The Environmental Molecular Sciences Laboratory (EMSL), Pacific Northwest National Laboratory (PNNL), Washington, WA, 99352, USA.
Advanced Computing, Computing and Analytics Division, PNNL, Richland, WA, 99352, USA.
Sci Rep. 2019 Feb 12;9(1):1858. doi: 10.1038/s41598-019-38483-0.
Predicting phenotypic expression from genomic and environmental information is arguably the greatest challenge in today's biology. Being able to survey genomic content, e.g., as single-nucleotide polymorphism data, within a diverse population and predict the phenotypes of external traits, represents the holy grail across genome-informed disciplines, from personal medicine and nutrition to plant breeding. In the present study, we propose a two-step procedure in bridging the genome to phenome gap where external phenotypes are viewed as emergent properties of internal phenotypes, such as molecular profiles, in interaction with the environment. Using biomass accumulation and shoot-root allometry as external traits in diverse genotypes of the model grass Brachypodium distachyon, we established correlative models between genotypes and metabolite profiles (metabotypes) as internal phenotypes, and between metabotypes and external phenotypes under two contrasting watering regimes. Our results demonstrate the potential for employing metabotypes as an integrator in predicting external phenotypes from genomic information.
从基因组和环境信息预测表型表达可以说是当今生物学面临的最大挑战。能够在多样化的人群中调查基因组内容（例如单核苷酸多态性数据），并预测外部特征的表型，这代表了从个性化医疗和营养到植物育种等基因组信息学科的圣杯。在本研究中，我们提出了一种两步程序，用于弥合基因组与表型之间的差距，即将外部表型视为与环境相互作用的内部表型（如分子谱）的涌现特性。我们使用生物量积累和茎叶异速生长作为模型草柳枝稷不同基因型的外部特征，建立了基因型和代谢物谱（代谢型）之间的相关模型，以及在两种对照浇水条件下代谢型和外部特征之间的相关模型。我们的结果表明，代谢型作为一种整合因子，有可能从基因组信息中预测外部表型。
