Multiple Types Of Omics Data Research Articles

Does combining numerous data types in multi-omics data improve or hinder performance in survival prediction? Insights from a large-scale benchmark study

BackgroundPredictive modeling based on multi-omics data, which incorporates several types of omics data for the same patients, has shown potential to outperform single-omics predictive modeling. Most research in this domain focuses on incorporating numerous data types, despite the complexity and cost of acquiring them. The prevailing assumption is that increasing the number of data types necessarily improves predictive performance. However, the integration of less informative or redundant data types could potentially hinder this performance. Therefore, identifying the most effective combinations of omics data types that enhance predictive performance is critical for cost-effective and accurate predictions.MethodsIn this study, we systematically evaluated the predictive performance of all 31 possible combinations including at least one of five genomic data types (mRNA, miRNA, methylation, DNAseq, and copy number variation) using 14 cancer datasets with right-censored survival outcomes, publicly available from the TCGA database. We employed various prediction methods and up-weighted clinical data in every model to leverage their predictive importance. Harrell’s C-index and the integrated Brier Score were used as performance measures. To assess the robustness of our findings, we performed a bootstrap analysis at the level of the included datasets. Statistical testing was conducted for key results, limiting the number of tests to ensure a low risk of false positives.ResultsContrary to expectations, we found that using only mRNA data or a combination of mRNA and miRNA data was sufficient for most cancer types. For some cancer types, the additional inclusion of methylation data led to improved prediction results. Far from enhancing performance, the introduction of more data types most often resulted in a decline in performance, which varied between the two performance measures.ConclusionsOur findings challenge the prevailing notion that combining multiple omics data types in multi-omics survival prediction improves predictive performance. Thus, the widespread approach in multi-omics prediction of incorporating as many data types as possible should be reconsidered to avoid suboptimal prediction results and unnecessary expenditure.

Smccnet 2.0: a comprehensive tool for multi-omics network inference with shiny visualization

SummarySparse multiple canonical correlation network analysis (SmCCNet) is a machine learning technique for integrating omics data along with a variable of interest (e.g., phenotype of complex disease), and reconstructing multi-omics networks that are specific to this variable. We present the second-generation SmCCNet (SmCCNet 2.0) that adeptly integrates single or multiple omics data types along with a quantitative or binary phenotype of interest. In addition, this new package offers a streamlined setup process that can be configured manually or automatically, ensuring a flexible and user-friendly experience.AvailabilityThis package is available in both CRAN: https://cran.r-project.org/web/packages/SmCCNet/index.html and Github: https://github.com/KechrisLab/SmCCNet under the MIT license. The network visualization tool is available at https://smccnet.shinyapps.io/smccnetnetwork/.

Abstract 3952: Multiomic single-cell tumor evolution models of minimal residual disease in pediatric B-cell acute lymphoblastic leukemia

