Abstract

Eukaryotic genomes are mostly composed of noncoding DNA whose role is still poorly understood. Studies in several organisms have shown correlations between the length of the intergenic and genic sequences of a gene and the expression of its corresponding mRNA transcript. Some studies have found a positive relationship between intergenic sequence length and expression diversity between tissues, and concluded that genes under greater regulatory control require more regulatory information in their intergenic sequences. Other reports found a negative relationship between expression level and gene length and the interpretation was that there is selection pressure for highly expressed genes to remain small. However, a correlation between gene sequence length and expression diversity, opposite to that observed for intergenic sequences, has also been reported, and to date there is no testable explanation for this observation. To shed light on these varied and sometimes conflicting results, we performed a thorough study of the relationships between sequence length and gene expression using cell-type (tissue) specific microarray data in Arabidopsis thaliana. We measured median gene expression across tissues (expression level), expression variability between tissues (expression pattern uniformity), and expression variability between replicates (expression noise). We found that intergenic (upstream and downstream) and genic (coding and noncoding) sequences have generally opposite relationships with respect to expression, whether it is tissue variability, median, or expression noise. To explain these results we propose a model, in which the lengths of the intergenic and genic sequences have opposite effects on the ability of the transcribed region of the gene to be epigenetically regulated for differential expression. These findings could shed light on the role and influence of noncoding sequences on gene expression.

Highlights

  • With the goal of precisely identifying genome-wide interdependencies between the length of the sequences associated with a gene and the expression of its mRNA transcript, we used the genome sequence of A. thaliana and gene expression data derived from microarray experiments

  • We did not separate intergenic sequences based on the orientation of their flanking genes because we did not observe any substantial effect for this factor on their relationship with gene expression

  • We found a positive relationship between the upstream and the downstream intergenic sequence lengths (Figure 2A), and a stronger positive relationship between the coding and gene noncoding sequences (Figure 2B)

Read more

Summary

Objectives

Our aim was to perform a thorough and objective study of the relationships between the mRNA expression of genes and the length of their associated noncoding and coding sequences

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call