ABI Innovation: Beyond dN/dS: Population Genetics, Genome Structure, and Protein Structure

完整的基因组序列，包括所有蛋白质编码基因的序列，现在存在于许多不同的物种中。在这些序列中存在着物种分化的分子基础。目前还不可能系统地比较来自不同物种的同源基因，并预测哪些基因改变了功能。将开发新的机制模型，描述潜在的种群遗传、进化和生化过程以及物种之间氨基酸变化的限制，并将其应用于基因组比较，包括人类和黑猩猩之间的差异。此外，该项目将为本科生和研究生提供教育机会，并提供博士后培训。此外，将制作一个流行的视频、互动教育网站和基于海报的比较基因组学博物馆展览，向大众科学观众解释比较基因组分析的概念。该项目包括与美国自然历史博物馆和新西兰奥塔哥大学的研究人员合作。目前在比较基因组学中广泛使用的方法只是寻找改变编码氨基酸的核苷酸变化的统计过量，通过同义核苷酸变化的期望（非同义核苷酸取代率与同义核苷酸取代率的比率）进行标准化，但不考虑氨基酸生物化学或结构生物学。他们也没有考虑基因组结构（连锁）或群体遗传对氨基酸变化固定概率的考虑。本研究将开发明确考虑群体遗传模型中染色体位点连锁的方法，并以不同的方式考虑氨基酸变化的生化和结构背景。新模型最初将用于比较人类和黑猩猩基因组蛋白质编码集，随后将在系统发育背景下实施，以便比较来自多个物种的基因家族。这种建模方法的潜在下游应用，能够更好地理解谱系特异性进化的机制，包括预测蛋白质或功能发生变化的途径。这可以指出使物种独特的蛋白质和途径，也可以验证模式生物何时可能成为特定分子系统的良好模型。与此项目有关的信息可在http://www.wyomingbioinformatics.org/LiberlesGroup上获得。

Full genome sequences, including the sequence for all protein-encoding genes, now exist for many different species. Within these sequences lie the molecular underpinnings of species differentiation. Presently it is not possible to systematically compare homologous genes from different species and predict which have changed function. New mechanistic models that describe the underlying population genetic, evolutionary, and biochemical processes and constraints on amino acid change between species will be developed and applied to genome comparisons, including to the divergence between humans and chimpanzees. Additionally, the project will provide educational opportunities for undergraduate and graduate students as well as postdoctoral training. Further, a popular video, interactive educational website, and poster-based museum exhibit on comparative genomics will be produced to explain the concepts of comparative genomic analysis to a popular science audience. The project involves collaboration with the American Museum of Natural History and researchers at University of Otago in New Zealand.Current widely used approaches in comparative genomics simply look for a statistical excess of nucleotide changes that change the encoded amino acid, normalized by the expectation from synonymous nucleotide changes (the ratio of nonsynonymous to synonymous nucleotide substitution rates), but do not look at amino acid biochemistry or structural biology. They also do not consider genome structure (linkage) or population genetic considerations of the probability of fixation of an amino acid change. This research will develop approaches that explicitly consider the linkage of sites on chromosomes in a population genetic model and in different ways, also consider the biochemical and structural context of amino acid changes. The new models will be used initially in comparisons of the human and chimpanzee genome protein coding set and will subsequently be implemented in a phylogenetic context to enable comparison of gene families from multiple species. Potential downstream applications of this modeling approach that enables better mechanistic understanding of lineage-specific evolution include prediction of proteins or pathways where functions have changed. This can point to proteins and pathways that make a species unique as well as validating when model organisms are likely to be good models for particular molecular systems. Information related to this project will be available at http://www.wyomingbioinformatics.org/LiberlesGroup.