Abstract Detection of minimal residual disease (MRD) after initial diagnosis (IDX) and therapy is a strong prognostic marker for relapse in pediatric B-cell acute lymphoblastic leukemia (B-ALL). Previous studies have explored clonal genetic and gene expression changes between initial diagnosis and relapse. However, studying tumor cells present at MRD is challenging due to their low abundance. Examining the genomic profiles of leukemia cells that resist induction chemotherapy and understanding their evolutionary relationship to cells present at diagnosis may lead to a better understanding of the cells that eventually seed relapse. We analyzed somatic point mutations, indels, and copy number variants from 41 pediatric B-ALL patients with high-risk B-ALL. We specifically focused on patients from the following four common high-risk subtypes: KMT2A-rearranged, Ph+, Ph-like, and iAMP21, who had detectable disease at the end of induction. Leukemic burden at end-induction (day 29) ranged from 0.03% to 66.00%. Data were generated in partnership with the National Cancer Institute’s Human Tumor Atlas Network and the Children’s Oncology Group. Our data set of 41 patients includes 39 with paired initial diagnostic and end-induction samples that we used to create tumor evolution models based on single cells across disease stages. We produced single-cell multiome gene expression (scRNA) and chromatin accessibility (scATAC) profiles (10x Genomics) with a median of 7428 leukemic blast cells at IDX (range: 2490-14219) and 1907 at MRD (range: 240-10167) after filtering. We called somatic mutations from scRNA and scATAC using SComatic and Monopogen, and we detected copy number changes using Numbat and EpiAneufinder. We also designed a targeted DNA sequencing panel including 335 variants based on B-ALL literature and database searches to produce scDNA (Mission Bio) for 24 patients (16 IDX-MRD pairs) with a median of 2346 cells profiled at IDX (range: 340-8849) and 1828 at MRD (range: 596-5133). We built tumor phylogenies based on somatic SNVs and CNVs with COMPASS. Using multiple tools and complementary single-cell omics data types to detect somatic changes improved our confidence in the events used for tumor evolution tree construction. Our approach allowed us to define recurrent genomic events and patterns across pediatric B-ALL subtypes and to describe the clonal evolution of cells measured at initial diagnostic and end-induction disease stages. We were also able to link cell-specific mutation events to their transcriptional and epigenomic signatures, including mapping mutation events along the B-cell development trajectory and within differentially expressed gene pathways. Some cases showed dramatic shifts in subclonal structure between IDX and MRD that may be related to therapy-induced clonal selection, such as a loss of NRAS hotspot mutation. Citation Format: Steven M. Foltz, Avi Loren, Changya Chen, Rushabh Mehta, Elizabeth Li, Jason Xu, Fatemeh Alikarami, Kathrin M. Bernt, Kai Tan. Multiomic single-cell tumor evolution models of minimal residual disease in pediatric B-cell acute lymphoblastic leukemia [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 3952.

Poly-omic risk scores predict inflammatory bowel disease diagnosis.

Inflammatory bowel disease (IBD) is characterized by complex etiology and a disrupted colonic ecosystem. We provide a framework for the analysis of multi-omic data, which we apply to study the gut ecosystem in IBD. Specifically, we train and validate models using data on the metagenome, metatranscriptome, virome, and metabolome from the Human Microbiome Project 2 IBD multi-omic database, with 1,785 repeated samples from 130 individuals (103 cases and 27 controls). After splitting the participants into training and testing groups, we used mixed-effects least absolute shrinkage and selection operator regression to select features for each omic. These features, with demographic covariates, were used to generate separate single-omic prediction scores. All four single-omic scores were then combined into a final regression to assess the relative importance of the individual omics and the predictive benefits when considered together. We identified several species, pathways, and metabolites known to be associated with IBD risk, and we explored the connections between data sets. Individually, metabolomic and viromic scores were more predictive than metagenomics or metatranscriptomics, and when all four scores were combined, we predicted disease diagnosis with a Nagelkerke's R2 of 0.46 and an area under the curve of 0.80 (95% confidence interval: 0.63, 0.98). Our work supports that some single-omic models for complex traits are more predictive than others, that incorporating multiple omic data sets may improve prediction, and that each omic data type provides a combination of unique and redundant information. This modeling framework can be extended to other complex traits and multi-omic data sets.IMPORTANCEComplex traits are characterized by many biological and environmental factors, such that multi-omic data sets are well-positioned to help us understand their underlying etiologies. We applied a prediction framework across multiple omics (metagenomics, metatranscriptomics, metabolomics, and viromics) from the gut ecosystem to predict inflammatory bowel disease (IBD) diagnosis. The predicted scores from our models highlighted key features and allowed us to compare the relative utility of each omic data set in single-omic versus multi-omic models. Our results emphasized the importance of metabolomics and viromics over metagenomics and metatranscriptomics for predicting IBD status. The greater predictive capability of metabolomics and viromics is likely because these omics serve as markers of lifestyle factors such as diet. This study provides a modeling framework for multi-omic data, and our results show the utility of combining multiple omic data types to disentangle complex disease etiologies and biological signatures.

