Gene expression patterns are correlated with genomic and genic structure in soybean

Authors: WOODY, JENNA - Iowa State University; SEVERIN, ANDREW - Iowa State University; Bolon, Yung Tsi; JOSEPH, BINDU - University Of California; DIERS, BRIAN - University Of Illinois; FARMER, ANDREW - National Center For Genome Research; Weeks, Nathan; MUEHLBAUER, GARY - University Of Minnesota; Nelson, Rex; Grant, David; SPECHT, JAMES - University Of Nebraska; Graham, Michelle; Cannon, Steven; MAY, GREGORY - National Center For Genome Research; Vance, Carroll; Shoemaker, Randy

Submitted to: Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/27/2010
Publication Date: 1/15/2011
Citation: Woody, J.L., Severin, A., Bolon, Y.E., Joseph, B., Diers, B.W., Farmer, A.T., Weeks, N.T., Muehlbauer, G.J., Nelson, R., Grant, D.M., Specht, J.E., Graham, M.A., Cannon, S.B., May, G.D., Vance, C.P., Shoemaker, R.C. 2011. Gene expression patterns are correlated with genomic and genic structure in soybean. Genome. 54(1):10-18.

Interpretive Summary: Reports of differences in the size of gene messages relative to how broadly or narrowly the genes are expressed have been contradictory. Similar contradictory information relative to how much gene message is produced has been reported. This has created confusion as to the possible role of structural differences in genes in patterns of gene expression. In this study the authors used gene expression patterns of all of the hereditary material of soybean for 14 tissues and stages of expression. They considered depth and breadth of gene expression simultaneously. In this way they were able to explain the seemingly conflicting reports. Opposite results were seen depending upon level of expression relative to breadth of expression (expressed in all 14 tissues, or only a few). These results clarify and explain previous reports. This study clearly shows that plants and animals do not differ in their relationship between gene message size and levels of expression. This information is important to geneticists studying the evolution and function of genes.

Technical Abstract: Studies have indicated that exon and intron size, and intergenic distance are correlated with gene expression levels and expression breadth. Previous studies on these correlations in plants and animals have been conflicting. In this study next-generation sequence data of the soybean transcriptome was generated and analyzed from 14 unique tissues or stages of development. Four categories of genes were considered based upon expression parameters; 1) genes that had low expression levels and were narrowly expressed (one tissue only), 2) genes that had low expression levels and were broadly expressed (expressed in all 14 tissues), 3) genes that had high expression levels and were narrowly expressed, and 4) genes that had high expression levels and were broadly expressed. We also considered low, intermediate, and high levels of expression in a linear correlation with one to 14 tissues. Our results revealed a novel dichotomy. At the low expression level an increase in expression breadth correlated with an increase in transcript size due to an increase in the number of exons and introns. No significant changes in intron or exon sizes were noted. Conversely, genes expressed at the intermediate to high expression levels displayed a decrease in transcript size as their expression breadth increased. This was due to smaller introns and exons, with no significant change in the number of introns and exons. Taking advantage of the known gene space of soybean, we evaluated the positioning of genes and found significant clustering of similarly expressed genes and regions with multiple clusters of similarly expressed genes. Identifying the correlations between the physical parameters of the individual genes can lead to uncovering the possible role of regulation due to nucleotide composition and could have potential impacts in identifying the unknown role of the noncoding regions.