Global analysis of human duplicated genes reveals the relative importance of whole-genome duplicates originated in the early vertebrate evolution

Abstract

Background: Gene duplication is a genetic mutation that creates functionally redundant gene copies that are initially relieved from selective pressures and may adapt themselves to new functions with time. The levels of gene duplication may vary from small-scale duplication (SSD) to whole genome duplication (WGD). Studies with yeast revealed ample differences between these duplicates: Yeast WGD pairs were functionally more similar, less divergent in subcellular localization and contained a lesser proportion of essential genes. In this study, we explored the differences in evolutionary genomic properties of human SSD and WGD genes, with the identifiable human duplicates coming from the two rounds of whole genome duplication occurred early in vertebrate evolution.

Results: We observed that these two groups of duplicates were also dissimilar in terms of their evolutionary and genomic properties. But interestingly, this is not like the same observed in yeast. The human WGDs were found to be functionally less similar, diverge more in subcellular level and contain a higher proportion of essential genes than the SSDs, all of which are opposite from yeast. Additionally, we explored that human WGDs were more divergent in their gene expression profile, have higher multifunctionality and are more often associated with disease, and are evolutionarily more conserved than human SSDs.

Conclusions: Our study suggests that human WGD duplicates are more divergent and entails the adaptation of WGDs to novel and important functions that consequently lead to their evolutionary conservation in the course of evolution.

Fig. 1

Functional similarity between human small-scale and whole-genome duplicate pairs. The SSDs are represented in brick red and WGDs are represented in blue. The red and blue lines represent the mean functional similarity of SSD and WGD pairs, respectively. The black line represents the mean functional similarity of all human duplicates. The functional similarities between different dN ranges were calculated using both GO domains. a. Biological Process and b. Molecular Function (For every dN range, P < 0.05). For exact P-values, refer Table 1

Fig. 2

Subcellular co-localization between human small-scale and whole-genome duplicate pairs. The SSDs are represented in brick red and WGDs are represented in blue. The red and blue lines represent the mean functional similarity of SSD and WGD pairs, respectively. The black line represents the mean functional similarity of all human duplicates (For every dN range, P < 0.05). For exact P-values, see Table 1

Fig. 3

Differences in gene expression correlation between human small-scale and whole-genome duplicate pairs. The gene expression correlation values of SSD and WGD pairs were calculated using RNA-seq gene expression data from a. Human Protein Atlas and b. Expression Atlas. The SSDs are represented in brick red and WGDs are represented in blue. The red and blue lines represent the mean gene expression correlation of SSD and WGD pairs, respectively. The black line represents the mean of gene expression correlation of all human duplicated gene pairs. (For every dN range, P < 0.05). Exact P-values are provided in Table 1

Fig. 4

Differences in evolutionary rates of human small-scale and whole genome duplicates using Mouse (Mus musculus) and Chimpanzee (Pan troglodytes) one-to-one orthologs. Both the dN (a) and the $\frac{dN}{dS}$ ratios (b) were used as the measurements of evolutionary rate. The SSDs are represented in brick red and WGDs are represented in blue. Exact P-values are provided in the figure and in Table 2

Fig. 5

Multifunctionality of human small-scale and whole-genome duplicates: a. Using their association with unique GO-Biological Processes. b. Using the number of Pfam domains. The SSDs are represented in brick red and WGDs are represented in blue. Exact P-values are provided in the figure