (MM, n = 28), metastatic-melanomas (n = 27), and nevi (n = 43)) provides insights into melanoma biology. Multivariate analysis reveals that PRKDC amplification is a prognostic molecule for melanomas. Further proteogenomic analysis combined with functional experiments reveals that the cis-effect of PRKDC amplification may lead to tumor proliferation through the activation of DNA repair and folate metabolism pathways. Proteome-based stratification of primary melanomas defines three prognosis-related subtypes, namely, the ECM subtype, angiogenesis subtype (with a high metastasis rate), and cell proliferation subtype, which provides an essential framework for the utilization of specific targeted therapies for particular melanoma subtypes. The immune classification identifies three immune subtypes. Further analysis combined with an independent anti-PD-1 treatment cohort reveals that upregulation of the MAPK7-NFKB signaling pathway may facilitate T-cell recruitment and increase the sensitivity of patients to immunotherapy. In contrast, PRKDC may reduce the sensitivity of melanoma patients to immunotherapy by promoting DNA repair in melanoma cells. These results emphasize the clinical value of multi-omics data and have the potential to improve the understanding of melanoma treatment.

Network-based integrative multi-omics approach reveals biosignatures specific to COVID-19 disease phases.

COVID-19 disease is characterized by a spectrum of disease phases (mild, moderate, and severe). Each disease phase is marked by changes in omics profiles with corresponding changes in the expression of features (biosignatures). However, integrative analysis of multiple omics data from different experiments across studies to investigate biosignatures at various disease phases is limited. Exploring an integrative multi-omics profile analysis through a network approach could be used to determine biosignatures associated with specific disease phases and enable the examination of the relationships between the biosignatures. To identify and characterize biosignatures underlying various COVID-19 disease phases in an integrative multi-omics data analysis. We leveraged a multi-omics network-based approach to integrate transcriptomics, metabolomics, proteomics, and lipidomics data. The World Health Organization Ordinal Scale WHO Ordinal Scale was used as a disease severity reference to harmonize COVID-19 patient metadata across two studies with independent data. A unified COVID-19 knowledge graph was constructed by assembling a disease-specific interactome from the literature and databases. Disease-state specific omics-graphs were constructed by integrating multi-omics data with the unified COVID-19 knowledge graph. We expanded on the network layers of multiXrank, a random walk with restart on multilayer network algorithm, to explore disease state omics-specific graphs and perform enrichment analysis. Network analysis revealed the biosignatures involved in inducing chemokines and inflammatory responses as hubs in the severe and moderate disease phases. We observed distinct biosignatures between severe and moderate disease phases as compared to mild-moderate and mild-severe disease phases. Mild COVID-19 cases were characterized by a unique biosignature comprising C-C Motif Chemokine Ligand 4 (CCL4), and Interferon Regulatory Factor 1 (IRF1). Hepatocyte Growth Factor (HGF), Matrix Metallopeptidase 12 (MMP12), Interleukin 10 (IL10), Nuclear Factor Kappa B Subunit 1 (NFKB1), and suberoylcarnitine form hubs in the omics network that characterizes the moderate disease state. The severe cases were marked by biosignatures such as Signal Transducer and Activator of Transcription 1 (STAT1), Superoxide Dismutase 2 (SOD2), HGF, taurine, lysophosphatidylcholine, diacylglycerol, triglycerides, and sphingomyelin that characterize the disease state. This study identified both biosignatures of different omics types enriched in disease-related pathways and their associated interactions (such as protein-protein, protein-transcript, protein-metabolite, transcript-metabolite, and lipid-lipid interactions) that are unique to mild, moderate, and severe COVID-19 disease states. These biosignatures include molecular features that underlie the observed clinical heterogeneity of COVID-19 and emphasize the need for disease-phase-specific treatment strategies. The approach implemented here can be used to find associations between transcripts, proteins, lipids, and metabolites in other diseases.