Identify Alzheimer’s disease subtypes and markers from multi‐omic data of human brain and blood with a subspace merging algorithm

AbstractBackgroundAlzheimer’s disease (AD) is a highly heterogeneous disease with different disease trajectory, brain pathological changes, and risk factors. Identifying subtypes of AD can have greatly impact AD diagnosis, treatment, and disease management. Our goal is to identify subtypes of AD by integrating multiple types of omics data from AD patients’ brain samples and blood samples.MethodWe applied a subspace merging algorithm on AD patient brain samples with matched transcriptomic, DNA methylation, and proteomics data collected from the ROSMAP study. This algorithm merges patient graphs generated from these three datasets on a Grassmann manifold followed by spectral clustering. Differential gene expression and pathway analysis were applied to establish the gene signatures between the subtypes (patient clusters) and to the transcriptomic data of the peripheral blood monocyte (PBMC) from the matching patients to identify possible blood markers.ResultOur analysis identified two patient clusters between which the cognitive test scores or the degree of AD pathology are significantly different. Differentially expressed genes between the two groups were identified from both brain and blood samples. Gene set enrichment analysis identified a group of genes significantly enriched with synaptic functions in the brain samples. The Ingenuity Pathway Analysis was used to examine any connections between the significantly differentially expressed genes in the brain samples with the ones identified from the PBMC samples. The low density lipoprotein receptor (LDLR) gene, which was an APOE receptor observed from the blood sample is connected with 12 genes from the brain sample. The results suggested that LDLR could be a novel blood biomarker for AD subtyping.ConclusionThe subspace merging algorithm integrated multimodal omics data from the patients and successfully identified two clusters with distinct clinical progression. Differential expression analysis identified 388 significant genes from the brain samples and 18 from the PBMC samples. The pathway analysis indicated that LDLR could be a novel blood biomarker for AD subtyping. Future work will focus on validating LDLR as a candidate blood marker for AD subtyping using other public datasets with similar study design. The clinical relevance and molecular mechanism basis for the findings will also be investigated.

Integrated multi-omics analysis of brain aging in female nonhuman primates reveals altered signaling pathways relevant to age-related disorders

The prefrontal cortex (PFC) has been implicated as a key brain region responsible for age-related cognitive decline. Little is known about aging-related molecular changes in PFC that may mediate these effects. To date, no studies have used untargeted discovery methods with integrated analyses to determine PFC molecular changes in healthy female primates. We quantified PFC changes associated with healthy aging in female baboons by integrating multiple omics data types (transcriptomics, proteomics, metabolomics) from samples across the adult age span. Our integrated omics approach using unbiased weighted gene co-expression network analysis to integrate data and treat age as a continuous variable, revealed highly interconnected known and novel pathways associated with PFC aging. We found Gamma-aminobutyric acid (GABA) tissue content associated with these signaling pathways, providing 1 potential biomarker to assess PFC changes with age. These highly coordinated pathway changes during aging may represent early steps for aging-related decline in PFC functions, such as learning and memory, and provide potential biomarkers to assess cognitive status in humans.

Advances in deep learning-based cancer outcome prediction using multi-omics data

Cancer prognosis reflects a complex biological process measured by multiple types of omics data. Deep learning frameworks have been proposed to integrate multi-omics data and predict patient outcomes in different cancer types, potentially revolutionizing cancer prognosis with superior performance. This minireview summarizes the advances in the strategies for multi-omics data integration and the performance of different deep learning models in prognosis prediction of diverse cancer types using multi-omics data published in the past 18 months. The challenges and limitations of deep learning models for predicting cancer outcomes based on multi-omics data are discussed.

Combining time-resolved transcriptomics and proteomics data for Adverse Outcome Pathway refinement in ecotoxicology

