SIB Swiss Institute of Bioinformatics, Geneva, Switzerland.
Methods Mol Biol. 2024;2836:285-298. doi: 10.1007/978-1-0716-4007-4_15.
The Gene Ontology (GO) project describes the functions of the gene products of organisms from all kingdoms of life in a standardized way, enabling powerful analyses of experiments involving genome-wide analysis. The scientific literature is used to convert experimental results into GO annotations that systematically classify gene products' functions. However, to address the fact that only a minor fraction of all genes has been characterized experimentally, multiple predictive methods to assign GO annotations have been developed since the inception of GO. Sequence homologies between novel genes and genes with known functions help to approximate the roles of these non-characterized genes. Here we describe the main sequence homology methods to produce annotations: pairwise comparison (BLAST), protein profile models (InterPro), and phylogenetic-based annotation (PAINT). Some of these methods can be implemented with genome analysis pipelines (BLAST and InterPro2GO), while PAINT is curated by the GO consortium.
GO 项目以标准化的方式描述来自所有生命领域的生物体的基因产物的功能，使涉及全基因组分析的实验能够进行强大的分析。该项目利用科学文献将实验结果转换为 GO 注释，系统地对基因产物的功能进行分类。然而，为了应对并非所有基因都经过实验鉴定这一事实，自 GO 成立以来，已经开发了多种预测方法来分配 GO 注释。新基因与具有已知功能的基因之间的序列同源性有助于推测这些未鉴定基因的作用。在这里，我们描述了产生注释的主要序列同源性方法：两两比较（BLAST）、蛋白质轮廓模型（InterPro）和基于系统发育的注释（PAINT）。这些方法中的一些可以与基因组分析管道（BLAST 和 InterPro2GO）一起实现，而 PAINT 则由 GO 联盟进行整理。
