What Signatures Dominantly Associate with Gene Age?

Abstract

As genes originate at different evolutionary times, they harbor distinctive genomic signatures of evolutionary ages. Although previous studies have investigated different gene age-related signatures, what signatures dominantly associate with gene age remains unresolved. Here we address this question via a combined approach of comprehensive assignment of gene ages, gene family identification, and multivariate analyses. We first provide a comprehensive and improved gene age assignment by combining homolog clustering with phylogeny inference and categorize human genes into 26 age classes spanning the whole tree of life. We then explore the dominant age-related signatures based on a collection of 10 potential signatures (including gene composition, gene length, selection pressure, expression level, connectivity in protein–protein interaction network and DNA methylation). Our results show that GC content and connectivity in protein–protein interaction network (PPIN) associate dominantly with gene age. Furthermore, we investigate the heterogeneity of dominant signatures in duplicates and singletons. We find that GC content is a consistent primary factor of gene age in duplicates and singletons, whereas PPIN is more strongly associated with gene age in singletons than in duplicates. Taken together, GC content and PPIN are two dominant signatures in close association with gene age, exhibiting heterogeneity in duplicates and singletons and presumably reflecting complex differential interplays between natural selection and mutation.

Systematic Analysis of Mouse Genome Reveals Distinct Evolutionary and Functional Properties Among Circadian and Ultradian Genes

“…The results of our study are globally in agreement with the general notion that ancient genes have a propensity for non-biased codon usage, as detected by a recent analysis on the human genome showing that gene age has a significant inverse correlation with SCU bias (with older genes having Codon Deviation Coefficient values toward 0) ( Yin et al, 2016 ). This issue is further complicated when gene expression is taken into account.…”

The borders ofcis-regulatory DNA sequences harbor the divergent transcription factor binding motifs in the human genome

“…(A) Comparison of mean MPIs between DHS-center (purple, left violin plot) and DHS-edge (blue, right violin plot) in six age categories of genes. The age of human genes arose at different evolutionary time were identified by combining homolog clustering with phylogeny inference according to Yin et al 2016 [47]. Accordingly, category 1 to 6 denoted the Primate origin (youngest genes), Mammalia, Vertabrata, Metazoan, Eukaryota; and Cellular organism (oldest genes), respectively.…”

Borders of Cis-Regulatory DNA Sequences Preferentially Harbor the Divergent Transcription Factor Binding Motifs in the Human Genome

“…The ages of human genes arising at different evolutionary times were identified by combining homolog clustering with phylogeny inference, as described in recent literature ( Yin et al, 2016 ). Gene category 1 denoted Primates origin, i.e., the youngest genes; category 2 denoted Mammalia origin; category 3 denoted Vertebrata origin; category 4 denoted Metazoan origin; category 5 denoted Eukaryota origin; and category 6 denoted cellular-organism origin, i.e., the oldest genes.…”