Predicting regulatory mutations and their target genes by new computational integrative analysis: A study of follicular lymphoma

Mutations in DNA regulatory regions are increasingly being recognized as important drivers of cancer and other complex diseases. These mutations can regulate gene expression by affecting DNA-protein binding and epigenetic profiles, such as DNA methylation in genome regulatory elements. However, identifying mutation hotspots associated with expression regulation and disease progression in non-coding DNA remains a challenge. Unlike most existing approaches that assign a mutation score to individual single nucleotide polymorphisms (SNP), a mutation block (MB)-based approach was introduced in this study to assess the collective impact of a cluster of SNPs on transcription factor-DNA binding affinity, differential gene expression (DEG), and nearby DNA methylation. Moreover, the long-distance target genes of functional MBs were identified using a new permutation-based algorithm that assessed the significance of correlations between DNA methylation at regulatory regions and target gene expression. Two new Python packages were developed. The Differential Methylation Region (DMR-analysis) analysis tool was used to detect DMR and map them to regulatory elements. The second tool, an integrated DMR, DEG, and SNP analysis tool (DDS-analysis), was used to combine the omics data to identify functional MBs and long-distance target genes. Both tools were validated in follicular lymphoma (FL) cohorts, where not only known functional MBs and their target genes (BCL2 and BCL6) were recovered, but also novel genes were found, including CDCA4 and JAG2, which may be associated with FL development. These genes are linked to target gene expression and are significantly correlated with the methylation of nearby DNA sequences in FL. The proposed computational integrative analysis of multiomics data holds promise for identifying regulatory mutations in cancer and other complex diseases.

Integrative analyses of multi-omics data constructing tumor microenvironment and immune-related molecular prognosis model in human colorectal cancer

The increasing prevalence and incidence of colorectal cancer (CRC), particularly in young adults, underscore the imperative to comprehend its fundamental mechanisms, discover novel diagnostic and prognostic markers, and enhance therapeutic strategies. Here, we integrated multi-omics data, including gene expression, somatic mutation data and DNA methylation data, to unravel the intricacies of tumor microenvironment (TME) in CRC and search for novel prognostic markers. By calculating the immune score for each patient from the expression profile, we delineated the differential immune cell fraction, constructed an immune-related multi-omics atlas, and identified molecular characteristics. The entire colorectal dataset (n = 343) was randomly divided into training (n = 249) and testing datasets (n = 94). We screened 144 immune-related genes, 6 mutant genes, and 38 methylation probes associated with overall survival (OS). These makers were then incorporated into a 10-gene prognostic model using Lasso and Cox regression in the training dataset, and the model's performance was evaluated in an independent validation dataset. The model exhibited satisfactory results (average concordance index [C-index] = 0.77), with the average 1-year, 3-year, and 5-year AUCs being 0.79, 0.76, and 0.76 in the training dataset and 0.74, 0.80, and 0.90 in the testing dataset. Furthermore, the prognostic model demonstrated applicability in guiding chemotherapy for CRC patients and exhibited a degree of pan-cancer utility in risk stratification. In conclusion, our integrated analysis of multi-omics data revealed immune-related genetic and epigenetic characteristics of the TME. We propose an integrative prognostic model that can stratify risk and guide chemotherapy for CRC patients. The generalizability of the model in risk stratification across different cancer types was validated in Pan-Cancer cohort.

Mechanistic safety assessment via multi-omic characterisation of systemic pathway perturbations following in vivo MAT2A inhibition

The tumour suppressor p16/CDKN2A and the metabolic gene, methyl-thio-adenosine phosphorylase (MTAP), are frequently co-deleted in some of the most aggressive and currently untreatable cancers. Cells with MTAP deletion are vulnerable to inhibition of the metabolic enzyme, methionine-adenosyl transferase 2A (MAT2A), and the protein arginine methyl transferase (PRMT5). This synthetic lethality has paved the way for the rapid development of drugs targeting the MAT2A/PRMT5 axis. MAT2A and its liver- and pancreas-specific isoform, MAT1A, generate the universal methyl donor S-adenosylmethionine (SAM) from ATP and methionine. Given the pleiotropic role SAM plays in methylation of diverse substrates, characterising the extent of SAM depletion and downstream perturbations following MAT2A/MAT1A inhibition (MATi) is critical for safety assessment. We have assessed in vivo target engagement and the resultant systemic phenotype using multi-omic tools to characterise response to a MAT2A inhibitor (AZ’9567). We observed significant SAM depletion and extensive methionine accumulation in the plasma, liver, brain and heart of treated rats, providing the first assessment of both global SAM depletion and evidence of hepatic MAT1A target engagement. An integrative analysis of multi-omic data from liver tissue identified broad perturbations in pathways covering one-carbon metabolism, trans-sulfuration and lipid metabolism. We infer that these pathway-wide perturbations represent adaptive responses to SAM depletion and confer a risk of oxidative stress, hepatic steatosis and an associated disturbance in plasma and cellular lipid homeostasis. The alterations also explain the dramatic increase in plasma and tissue methionine, which could be used as a safety and PD biomarker going forward to the clinic.

