Integrated human genome annotation: generation of a reference gene set

本提案的具体目的是在人类基因组参考序列上高精度地注释所有循证基因特征。这包括鉴定所有具有相关替代变异的蛋白质编码位点、在公共核苷酸数据库（NCBI/EMBL/DDBJ）中有转录证据的非编码位点和假基因。为了实现这一目标，我们将整合计算方法，包括最近的比较方法，专家手工注释，能够整合文献信息，以及有针对性的实验方法。基于我们对ENCODE区域的初始GENCODE注释的详尽实验和计算调查，我们有信心我们可以提供具有高特异性和敏感性的基因集，这将为其他生物学家和其他ENCODE组提供关键信息。作为这个过程的一部分，我们将清楚地标记所有明显的基因位点，根据它们可能的当前功能状态对它们进行分类，这样用户就可以知道哪些区域看起来像基因，哪些区域最有可能是假基因，哪些区域的转录证据最有可能是人造的。在定义人类基因组的蛋白质编码基因方面，有许多积极的团队在工作。本建议包括大多数这类团体，并与其他关键团体协调。至关重要的是，所有小组都带来了丰富的数据集成和评估经验，从而通过多种方法解决注释差异。这给了我们信心，通过这个综合项目，我们将能够消除许多关于人类基因组中基因及其组成外显子和转录结构精确位置的剩余不确定性。

The specific aim of this proposal is to annotate all the evidence-based gene features at high accuracy on the human genome reference sequence. This includes identifying all the protein-coding loci with associated alternative variants, non-coding loci which have transcript evidence available in the public nucleotide database (NCBI/EMBL/DDBJ) and pseudogenes. To achieve this goal we will integrate computational approaches, including recent comparative methods, expert manual annotation, able to integrate literature information, and targeted experimental approaches. Based on the exhaustive experimental and computation investigation of our initial GENCODE annotation of the ENCODE regions we are confident that we can deliver a gene set with high specificity and sensitivity that will provide critical information to other biologists and other ENCODE groups. As part of this process we will label all apparent gene loci clearly, classifying them according to their likely current functional status, so users are informed where regions that appear gene like are most likely pseudogenes or where transcript evidence is most likely artefactual. There are a number of motivated groups working in the area of defining protein coding genes for the human genome. This proposal includes most such groups and coordinates with other key groups. Critically, all the groups bring extensive experience of data integration and evaluation, leading to the resolution of annotation discrepancies by multiple approaches. This gives us confidence that through this integrated project we will be able to eliminate many of the remaining uncertainties about the precise location of genes and their component exons and transcript structure in the human genome. Genome-wide, highly accurate transcript definition will be of enormous value to the myriad of researchers working on the human genome. It will both have large cost savings worldwide due to increased specificity of reagent design and provide a more complete view of human genes, in particular those associated with disease. From this foundation, more accurate descriptions of the genetic causes of disease can be discovered.