A Markov random field model-based approach for differentially expressed gene detection from single-cell RNA-seq data

Abstract

The development of single-cell RNA-sequencing (scRNA-seq) technologies has offered insights into complex biological systems at the single-cell resolution. In particular, these techniques facilitate the identifications of genes showing cell-type-specific differential expressions (DE). In this paper, we introduce MARBLES, a novel statistical model for cross-condition DE gene detection from scRNA-seq data. MARBLES employs a Markov Random Field model to borrow information across similar cell types and utilizes cell-type-specific pseudobulk count to account for sample-level variability. Our simulation results showed that MARBLES is more powerful than existing methods to detect DE genes with an appropriate control of false positive rate. Applications of MARBLES to real data identified novel disease-related DE genes and biological pathways from both a single-cell lipopolysaccharide mouse dataset with 24 381 cells and 11 076 genes and a Parkinson’s disease human data set with 76 212 cells and 15 891 genes. Overall, MARBLES is a powerful tool to identify cell-type-specific DE genes across conditions from scRNA-seq data.

Functional annotation of the human PTSD methylome identifies tissue-specific epigenetic variation across subcortical brain regions

“…Univariate analysis of our dataset identified a handful of genes with significant DMCs, a caveat noted in other postmortem methylation studies 44 . Previous work from our group and others have successfully applied "Joint Analysis" to transcriptomic and other high dimensional genomic data to identify significant biological signals where power maybe lacking 30,31,45,46 . Therefore, we employed a Markov random field model joint analysis to take advantage of the regional colocalization of sub nuclei of the amygdala and subregions of the hippocampus and integrate univariate summary statistics from differential methylation analysis with topology information from these regions in order to improve our power in detecting biologically significant DMCs.…”

“…In fact, we were only able to find a limited number of DMCs (and associated genes) between PTSD and controls in each brain region after the Benjamini-Hochberg correction (Figure 2). The Markov Random Field (MRF) model has been applied to both genome-wide association studies and bulk RNA-seq studies to model biological dependencies/networks in genomic and transcriptomic data [30][31][32] . In these previous Figure 2 | Univariate analysis reveals regional and sex-specific differences in CpG methylation between PTSD cases and controls (A) Two-sided Manhattan plot for PTSD case-control differential methylation using all samples included in the study.…”

Development of a Myeloid-Derived Suppressor Cells-Related Genes Predictive Signature for Prognostic Survival and Immunotherapeutic Responses in Hepatocellular Carcinoma

“…The use of single cell sequencing technology has attracted considerable interest in the biomedical research and clinical practice eld (Zhu et al 2022). This technology enables a comprehensive depiction of cell subpopulations, states, and lineages within heterogeneous tumors.…”