Conventional Environmental Risk Assessment (ERA) of pesticide pollution is based on soil concentrations and apical endpoints, such as the reproduction of test organisms, but has traditionally disregarded information along the organismal response cascade leading to an adverse outcome. The Adverse Outcome Pathway (AOP) framework includes response information at any level of biological organization, providing opportunities to use intermediate responses as a predictive read-out for adverse outcomes instead. Transcriptomic and proteomic data can provide thousands of data points on the response to toxic exposure. Combining multiple omics data types is necessary for a comprehensive overview of the response cascade and, therefore, AOP development. However, it is unclear if transcript and protein responses are synchronized in time or time lagged. To understand if analysis of multi-omics data obtained at the same timepoint reveal one synchronized response cascade, we studied time-resolved shifts in gene transcript and protein abundance in the springtail Folsomia candida, a soil ecotoxicological model, after exposure to the neonicotinoid insecticide imidacloprid. We analyzed transcriptome and proteome data every 12 h up to 72 h after onset of exposure. The most pronounced shift in both transcript and protein abundances was observed after 48 h exposure. Moreover, cross-correlation analyses indicate that most genes displayed the highest correlation between transcript and protein abundances without a time-lag. This demonstrates that a combined analysis of transcriptomic and proteomic data from the same time-point can be used for AOP improvement. This data will promote the development of biomarkers for the presence of neonicotinoid insecticides or chemicals with a similar mechanism of action in soils.

Using graph-based model to identify cell specific synthetic lethal effects

Synthetic lethal (SL) pairs are pairs of genes whose simultaneous loss-of-function results in cell death, while a damaging mutation of either gene alone does not affect the cell’s survival. This makes SL pairs attractive targets for precision cancer therapies, as targeting the unimpaired gene of the SL pair can selectively kill cancer cells that already harbor the impaired gene. Limited by the difficulty of finding true SL pairs, especially on specific cell types, current computational approaches provide only limited insights because of overlooking the crucial aspects of cellular context dependency and mechanistic understanding of SL pairs. As a result, the identification of SL targets still relies on expensive, time-consuming experimental approaches. In this work, we applied cell-line specific multi-omics data to a specially designed deep learning model to predict cell-line specific SL pairs. Through incorporating multiple types of cell-specific omics data with a self-attention module, we represent gene relationships as graphs. Our approach achieves the prediction of SL pairs in a cell-specific manner and demonstrates the potential to facilitate the discovery of cell-specific SL targets for cancer therapeutics, providing a tool to unearth mechanisms underlying the origin of SL in cancer biology. The code and data of our approach can be found at https://github.com/promethiume/SLwise

A Practical Guide to Inferring Multi-Omics Networks in Plant Systems.

The inference of gene regulatory networks can reveal molecular connections underlying biological processes and improve our understanding of complex biological phenomena in plants. Many previous network studies have inferred networks using only one type of omics data, such as transcriptomics. However, given more recent work applying multi-omics integration in plant biology, such as combining (phospho)proteomics with transcriptomics, it may be advantageous to integrate multiple omics data types into a comprehensive network prediction. Here, we describe a state-of-the-art approach for integrating multi-omics data with gene regulatory network inference to describe signaling pathways and uncover novel regulators. We detail how to download and process transcriptomics and (phospho)proteomics data for network inference, using an example dataset from the plant hormone signaling field. We provide a step-by-step protocol for inference, visualization, and analysis of an integrative multi-omics network using currently available methods. This chapter serves as an accessible guide for novice and intermediate bioinformaticians to analyze their own datasets and reanalyze published work.

Advances in fermented foods revealed by multi-omics: A new direction toward precisely clarifying the roles of microorganisms.

Fermented foods generally comprise a complex micro-ecosystem with beneficial microbiota, functional products, and special flavors and qualities that are welcomed globally. Single-omics analysis allows for a comprehensive characterization of the main microbial factors influencing the function, flavor, and quality of fermented foods. However, the species, relative abundance, viability, growth patterns, and metabolic processes of microorganisms vary with changes in processing and environmental conditions during fermentation. Furthermore, the mechanisms underlying the complex interaction among microorganisms are still difficult to completely understand and analyze. Recently, multi-omics analysis and the integration of multiple types of omics data allowed researchers to more comprehensively explore microbial communities and understand the precise relationship between fermented foods and their functions, flavors, and qualities. Multi-omics approaches might help clarify the mechanisms underpinning the fermentation processes, metabolites, and functional components of these communities. This review clarified the recent advances in the roles of microorganisms in fermented foods based on multi-omics data. Current research achievements may allow for the precise control of the whole industrial processing technology of fermented foods, meeting consumers' expectations of healthy products.

