Global coordination level in single-cell transcriptomic data

Abstract

Genes are linked by underlying regulatory mechanisms and by jointly implementing biological functions, working in coordination to apply different tasks in the cells. Assessing the coordination level between genes from single-cell transcriptomic data, without a priori knowledge of the map of gene regulatory interactions, is a challenge. A ‘top-down’ approach has recently been developed to analyze single-cell transcriptomic data by evaluating the global coordination level between genes (called GCL). Here, we systematically analyze the performance of the GCL in typical scenarios of single-cell RNA sequencing (scRNA-seq) data. We show that an individual anomalous cell can have a disproportionate effect on the GCL calculated over a cohort of cells. In addition, we demonstrate how the GCL is affected by the presence of clusters, which are very common in scRNA-seq data. Finally, we analyze the effect of the sampling size of the Jackknife procedure on the GCL statistics. The manuscript is accompanied by a description of a custom-built Python package for calculating the GCL. These results provide practical guidelines for properly pre-processing and applying the GCL measure in transcriptional data.

Nature of epigenetic aging from a single-cell perspective

“…To contrast co-regulation and stochastic patterns during aging, we analyze embryonic development scDNAm data as a control for co-regulated changes because of a tightly controlled genetic program that governs development. Finally, in scRNAseq data, we find supportive evidence of significant global coordination levels (GCL) 29,30 of gene expression for genes associated with co-regulated CpG clusters compared to non-significant GCL for stochastic CpG clusters.…”

“…To test the hypothesis that epigenetic co-regulation affects transcriptomic coordination, we examined the single-cell RNAseq data corresponding to the scDNAm data analyzed above from the multi-omic aging muscle stem cells (muSC) 49 . The transcriptional coordination of a set of genes was measured with the help of the global coordination level (GCL) 29,30 . First, we mapped the list of CpGs comprising co-regulated or stochastic clusters to genes ( Fig.…”

“…The CpGs with zero coverage were not used for calculating the Pearson correlation. The method of identifying co-regulated clusters is conceptually similar to the global coordination level (GCL) metric developed independently for scRNAseq data analysis 29,30 . The majority of age-related CpG sites, 92%, changed according to the stochastic scenario, whereas 8% of the CpG sites changed in a co-regulated manner.…”

Progress in Discovering Transcriptional Noise in Aging

“…In addition to the implementation of existing measurements of data variation, novel measurements have been developed specifically for transcriptional noise. These include Bayesian analysis of single-cell sequencing (BASiCS) [ 47 ], the method of Isildak et al (which we call residual-based correlation; RBC) [ 48 ], global coordination level (GCL) [ 26 , 49 ], Decibel [ 39 ], Scallop [ 39 ], and the method of Marti et al (which we call technically induced noise approximation; TINA) [ 28 ]. Because these methods involve more comprehensive definitions than CoV and FF, the reader can refer to the original papers for their mathematical or statistical equations.…”

“…A higher GCL value indicates lower transcriptional noise. While GCL analysis is a promising new method that incorporates the detection of an age-related increase in transcriptional noise in a wide variety of different species and cell types, pre-filtering the input of scRNA-seq data has been shown to be critical because “outlier” cells and cell clustering can significantly impact its results [ 49 ].…”