Statistical analysis of fractionation resistance by functional category and expression

Abstract

Background: The current literature establishes the importance of gene functional category and expression in promoting or suppressing duplicate gene loss after whole genome doubling in plants, a process known as fractionation. Inspired by studies that have reported gene expression to be the dominating factor in preventing duplicate gene loss, we analyzed the relative effect of functional category and expression.

Methods: We use multivariate methods to study data sets on gene retention, function and expression in rosids and asterids to estimate effects and assess their interaction.

Results: Our results suggest that the effect on duplicate gene retention fractionation by functional category and expression are independent and have no statistical interaction.

Conclusion: In plants, functional category is the more dominant factor in explaining duplicate gene loss.

Keywords: Angiosperms; Expression level; Gene loss; Gene ontology; Whole genome duplication.

Fig. 1

Whole genome duplication history. Star symbols mean whole genome triplication events while triangle symbols are duplication events [3, 20, 30]. Phylogeny branch lengths not to scale

Fig. 2

Based on retention of paralogs resulting from ancient polyploidization in three rosids and three asterids. Retained genes identified in homeologous syntenic blocks detected by SynMap [17, 18]. “Increase in fractionation resistance” ranges from 1 (singleton in all three species) to 4 (three paralogs retained in all species). “Normalized proportions” measures how many of the gene paralogy groups with a given fractionation resistance are annotated by a specific Gene Ontology (GO) term. E.g., in the rosids, 80% of the paralogy groups with fractionation resistance 1 are annotated with the GO term “Cellular Process”. From [14], Figure 3

Fig. 3

The Paramecium genes are filtered by GO terms before putting inside the expression bins. The Y-axis describes the retention rate of genes inside the expression bins. From [16], Figure S3

Fig. 4

Number of gene families of in each fractionation resistance categories. “All singletons” have retention index of zero, “mostly singletons” have retention index of one, “mostly duplicates” have retention index of two, and “all duplicates” have retention index of three

Fig. 5

Example of nested structure of GO terms. Starting at a low-level GO term “protein secretion”, it is inherited by two higher GO terms “secretion by cell” and “protein transport”. After a few more levels of GO terms (represented by dashed lines), the starting GO term is now inheriting two high level terms “cellular process” and “localization”. These high level terms are then linked to the root term, “biological process”. There are three root terms in gene ontology, they are “biological process”, “cellular component”, and “molecular function” [27]

Fig. 6

Summary of average retentions indices in grape and tomato. Each functional category has two data points: average retention index under low expression (LowExp) and average retention index under high expression index (HighExp)

Fig. 7

Tukey’s honest significant difference test. The horizontal bar indicate the Tukey test statistics (which include corrections for multiple comparisons) of the estimated difference between labelled categories. The vertical lines indicate the 95% confidence interval. In both grape and tomato, category Z2 and Z1, in red, are not significantly different from each other. In grape, category Z3 and Z2, in light blue is not highly significant (adjusted p-value is 0.06887)