Molecular biology and coevolution

点击翻译按钮获取中文摘要

Efficient biosynthesis depends much on efficient RNA processing and translation. My research program focuses on 1) coevolution between protein-coding genes and translation machinery in bacteria, and 2) intron-splicing in eukaryotes. In bacterial translation, coevolution occurs between 1) Shine-Dalgarno (SD) sequences on mRNA and anti-SD (aSD) sequences on small subunit (SSU) rRNA, 2) sense codon usage and differential tRNA abundance, and 3) stop codon usage and differential abundance of release factors (RF1 and RF2). All these involve a motif signal and its decoders, i.e., SD decoded by aSD, sense codon by tRNA, and stop codon by release factors. Coevolution between the signal motifs and their decoders are confounded by many other factors. SDs, as well as start and stop codons, can all be embedded in a secondary structure and become invisible to their respective decoders. The effect of selection is more visible in highly expressed genes than in lowly expressed genes. Requirement of SD/aSD pairing differs between the first gene in an operon and the following genes. Different nucleotides downstream of stop codons affect translation terminate efficiency. Stop codon can be misread by tRNAs and sense codons misread by release factors. We need accurate quantification of motif signals and decoders to achieve an integrated understanding of their relationships. We have been using RNA-Seq data to characterize the abundance of mRNA, tRNA and SSU rRNA with different 3' tails generated by continuous degradation of 3' exoribonucleases, and statistical modeling to quantify their relationships Intron splicing signals consist of 5' and 3' splice sites and branchpoint sites, as well as exonic and intronic splice enhancers. Their signal strength also depends on secondary structure, flanking nucleotides, and gene expression. Many yeast spliceosomes are recruited by intronless mRNAs because these mRNAs harbor false splice site signals, although such false signals are selected against in highly transcribed genes. We will 1) characterize splicing signal strength by using position weight matrix and self-organizing map algorithms, 2) measure splicing efficiency (SE) by using RNA-Seq data to obtain abundance of exon-exon junctions (AEE), exon-intron junctions (AEI), and total mRNA (AT) so that SE = AEE/AT and AT = AEE + AEI, and 3) model how SE is affected by factors such as splicing strength, secondary structure, presence of splicing enhancers, etc. We will apply this method first to yeast (Saccharomyces cerevisiae) where most intron-containing genes have just one intron, and then extend the method to study genes with multiple introns in multicellular eukaryotes. Our research will not only advance knowledge of translation and splicing, but also lead to better design of biosynthesis in industry. Two PhD, 4 MSc and 4 undergrad student currently participate in the program.