Comprehensive analysis of TCGA data reveals correlation between DNA methylation and alternative splicing

The effect of DNA methylation on the regulation of gene expression has been extensively discussed in the literature. However, the potential association between DNA methylation and alternative splicing is not understood well. In this study, we integrated multiple omics data types from The Cancer Genome Atlas (TCGA) and systematically examined the relationship between DNA methylation and alternative splicing. Using the methylation data and exon expression data, we identified many CpG sites significantly associated with exon expression in various types of cancers. We further observed that the direction and strength of significant CpG-exon correlation tended to be consistent across different cancer contexts, indicating that some CpG-exon correlation patterns reflect fundamental biological mechanisms that transcend tissue- and cancer- types. We also discovered that CpG sites correlated with exon expressions were more likely to be associated with patient survival outcomes compared to CpG sites that did not correlate with exon expressions. Furthermore, we found that CpG sites were more strongly correlated with exon expression than expression of isoforms harboring the corresponding exons. This observation suggests that a major effect of CpG methylation on alternative splicing may be related to the inclusion or exclusion of exons, which subsequently impacts the relative usage of various isoforms. Overall, our study revealed correlation patterns between DNA methylation and alternative splicing, which provides new insights into the role of methylation in the transcriptional process.

Connecting omics signatures and revealing biological mechanisms with iLINCS

There are only a few platforms that integrate multiple omics data types, bioinformatics tools, and interfaces for integrative analyses and visualization that do not require programming skills. Here we present iLINCS (http://ilincs.org), an integrative web-based platform for analysis of omics data and signatures of cellular perturbations. The platform facilitates mining and re-analysis of the large collection of omics datasets (>34,000), pre-computed signatures (>200,000), and their connections, as well as the analysis of user-submitted omics signatures of diseases and cellular perturbations. iLINCS analysis workflows integrate vast omics data resources and a range of analytics and interactive visualization tools into a comprehensive platform for analysis of omics signatures. iLINCS user-friendly interfaces enable execution of sophisticated analyses of omics signatures, mechanism of action analysis, and signature-driven drug repositioning. We illustrate the utility of iLINCS with three use cases involving analysis of cancer proteogenomic signatures, COVID 19 transcriptomic signatures and mTOR signaling.

PathwayMultiomics: An R Package for Efficient Integrative Analysis of Multi-Omics Datasets With Matched or Un-matched Samples.

Recent advances in technology have made multi-omics datasets increasingly available to researchers. To leverage the wealth of information in multi-omics data, a number of integrative analysis strategies have been proposed recently. However, effectively extracting biological insights from these large, complex datasets remains challenging. In particular, matched samples with multiple types of omics data measured on each sample are often required for multi-omics analysis tools, which can significantly reduce the sample size. Another challenge is that analysis techniques such as dimension reductions, which extract association signals in high dimensional datasets by estimating a few variables that explain most of the variations in the samples, are typically applied to whole-genome data, which can be computationally demanding. Here we present pathwayMultiomics, a pathway-based approach for integrative analysis of multi-omics data with categorical, continuous, or survival outcome variables. The input of pathwayMultiomics is pathway p-values for individual omics data types, which are then integrated using a novel statistic, the MiniMax statistic, to prioritize pathways dysregulated in multiple types of omics datasets. Importantly, pathwayMultiomics is computationally efficient and does not require matched samples in multi-omics data. We performed a comprehensive simulation study to show that pathwayMultiomics significantly outperformed currently available multi-omics tools with improved power and well-controlled false-positive rates. In addition, we also analyzed real multi-omics datasets to show that pathwayMultiomics was able to recover known biology by nominating biologically meaningful pathways in complex diseases such as Alzheimer’s disease.

