On the Evolution of the Avian Transcriptome

Sammanfattning: Change in gene expression is a powerful tool for evolution, because seemingly small expression changes can contribute important steps towards adaptation without necessarily affecting the whole organism. There is still much to learn about how gene expression evolves on genome- and population-wide levels, especially in non-model organisms. This thesis addresses some important questions in gene expression evolution via the quantitative measurement of RNA and protein levels in birds.First, I confirmed the state of incomplete dosage compensation in birds by sequencing the transcriptome of collared flycatchers (Ficedula albicollis). I showed that pleiotropy governs the evolution of expression male-bias from the Z chromosome. Sex-linked genes in females were more highly expressed than half the male expression level, indicative of a partial up-regulation. A comparison with data from ostrich (Struthio camelus), a bird with non-degenerated sex chromosomes, showed that sex-linked expression male-bias evolved following sex chromosome degradation.Second, using a combination of RNA sequencing and proteome mass spectrometry in chicken (Gallus gallus), I asked whether complete dosage compensation was achieved through regulation at translation. I showed that this was not the case and that incomplete dosage compensation extends to the protein level in birds. In addition, sex-linked genes showed more often an increased amount of regulation at translational level than autosomal genes.Third, I investigated gene expression divergence between collared and pied flycatchers (Ficedula hypoleuca) using RNA sequencing in multiple tissues and individuals. Tissues differed in the degree of expression variance and in the number of divergent genes, which I identified using expression QST. Variance within species was negatively correlated with expression breadth and protein interactivity, indicating that evolutionary constraints act predominantly within interbreeding populations. Among genes unique to one of the species, I identified one gene, DPP7, falling into a large genomic deletion fixed in pied flycatchers.Fourth, I investigated allele-specific expression (ASE) in the two flycatcher populations. ASE was identified from genetic variants within transcripts using RNA sequencing reads. We developed a Bayesian negative binomial approach that gained statistical power by estimating expression variance from combined SNPs within a transcript and overdispersion from the whole dataset.