Knowledge-guided learning methods for integrative analysis of multi-omics data

Integrative analysis of multi-omics data has the potential to yield valuable and comprehensive insights into the molecular mechanisms underlying complex diseases such as cancer and Alzheimer's disease. However, a number of analytical challenges complicate multi-omics data integration. For instance, -omics data are usually high-dimensional, and sample sizes in multi-omics studies tend to be modest. Furthermore, when genes in an important pathway have relatively weak signal, it can be difficult to detect them individually. There is a growing body of literature on knowledge-guided learning methods that can address these challenges by incorporating biological knowledge such as functional genomics and functional proteomics into multi-omics data analysis. These methods have been shown to outperform their counterparts that do not utilize biological knowledge in tasks including prediction, feature selection, clustering, and dimension reduction. In this review, we survey recently developed methods and applications of knowledge-guided multi-omics data integration methods and discuss future research directions.

Integrated Analysis of Multi-Omic Data Reveals Regulatory Mechanisms and Network Characteristics in Breast Cancer.

Breast cancer is a complex and heterogeneous disease, and understanding its regulatory mechanisms and network characteristics is essential for identifying therapeutic targets and developing effective treatment strategies. This study aimed to unravel the intricate network of interactions involving differentially expressed genes, microribonucleic acid (miRNAs), and proteins in breast cancer through an integrative analysis of multi-omic data from Cancer Genome Atlas Breast Invasive Carcinoma (TCGA-BRCA) dataset. The TCGA-BRCA dataset was used for data acquisition, which included RNA sequencing data for gene expression, miRNA sequencing data for miRNA expression, and protein expression quantification data. Various R packages, such as TCGAbiolinks, limma, and RPPA, were employed for data preprocessing and integration. Differential expression analysis, network construction, miRNA regulation exploration, pathway enrichment analysis, and independent dataset validation were performed. Eight consistently upregulated hub genes-including ACTB, HSP90AA1, FN1, HSPA8, CDC42, CDH1, UBC, and EP300-were identified in breast cancer, indicating their potential significance in driving the disease. Pathway enrichment analysis revealed highly enriched pathways in breast cancer, including proteoglycans in cancer, PI3K-Akt, and mitogen-activated protein kinase signaling. This integrated multi-omic data analysis provides valuable insights into the regulatory mechanisms, network characteristics, and functional roles of genes, miRNAs, and proteins in breast cancer. The findings contribute to our understanding of the molecular landscape of breast cancer, facilitate the identification of potential therapeutic targets, and inform strategies for effective treatment.

Abstract 7355: Integration analysis of lncRNAs and R-loops in 3D genome organization and gene regulation in the cancer genome

Abstract Existing evidence shows that some long non-coding RNAs (lncRNAs) can recruit chromatin regulators and form R-loops to mediate CCTCC-binding factor (CTCF) binding at topologically associated domain (TAD) boundaries to activate transcription of target genes. However, it remains unclear to what extent such lncRNA-dependent chromatin organization events occur genome-wide in normal and cancer cells. By curated public data of lncRNA, CTCF, RAD21 profiles and integrative analysis of multi-omics data using innovative computational approaches, we investigated global associations of lncRNAs and R-loops with activities of transcription factors and chromatin regulators in gene expression regulations across multiple human cell types. We developed a comprehensive computational pipeline for cross-cell-type analysis of lncRNA and protein-coding gene expression profiles and transcription factor binding profiles. We identified a total of 542 candidate genomic regions where lncRNAs play a pivotal role in regulating target genes by facilitating CTCF/cohesin binding. R-loops are enriched at CTCF/cohesin binding sites, suggesting their potential contribution to CTCF binding. Additionally, our findings underscore the significant cell-type specificity of CTCF binding patterns across the genome. This work adds knowledge to cancer genome organization and gene regulation, and builds foundations for further functional and mechanistic studies of lncRNAs and R-loops in the human genome. Citation Format: Shengyuan Wang, Zhenjia Wang, Yuh-Hwa Wang, Hui Li, Peter Todd Stukenburg, Chongzhi Zang. Integration analysis of lncRNAs and R-loops in 3D genome organization and gene regulation in the cancer genome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7355.