Pan-cancer evaluation of gene expression and somatic alteration data for cancer prognosis prediction

BackgroundOver the past decades, approaches for diagnosing and treating cancer have seen significant improvement. However, the variability of patient and tumor characteristics has limited progress on methods for prognosis prediction. The development of high-throughput omics technologies now provides multiple approaches for characterizing tumors. Although a large number of published studies have focused on integration of multi-omics data and use of pathway-level models for cancer prognosis prediction, there still exists a gap of knowledge regarding the prognostic landscape across multi-omics data for multiple cancer types using both gene-level and pathway-level predictors.MethodsIn this study, we systematically evaluated three often available types of omics data (gene expression, copy number variation and somatic point mutation) covering both DNA-level and RNA-level features. We evaluated the landscape of predictive performance of these three omics modalities for 33 cancer types in the TCGA using a Lasso or Group Lasso-penalized Cox model and either gene or pathway level predictors.ResultsWe constructed the prognostic landscape using three types of omics data for 33 cancer types on both the gene and pathway levels. Based on this landscape, we found that predictive performance is cancer type dependent and we also highlighted the cancer types and omics modalities that support the most accurate prognostic models. In general, models estimated on gene expression data provide the best predictive performance on either gene or pathway level and adding copy number variation or somatic point mutation data to gene expression data does not improve predictive performance, with some exceptional cohorts including low grade glioma and thyroid cancer. In general, pathway-level models have better interpretative performance, higher stability and smaller model size across multiple cancer types and omics data types relative to gene-level models.ConclusionsBased on this landscape and comprehensively comparison, models estimated on gene expression data provide the best predictive performance on either gene or pathway level. Pathway-level models have better interpretative performance, higher stability and smaller model size relative to gene-level models.

OmiEmbed: A Unified Multi-Task Deep Learning Framework for Multi-Omics Data

Simple SummaryOmiEmbed is a unified multi-task deep learning framework for multi-omics data, supporting dimensionality reduction, multi-omics integration, tumour type classification, phenotypic feature reconstruction and survival prediction. The framework is comprised of deep embedding and downstream task modules to capture biomedical information from high-dimensional omics data. OmiEmbed outperformed state-of-the-art methods on all three types of downstream tasks: classification, regression and survival prediction. Better performance was achieved using the multi-task training strategy compared to training each downstream task individually. With the multi-task strategy, OmiEmbed learnt a comprehensive omics embedding containing information from multiple tasks. OmiEmbed is open source, well-organised and convenient to be extended to other customised input data, network structures and downstream tasks, which has promising potential to facilitate more accurate and personalised clinical decision making.High-dimensional omics data contain intrinsic biomedical information that is crucial for personalised medicine. Nevertheless, it is challenging to capture them from the genome-wide data, due to the large number of molecular features and small number of available samples, which is also called “the curse of dimensionality” in machine learning. To tackle this problem and pave the way for machine learning-aided precision medicine, we proposed a unified multi-task deep learning framework named OmiEmbed to capture biomedical information from high-dimensional omics data with the deep embedding and downstream task modules. The deep embedding module learnt an omics embedding that mapped multiple omics data types into a latent space with lower dimensionality. Based on the new representation of multi-omics data, different downstream task modules were trained simultaneously and efficiently with the multi-task strategy to predict the comprehensive phenotype profile of each sample. OmiEmbed supports multiple tasks for omics data including dimensionality reduction, tumour type classification, multi-omics integration, demographic and clinical feature reconstruction, and survival prediction. The framework outperformed other methods on all three types of downstream tasks and achieved better performance with the multi-task strategy compared to training them individually. OmiEmbed is a powerful and unified framework that can be widely adapted to various applications of high-dimensional omics data and has great potential to facilitate more accurate and personalised clinical decision making.

MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification

To fully utilize the advances in omics technologies and achieve a more comprehensive understanding of human diseases, novel computational methods are required for integrative analysis of multiple types of omics data. Here, we present a novel multi-omics integrative method named Multi-Omics Graph cOnvolutional NETworks (MOGONET) for biomedical classification. MOGONET jointly explores omics-specific learning and cross-omics correlation learning for effective multi-omics data classification. We demonstrate that MOGONET outperforms other state-of-the-art supervised multi-omics integrative analysis approaches from different biomedical classification applications using mRNA expression data, DNA methylation data, and microRNA expression data. Furthermore, MOGONET can identify important biomarkers from different omics data types related to the investigated biomedical problems.

