| This dissertation mainly uses bioinformatic and comparative analysis on eukaryotic genomes to address several important themes about protein translation.; Translation Termination is a key step of protein synthesis, but many aspects about this process remain unclear. The findings about this topic can be summarized as follows. (1) Ivanov and his coworkers proposed an interesting model for stop codon recognition in eubacteria, where nucleotide triplets in the large subunit rRNA recognize stop codons. A comprehensive evaluation was performed on this model, and the results cast considerable doubt on the model. (2) Using ciliates as a model system, a set of specific amino acid sites important for stop codon recognition in eukaryotic release factor 1 was identified, and a set of cooperative paired substitutions that may underlie stop codon reassignment was also proposed. (3) The presence of a tandem stop codon may provide a buffer in case the ribosome fails to recognize the first stop. For the first time, the presence and conservation of tandem stop codons was demonstrated in several yeast species. (4) To study tRNA molecular mimicry, a novel bioinformatic approach was developed to quantitatively study the structural similarity between two different types of biopolymers---proteins and RNA, based on the spatial distribution of conserved elements. This work may not only uncover one important architectural rule for RNA mimicry, but also calls to attention the evolutionary relationship between protein and RNA.; Dual coding regions are shared among different transcripts, but encode amino acids in different reading frames. A systematic search was performed through a set of high-quality human transcripts, showing that approximately 7% of alternatively spliced genes contain dual (multiple) coding regions. Using a comparative genomic approach, it was also found that in these regions, most secondary reading frames evolved recently in mammals.; To explore different mechanisms underlying coordinated regulation, several candidate genes were identified, where their RNA editing sites may have the potential to block putative microRNA:target gene regulation. In a second study, based on the co-occurrence of motifs in promoter and 3'UTR regions, the large-scale coordination was revealed between regulation at the transcriptional and post-transcriptional levels. |