Integrative Analysis of Multi-Omic Data for the Characteristics of Endometrial Cancer.

Endometrial cancer (EC) is a frequently diagnosed gynecologic cancer. Identifying reliable prognostic genes for predicting EC onset is crucial for reducing patient morbidity and mortality. Here, a comprehensive strategy with transcriptomic and proteomic data was performed to measure EC's characteristics. Based on the publicly available RNA-seq data, death-associated protein kinase 3, recombination signal-binding protein for the immunoglobulin kappa J region, and myosin light chain 9 were screened out as potential biomarkers that affect the EC patients' prognosis. A linear model was further constructed by multivariate Cox regression for the prediction of the risk of being malignant. From further integrative analysis, exosomes were found to have a highly enriched role that might participate in EC occurrence. The findings were validated by qRT-polymerase chain reaction (PCR) and western blotting. Collectively, we constructed a prognostic-gene-based model for EC prediction and found that exosomes participate in EC incidents, revealing significantly promising support for the diagnosis of EC.

METTL16 promotes liver cancer stem cell self-renewal via controlling ribosome biogenesis and mRNA translation

BackgroundWhile liver cancer stem cells (CSCs) play a crucial role in hepatocellular carcinoma (HCC) initiation, progression, recurrence, and treatment resistance, the mechanism underlying liver CSC self-renewal remains elusive. We aim to characterize the role of Methyltransferase 16 (METTL16), a recently identified RNA N6-methyladenosine (m6A) methyltransferase, in HCC development/maintenance, CSC stemness, as well as normal hepatogenesis.MethodsLiver-specific Mettl16 conditional KO (cKO) mice were generated to assess its role in HCC pathogenesis and normal hepatogenesis. Hydrodynamic tail-vein injection (HDTVi)-induced de novo hepatocarcinogenesis and xenograft models were utilized to determine the role of METTL16 in HCC initiation and progression. A limiting dilution assay was utilized to evaluate CSC frequency. Functionally essential targets were revealed via integrative analysis of multi-omics data, including RNA-seq, RNA immunoprecipitation (RIP)-seq, and ribosome profiling.ResultsMETTL16 is highly expressed in liver CSCs and its depletion dramatically decreased CSC frequency in vitro and in vivo. Mettl16 KO significantly attenuated HCC initiation and progression, yet only slightly influenced normal hepatogenesis. Mechanistic studies, including high-throughput sequencing, unveiled METTL16 as a key regulator of ribosomal RNA (rRNA) maturation and mRNA translation and identified eukaryotic translation initiation factor 3 subunit a (eIF3a) transcript as a bona-fide target of METTL16 in HCC. In addition, the functionally essential regions of METTL16 were revealed by CRISPR gene tiling scan, which will pave the way for the development of potential inhibitor(s).ConclusionsOur findings highlight the crucial oncogenic role of METTL16 in promoting HCC pathogenesis and enhancing liver CSC self-renewal through augmenting mRNA translation efficiency.

Multi-omics integration with weighted affinity and self-diffusion applied for cancer subtypes identification

BackgroundCharacterizing cancer molecular subtypes is crucial for improving prognosis and individualized treatment. Integrative analysis of multi-omics data has become an important approach for disease subtyping, yielding better understanding of the complex biology. Current multi-omics integration tools and methods for cancer subtyping often suffer challenges of high computational efficiency as well as the problem of weight assignment on data types.ResultsHere, we present an efficient multi-omics integration via weighted affinity and self-diffusion (MOSD) to dissect cancer heterogeneity. MOSD first construct local scaling affinity on each data type and then integrate all affinities by weighted linear combination, followed by the self-diffusion to further improve the patients’ similarities for the downstream clustering analysis. To demonstrate the effectiveness and usefulness for cancer subtyping, we apply MOSD across ten cancer types with three measurements (Gene expression, DNA methylation, miRNA).ConclusionsOur approach exhibits more significant differences in patient survival and computationally efficient benchmarking against several state-of-art integration methods and the identified molecular subtypes reveal strongly biological interpretability. The code as well as its implementation are available in GitHub: https://github.com/DXCODEE/MOSD.