Clustering of cancer data based on Stiefel manifold for multiple views

BackgroundIn recent years, various sequencing techniques have been used to collect biomedical omics datasets. It is usually possible to obtain multiple types of omics data from a single patient sample. Clustering of omics data plays an indispensable role in biological and medical research, and it is helpful to reveal data structures from multiple collections. Nevertheless, clustering of omics data consists of many challenges. The primary challenges in omics data analysis come from high dimension of data and small size of sample. Therefore, it is difficult to find a suitable integration method for structural analysis of multiple datasets.ResultsIn this paper, a multi-view clustering based on Stiefel manifold method (MCSM) is proposed. The MCSM method comprises three core steps. Firstly, we established a binary optimization model for the simultaneous clustering problem. Secondly, we solved the optimization problem by linear search algorithm based on Stiefel manifold. Finally, we integrated the clustering results obtained from three omics by using k-nearest neighbor method. We applied this approach to four cancer datasets on TCGA. The result shows that our method is superior to several state-of-art methods, which depends on the hypothesis that the underlying omics cluster class is the same.ConclusionParticularly, our approach has better performance than compared approaches when the underlying clusters are inconsistent. For patients with different subtypes, both consistent and differential clusters can be identified at the same time.

Integrating multi-omics data through deep learning for accurate cancer prognosis prediction

BackgroundGenomic information is nowadays widely used for precise cancer treatments. Since the individual type of omics data only represents a single view that suffers from data noise and bias, multiple types of omics data are required for accurate cancer prognosis prediction. However, it is challenging to effectively integrate multi-omics data due to the large number of redundant variables but relatively small sample size. With the recent progress in deep learning techniques, Autoencoder was used to integrate multi-omics data for extracting representative features. Nevertheless, the generated model is fragile from data noises. Additionally, previous studies usually focused on individual cancer types without making comprehensive tests on pan-cancer. Here, we employed the denoising Autoencoder to get a robust representation of the multi-omics data, and then used the learned representative features to estimate patients’ risks. ResultsBy applying to 15 cancers from The Cancer Genome Atlas (TCGA), our method was shown to improve the C-index values over previous methods by 6.5% on average. Considering the difficulty to obtain multi-omics data in practice, we further used only mRNA data to fit the estimated risks by training XGboost models, and found the models could achieve an average C-index value of 0.627. As a case study, the breast cancer prognosis prediction model was independently tested on three datasets from the Gene Expression Omnibus (GEO), and shown able to significantly separate high-risk patients from low-risk ones (C-index>0.6, p-values<0.05). Based on the risk subgroups divided by our method, we identified nine prognostic markers highly associated with breast cancer, among which seven genes have been proved by literature review. ConclusionOur comprehensive tests indicated that we have constructed an accurate and robust framework to integrate multi-omics data for cancer prognosis prediction. Moreover, it is an effective way to discover cancer prognosis-related genes.

Multi-Staged Data-Integrated Multi-Omics Analysis for Symptom Science Research.

The incorporation of omics approaches into symptom science research can provide researchers with information about the molecular mechanisms that underlie symptoms. Most of the omics analyses in symptom science have used a single omics approach. Therefore, these analyses are limited by the information contained within a specific omics domain (e.g., genomics and inherited variations, transcriptomics and gene function). A multi-staged data-integrated multi-omics (MS-DIMO) analysis integrates multiple types of omics data in a single study. With this integration, a MS-DIMO analysis can provide a more comprehensive picture of the complex biological mechanisms that underlie symptoms. The results of a MS-DIMO analysis can be used to refine mechanistic hypotheses and/or discover therapeutic targets for specific symptoms. The purposes of this paper are to: (1) describe a MS-DIMO analysis using "Symptom X" as an example; (2) discuss a number of challenges associated with specific omics analyses and how a MS-DIMO analysis can address them; (3) describe the various orders of omics data that can be used in a MS-DIMO analysis; (4) describe omics analysis tools; and (5) review case exemplars of MS-DIMO analyses in symptom science. This paper provides information on how a MS-DIMO analysis can strengthen symptom science research through the prioritization of functional genes and biological processes associated with a specific symptom.