Inferring Regulatory Networks From Mixed Observational Data Using Directed Acyclic Graphs

Inferring gene regulatory networks from single-cell multiome data using atlas-scale external data

Existing methods for gene regulatory network (GRN) inference rely on gene expression data alone or on lower resolution bulk data. Despite the recent integration of chromatin accessibility and RNA sequencing data, learning complex mechanisms from limited independent data points still presents a daunting challenge. Here we present LINGER (Lifelong neural network for gene regulation), a machine-learning method to infer GRNs from single-cell paired gene expression and chromatin accessibility data. LINGER incorporates atlas-scale external bulk data across diverse cellular contexts and prior knowledge of transcription factor motifs as a manifold regularization. LINGER achieves a fourfold to sevenfold relative increase in accuracy over existing methods and reveals a complex regulatory landscape of genome-wide association studies, enabling enhanced interpretation of disease-associated variants and genes. Following the GRN inference from reference single-cell multiome data, LINGER enables the estimation of transcription factor activity solely from bulk or single-cell gene expression data, leveraging the abundance of available gene expression data to identify driver regulators from case-control studies.

Decomposition of variation of mixed variables by a latent mixed Gaussian copula model

Many biomedical studies collect data of mixed types of variables from multiple groups of subjects. Some of these studies aim to find the group-specific and the common variation among all these variables. Even though similar problems have been studied by some previous works, their methods mainly rely on the Pearson correlation, which cannot handle mixed data. To address this issue, we propose a latent mixed Gaussian copula (LMGC) model that can quantify the correlations among binary, ordinal, continuous, and truncated variables in a unified framework. We also provide a tool to decompose the variation into the group-specific and the common variation over multiple groups via solving a regularized M-estimation problem. We conduct extensive simulation studies to show the advantage of our proposed method over the Pearson correlation-based methods. We also demonstrate that by jointly solving the M-estimation problem over multiple groups, our method is better than decomposing the variation group by group. We also apply our method to a Chlamydia trachomatis genital tract infection study to demonstrate how it can be used to discover informative biomarkers that differentiate patients.

DAGBagM: learning directed acyclic graphs of mixed variables with an application to identify protein biomarkers for treatment response in ovarian cancer

BACKGROUND

Applying directed acyclic graph (DAG) models to proteogenomic data has been shown effective for detecting causal biomarkers of complex diseases. However, there remain unsolved challenges in DAG learning to jointly model binary clinical outcome variables and continuous biomarker measurements.

RESULTS

In this paper, we propose a new tool, DAGBagM, to learn DAGs with both continuous and binary nodes. By using appropriate models, DAGBagM allows for either continuous or binary nodes to be parent or child nodes. It employs a bootstrap aggregating strategy to reduce false positives in edge inference. At the same time, the aggregation procedure provides a flexible framework to robustly incorporate prior information on edges.

CONCLUSIONS

Through extensive simulation experiments, we demonstrate that DAGBagM has superior performance compared to alternative strategies for modeling mixed types of nodes. In addition, DAGBagM is computationally more efficient than two competing methods. When applying DAGBagM to proteogenomic datasets from ovarian cancer studies, we identify potential protein biomarkers for platinum refractory/resistant response in ovarian cancer. DAGBagM is made available as a github repository at https://github.com/jie108/dagbagM .

Integrative eQTL-weighted hierarchical Cox models for SNP-set based time-to-event association studies

BACKGROUND

Integrating functional annotations into SNP-set association studies has been proven a powerful analysis strategy. Statistical methods for such integration have been developed for continuous and binary phenotypes; however, the SNP-set integrative approaches for time-to-event or survival outcomes are lacking.

METHODS

We here propose IEHC, an integrative eQTL (expression quantitative trait loci) hierarchical Cox regression, for SNP-set based survival association analysis by modeling effect sizes of genetic variants as a function of eQTL via a hierarchical manner. Three p-values combination tests are developed to examine the joint effects of eQTL and genetic variants after a novel decorrelated modification of statistics for the two components. An omnibus test (IEHC-ACAT) is further adapted to aggregate the strengths of all available tests.

RESULTS

Simulations demonstrated that the IEHC joint tests were more powerful if both eQTL and genetic variants contributed to association signal, while IEHC-ACAT was robust and often outperformed other approaches across various simulation scenarios. When applying IEHC to ten TCGA cancers by incorporating eQTL from relevant tissues of GTEx, we revealed that substantial correlations existed between the two types of effect sizes of genetic variants from TCGA and GTEx, and identified 21 (9 unique) cancer-associated genes which would otherwise be missed by approaches not incorporating eQTL.

CONCLUSION

IEHC represents a flexible, robust, and powerful approach to integrate functional omics information to enhance the power of identifying association signals for the survival risk of complex human cancers.

Integration strategies of multi-omics data for machine learning analysis

Increased availability of high-throughput technologies has generated an ever-growing number of omics data that seek to portray many different but complementary biological layers including genomics, epigenomics, transcriptomics, proteomics, and metabolomics. New insight from these data have been obtained by machine learning algorithms that have produced diagnostic and classification biomarkers. Most biomarkers obtained to date however only include one omic measurement at a time and thus do not take full advantage of recent multi-omics experiments that now capture the entire complexity of biological systems.

Multi-omics data integration strategies are needed to combine the complementary knowledge brought by each omics layer. We have summarized the most recent data integration methods/ frameworks into five different integration strategies: early, mixed, intermediate, late and hierarchical. In this mini-review, we focus on challenges and existing multi-omics integration strategies by paying special attention to machine learning applications.

Regulation of Canonical Oncogenic Signaling Pathways in Cancer via DNA Methylation

Disruption of signaling pathways that plays a role in the normal development and cellular homeostasis may lead to the dysregulation of cellular signaling and bring about the onset of different diseases, including cancer. In addition to genetic aberrations, DNA methylation also acts as an epigenetic modifier to drive the onset and progression of cancer by mediating the reversible transcription of related genes. Although the role of DNA methylation as an alternative driver of carcinogenesis has been well-established, the global effects of DNA methylation on oncogenic signaling pathways and the presentation of cancer is only emerging. In this article, we introduced a differential methylation parsing pipeline (MethylMine) which mined for epigenetic biomarkers based on feature selection. This pipeline was used to mine for biomarkers, which presented a substantial difference in methylation between the tumor and the matching normal tissue samples. Combined with the Data Integration Analysis for Biomarker discovery (DIABLO) framework for machine learning and multi-omic analysis, we revisited the TCGA DNA methylation and RNA-Seq datasets for breast, colorectal, lung, and prostate cancer, and identified differentially methylated genes within the NRF2-KEAP1/PI3K oncogenic pathway, which regulates the expression of cytoprotective genes, that serve as potential therapeutic targets to treat different cancers.