Ongoing large-scale genome sequencing projects are forecasting a data deluge that will almost certainly overwhelm current analytical capabilities of evolutionary genomics. In contrast to population genomics, there are no standardized methods in evolutionary genomics for extracting evolutionary and functional (e.g. gene-trait association) signal from genomic data. Here, we examine how current practices of multi-species comparative genomics perform in this aspect and point out that many genomic datasets are under-utilized due to the lack of powerful methodologies. As a result, many current analyses emphasize gene families for which some functional data is already available, resulting in a growing gap between functionally well-characterized genes/organisms and the universe of unknowns. This leaves unknown genes on the 'dark side' of genomes, a problem that will not be mitigated by sequencing more and more genomes, unless we develop tools to infer functional hypotheses for unknown genes in a systematic manner. We provide an inventory of recently developed methods capable of predicting gene-gene and gene-trait associations based on comparative data, then argue that realizing the full potential of whole genome datasets requires the integration of phylogenetic comparative methods into genomics, a rich but underutilized toolbox for looking into the past.

中文翻译：

---

在大规模基因组测序时代，需要新的系统发育方法来了解基因功能。


正在进行的大规模基因组测序项目预测数据洪流几乎肯定会压倒进化基因组学当前的分析能力。与群体基因组学相反，进化基因组学中没有标准化的方法来从基因组数据中提取进化和功能（例如基因-性状关联）信号。在这里，我们研究了多物种比较基因组学的当前实践在这方面的表现，并指出由于缺乏强大的方法论，许多基因组数据集未得到充分利用。因此，当前的许多分析都强调已经有一些功能数据的基因家族，导致功能良好表征的基因/生物体与未知世界之间的差距越来越大。这使得未知基因处于基因组的“黑暗面”，这个问题不会通过对越来越多的基因组进行测序来缓解，除非我们开发出工具以系统的方式推断未知基因的功能假设。我们提供了最近开发的能够基于比较数据预测基因-基因和基因-性状关联的方法的清单，然后认为，实现全基因组数据集的全部潜力需要将系统发育比较方法整合到基因组学中，这是一个丰富但未充分利用的工具箱为了回顾过去。