• 在HSC中没有观察到cell-cell variation随着衰老的改变；For example, despite observations of genetic and epigenteic damage in ageing haematopoietic stem cells(HSCs); cell to cell transcriptional variability is not observed.

  • 基因共表达方法的缺陷：

    • First, co-expression networks estimates the direct or indirect correlations between pairs of genes, while an individual gene may be controlled by multiple regulators.

    • Second, each co-expression measure is designed to capture a specific feature that is not necessarily optimal for depicting all types of gene-to-gene transcriptional interrelations ( PCC, linear relatiosnhips)

    • Third, large calculated coexpresion matrices contain a considerable amount of noise, which raises an additional difficulty in explporing their differentiation across cohorts

  • Gcl 本质上是对bcdcorr的bootstrap.

  • 一个重要的观察和建设： As an illustration of the coordination measured by the GCL, consider the expression profiles of cells with N genes, that are represented as points in an N-dimensional space. If the gene expression levels are not independent, the set of points do not fill the N- dimensional space but are rather located near lower-dimensional manifold.

  • The GCL measures has two main advantages.

    • The dependency level is defined with respect to a general dependency form, not specific relations ( such as linear)
    • It can include high-order dependencies between multiple variables