Abstract
Gene birth is the process through which new genes appear. For a long time it was argued that the natural way of generating new genes was from copies of existing genes, and the possibility of de novo gene emergence was neglected. However, recent evidence has forced to reconsider old models and de novo gene birth gained recognition as a widespread phenomenon. De novo gene birth is the process by which a non-genic sequence is able to gain gene-like features through few mutations.
The following work is a compilation of analyses that seek to highlight the importance and prevalence of de novo gene birth in genomes, suggesting that this is a process that is present at all times and which becomes very relevant upon ecological shifts.
In the first chapter, I showed through phylostratigraphic analyses that new genes are substantially simpler than older, a trend which was consistent for several features and organisms, and suggestive of a frequent emergence of new genes through non-duplicative processes. In addition to this, I detected a strong association between gene birth and high transcriptional activity and chromosomal proximity. As part of this work, I was also able to use phylostratigraphy to evaluate a different model of gene birth, overprinting of alternative reading frames.
In the following chapters of this dissertation, I made use of high-throughput sequencing of transcriptomes and genomes to ask questions about the origin and change of genes at closer time divergences than ever before, ranging from nearly 3000 years to 10 million years of divergence. I was able to detect the theoretically predicted effects of short time scale comparisons on the rate of protein evolution. Also, I contribute evidence that genes of different ages show different selective constraints even after only a few thousand years of divergence.
Finally, in the last part of this thesis I evaluated the role of transcription in gene birth dynamics. Transcription seems to be a predominant feature of genomes, as most of the genome showed some level of transcription. In terms of de novo gene birth, I was able to identify 663 candidate loci from presence and absence of transcription. Analyses of these candidate loci indicated that gains are rather stable, meaning that subsequent losses were rarely found. In agreement with previous studies, I confirmed the role of testis as a driver of new genes.
These results indicate that transcription is not a limiting factor in the emergence of new genes, and that our knowledge about the key regulatory elements of transcription and their turnover is still limited to explain why new genes seem to arise at a higher rate than they decay.