MosGraphGen: a novel tool to generate multi-omics signaling graphs to facilitate integrative and interpretable graph AI model development.

Multi-omics data, i.e. genomics, epigenomics, transcriptomics, proteomics, characterize cellular complex signaling systems from multi-level and multi-view and provide a holistic view of complex cellular signaling pathways. However, it remains challenging to integrate and interpret multi-omics data for mining critical biomarkers. Graph AI models have been widely used to analyze graph-structure datasets, and are ideal for integrative multi-omics data analysis because they can naturally integrate and represent multi-omics data as a biologically meaningful multi-level signaling graph and interpret multi-omics data via graph node and edge ranking analysis. Nevertheless, it is nontrivial for graph-AI model developers to pre-analyze multi-omics data and convert the data into biologically meaningful graphs, which can be directly fed into graph-AI models. To resolve this challenge, we developed mosGraphGen (multi-omics signaling graph generator), generating Multi-omics Signaling graphs (mos-graph) of individual samples by mapping multi-omics data onto a biologically meaningful multi-level background signaling network with data normalization by aggregating measurements and aligning to the reference genome. With mosGraphGen, AI model developers can directly apply and evaluate their models using these mos-graphs. In the results, mosGraphGen was used and illustrated using two widely used multi-omics datasets of The Cancer Genome Atlas (TCGA) and Alzheimer's disease (AD) samples. The code of mosGraphGen is open-source and publicly available via GitHub: https://github.com/FuhaiLiAiLab/mosGraphGen.

Machine learning combining multi-omics data and network algorithms identifies adrenocortical carcinoma prognostic biomarkers.

Background: Rare endocrine cancers such as Adrenocortical Carcinoma (ACC) present a serious diagnostic and prognostication challenge. The knowledge about ACC pathogenesis is incomplete, and patients have limited therapeutic options. Identification of molecular drivers and effective biomarkers is required for timely diagnosis of the disease and stratify patients to offer the most beneficial treatments. In this study we demonstrate how machine learning methods integrating multi-omics data, in combination with system biology tools, can contribute to the identification of new prognostic biomarkers for ACC. Methods: ACC gene expression and DNA methylation datasets were downloaded from the Xena Browser (GDC TCGA Adrenocortical Carcinoma cohort). A highly correlated multi-omics signature discriminating groups of samples was identified with the data integration analysis for biomarker discovery using latent components (DIABLO) method. Additional regulators of the identified signature were discovered using Clarivate CBDD (Computational Biology for Drug Discovery) network propagation and hidden nodes algorithms on a curated network of molecular interactions (MetaBase™). The discriminative power of the multi-omics signature and their regulators was delineated by training a random forest classifier using 55 samples, by employing a 10-fold cross validation with five iterations. The prognostic value of the identified biomarkers was further assessed on an external ACC dataset obtained from GEO (GSE49280) using the Kaplan-Meier estimator method. An optimal prognostic signature was finally derived using the stepwise Akaike Information Criterion (AIC) that allowed categorization of samples into high and low-risk groups. Results: A multi-omics signature including genes, micro RNA's and methylation sites was generated. Systems biology tools identified additional genes regulating the features included in the multi-omics signature. RNA-seq, miRNA-seq and DNA methylation sets of features revealed a high power to classify patients from stages I-II and stages III-IV, outperforming previously identified prognostic biomarkers. Using an independent dataset, associations of the genes included in the signature with Overall Survival (OS) data demonstrated that patients with differential expression levels of 8 genes and 4 micro RNA's showed a statistically significant decrease in OS. We also found an independent prognostic signature for ACC with potential use in clinical practice, combining 9-gene/micro RNA features, that successfully predicted high-risk ACC cancer patients. Conclusion: Machine learning and integrative analysis of multi-omics data, in combination with Clarivate CBDD systems biology tools, identified a set of biomarkers with high prognostic value for ACC disease. Multi-omics data is a promising resource for the identification of drivers and new prognostic biomarkers in rare diseases that could be used in clinical practice.