Priya Piyush, Yadav Archana, Chand Jyoti, Yadav Gitanjali
Computational Biology Laboratory, National Institute of Plant Genome Research, Aruna Asaf Ali Marg, New Delhi, 110067 India.
Department of Plant Sciences, University of Cambridge, Downing Site, Cambridge, CB2 3EA UK.
Plant Methods. 2018 Jan 10;14:4. doi: 10.1186/s13007-017-0269-0. eCollection 2018.
Terpenoid hydrocarbons represent the largest and most ancient group of phytochemicals, such that the entire chemical library of a plant is often referred to as its 'terpenome'. Besides having numerous pharmacological properties, terpenes contribute to the scent of the rose, the flavors of cinnamon and the yellow of sunflowers. Rapidly increasing -omics datasets provide an unprecedented opportunity for terpenome detection, paving the way for automated web resources dedicated to phytochemical predictions in genomic data.
We have developed Terzyme, a predictive algorithm for identification, classification and assignment of broad substrate unit to terpene synthase (TPS) and prenyl transferase (PT) enzymes, known to generate the enormous structural and functional diversity of terpenoid compounds across the plant kingdom. Terzyme uses sequence information, plant taxonomy and machine learning methods for predicting TPSs and PTs in genome and proteome datasets. We demonstrate a significant enrichment of the currently identified terpenome by running Terzyme on more than 40 plants.
Terzyme is the result of a rigorous analysis of evolutionary relationships between hundreds of characterized sequences of TPSs and PTs with known specificities, followed by analysis of genome-wide gene distribution patterns, ontology based clustering and optimization of various parameters for building accurate profile Hidden Markov Models. The predictive webserver and database is freely available at http://nipgr.res.in/terzyme.html and would serve as a useful tool for deciphering the species-specific phytochemical potential of plant genomes.
萜类碳氢化合物是最大且最古老的一类植物化学物质，以至于植物的整个化学文库常被称为其“萜类组”。除了具有众多药理特性外，萜类物质还赋予了玫瑰的香气、肉桂的风味以及向日葵的黄色。快速增长的组学数据集为萜类组检测提供了前所未有的机遇，为致力于基因组数据中植物化学物质预测的自动化网络资源铺平了道路。
我们开发了Terzyme，这是一种预测算法，用于对萜烯合酶（TPS）和异戊烯基转移酶（PT）进行鉴定、分类以及将宽泛的底物单元分配给它们，已知这些酶能在整个植物界产生萜类化合物巨大的结构和功能多样性。Terzyme利用序列信息、植物分类学和机器学习方法在基因组和蛋白质组数据集中预测TPS和PT。通过在40多种植物上运行Terzyme，我们证明了当前鉴定出的萜类组有显著富集。
Terzyme是对数百个具有已知特异性的TPS和PT特征序列之间的进化关系进行严格分析的结果，随后分析全基因组基因分布模式、基于本体的聚类以及优化各种参数以构建准确的轮廓隐马尔可夫模型。预测性网络服务器和数据库可在http://nipgr.res.in/terzyme.html免费获取，将成为解读植物基因组物种特异性植物化学潜力的有用工具。
