Molecular evolution and optimization of the translation machinery

翻译是连接基因型和表型的关键环节。准确和有效的翻译起始和延伸是包括病毒在内的生物体成功的必要条件。我将在这个项目中解决的研究问题包括：1。翻译起始：在酵母等模式生物中，5' UTR应该具有哪些序列特征来提高翻译起始的效率（通过核糖体密度测量）？5' UTR中翻译相关基序是如何随着翻译起始因子和多聚A结合蛋白的变化而进化的？5' UTR中调控翻译起始的基序是什么？在哪里？翻译延伸：许多实验研究监测翻译效率后替换同义密码子或最优选或最不优选的密码子。为什么这些实验经常有相互矛盾的结果，例如，一些发现用优化的密码子显著提高了蛋白质的产量，而另一些发现没有改善？这些研究忽略了翻译起始，然而密码子优化只有在翻译起始有效的情况下才能导致蛋白质产量的提高。我们假设，那些发现优化密码子没有改善蛋白质生产的研究涉及低或中等翻译起始效率的mrna，并将通过使用最近可用的核糖体装载数据来验证这一假设。3. 研究翻译的生物信息学工具：我们应该如何开发和改进生物信息学工具，以更好地量化调控基序（目前主要通过位置权重矩阵和隐马尔可夫模型的轮廓来表征）和密码子使用偏差（如密码子适应指数、有效密码子数等），从而促进翻译起始、延伸和终止的研究？例如，目前的密码子使用偏差指数不能校正背景核苷酸频率，也不能区分突变和选择的影响。我们的研究将为从事蛋白质生产的制药行业做出贡献。

Translation is the key link between genotype and phenotype. Accurate and efficient translation initiation and elongation are essential for the success of living organisms including viruses. Research questions I will address in this project include:1. Translation initiation: What sequence features should the 5' UTR have to increase the efficiency of translation initiation (measured by ribosomal density) in model organisms such as the yeast? How do translation-related motifs in 5' UTR evolve in response to changes in translation initiation factors and poly(A) binding protein? What and where are the motifs in 5' UTR that modulate translation initiation?2. Translation elongation: Many experimental studies have monitored translation efficiency after replacing synonymous codons either to the most preferred or the least preferred codons. Why do these experiments often have contradictory results, e.g., some finding dramatically improved protein production with optimized codons and some finding no improvement? These studies ignored translation initiation, yet codon optimization can lead to improved protein production only when translation initiation is efficient. We hypothesize that those studies finding no improved protein production with optimized codons involve mRNAs of low or mediocre translation initiation efficiency and will test the hypothesis by using ribosomal loading data available recently. 3. Bioinformatics tools for study translation: How should we develop and improve bioinformatic tools for better quantification of regulatory motifs (currently characterized mainly by position weight matrix and profiles from hidden Markov models) and codon usage bias (e.g., codon adaptation index, effective number of codons, etc.) in order to facilitate studies on translation initiation, elongation and termination? For example, current codon usage bias indices do not correct for background nucleotide frequencies and do not distinguish the effect of mutation and selection.Our research will contribute to pharmaceutical industry engaged in protein production.