Cloudy with a Chance of Insights: Context Dependent Gene Regulation and Implications for Evolutionary Studies

Abstract

Research in various fields of evolutionary biology has shown that divergence in gene expression is a key driver for phenotypic evolution. An exceptional contribution of cis-regulatory divergence has been found to contribute to morphological diversification. In the light of these findings, the analysis of genome-wide expression data has become one of the central tools to link genotype and phenotype information on a more mechanistic level. However, in many studies, especially if general conclusions are drawn from such data, a key feature of gene regulation is often neglected. With our article, we want to raise awareness that gene regulation and thus gene expression is highly context dependent. Genes show tissue- and stage-specific expression. We argue that the regulatory context must be considered in comparative expression studies.

Keywords: Assay for Transposase-Accessible Chromatin using sequencing (ATACseq); ChIPseq; RNAseq; allele specific expression; chromatin; evolution; expression quantitative trait loci (eQTL); gene expression; gene regulation; genotype–phenotype map.

Figure 1

Gene expression is regulated on various levels. (A) The DNA is compressed in the nucleus of the cell. (B) The DNA in the nucleus is compressed by binding of histone proteins. The chromatin contains easily accessible euchromatin regions and highly compact and inaccessible heterochromatin regions. The status of the chromatin is influenced by post-translational histone modifications. Gene expression is modulated by the chromatin state and DNA modifications, such as methylations. (C) Key steps of gene expression (a–d). Transcription factors (TFs) bind to the DNA at specific sequences (1). TF binding activates the transcription initiation complex (2) through conformation changes (looping) of the DNA (3). TFs can also repress transcription, for instance by binding of a co-factor (4). Next generation sequencing (NGS)-based methods that can be applied to study certain aspects of gene regulation are mentioned in red in brackets. See Table 1 for an overview of the methods mentioned here.

Figure 2

Generic factors that are expressed across different tissue can be excluded in correlation studies. (A) If specific candidate genes that are differentially expressed between phenotype A and B are supposed to be revealed, one can generate a comparable dataset for additional tissues. (B) Each pairwise comparison will reveal a certain number of differentially expressed genes (DEGs). The DEGs that are common in two (1–3) or all organs (4) are most likely generic factors that may be less informative for follow-up analyses.