GLYCO, A HIGHLY EXPRESSIVE ONTOLOGY FOR THE GLYCOBIOLOGY DOMAIN

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得了主要资金， 因此可以在其他CRISP条目中表示。所列机构为 研究中心，而研究中心不一定是研究者所在的机构。 GlycO是一个高度表达的本体，它体现了聚糖结构的知识以及聚糖结构与其参与生物过程之间的关系。与GlycO，我们的目标是在糖生物学域，包括一个强大的模式和大型知识库的一般描述。该模式概念性地定义了类（例如，“包含所有N-聚糖的类”），特定实例被分配给该类并且知识库由实例组成（例如，特定的聚糖结构）和实例之间的特定关系。该模式允许推理的概念，利用Web本体语言OWL-DL（基于描述逻辑）的关系的限制。这为知识库的自动填充提供了基础，这是一个添加和分类新实例的过程。填充GlycO知识库所需的信息可以从几个部分重叠的来源自动提取，包括基因和基因组的京都百科全书（KEGG）、Glycosciences.de数据库（SweetDB）和复杂碳水化合物结构数据库（CARBBANK）。为了避免相同结构的多个条目，应用了转换和消歧技术。最终目标是生成一个大型本体，可用于注释，检索和处理有关聚糖结构-功能关系的信息，并发现该信息中隐含的知识。在GlycO中，结构信息在实例级模块化。也就是说，结构由可以在不同化学环境中重复使用的规范构建块组成。更大的结构（例如，聚糖（“聚糖”）由较小的规范结构单元（carbohydrate_residues）组成，这些规范结构单元又由甚至更小的规范结构单元（carbohydrate_residue_atoms）组成。结构单元包含特定的结构信息，例如碳水化合物残基的绝对构型、环形式和端基异构体构型，以及背景信息，例如残基在聚糖中的位置。对于在其生物合成中具有显著重叠的聚糖（诸如N-聚糖）的集合，每个组成规范残基体现可以与其与生物合成酶和其他生物化学实体的相互作用相关的背景信息。因此，简单地列出组成聚糖的残基提供了聚糖结构的描述和生物过程的隐含描述（例如，生物合成，催化剂，信号转导），它参与。在GlycO中，典型残基之间的连接表示为成熟节点。这种边到节点的提升通常被称为关系的“具体化”。这种模式允许具体化的链接以层次术语来指定。也就是说，两个典型残基之间的连接（例如，近端核-GlcpNAc残基和Asn残基）明确地体现了两个原子之间的更细粒度的连接（即，典型碳水化合物_residue_C1_atom和典型氨基酸_residue_N4_atom）。这些链接本身就是规范对象，可以以模块化的方式重用。例如，特定N-聚糖和给定肽之间的连接对应于更高水平的连接，其进而体现了典型核心-GlcNAc残基和典型Asn残基之间的典型连接。因此，模式提供了一个复杂结构的完整的，明确的规范，只需列出其最高级别的组件和链接。这不仅提供了增强的可计算性和逻辑一致性，还提供了可以与特定进程相关联的新的对象（链接）类。例如，上述连接是生物合成形成的，并由特定的酶（分别为寡糖基转移酶和PNGase-F）水解。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. GlycO is a highly expressive ontology that embodies knowledge of glycan structure and the relationships between the structure of glycans and their participation in biological processes. With GlycO we are aiming at a general description of the glycobiology domain that consists of a robust schema and large knowledgebase. The schema conceptually defines classes (e.g., "the class containing all N-glycans") to which specific instances are assigned and the knowledgebase is comprised of instances (e.g., a specific glycan structure) and specific relationships between instances. The schema allows reasoning about the concepts by exploiting the Web Ontology Language OWL-DL (based on Description Logic) to place restrictions on relationships. This provides the basis for automated population of the knowledge base, a process in which new instances are added and classified. The information needed to populate the GlycO knowledgebase can be automatically extracted from several partially overlapping sources, including the Kyoto Encyclopedia of Genes and Genomes (KEGG), Glycosciences.de databases (SweetDB), and the Complex Carbohydrate Structural Database (CARBBANK). In order to avoid multiple entries of identical structures, transformation and disambiguation techniques are applied. The ultimate goal is to generate a large ontology that can be used for the annotation, retrieval and processing of information regarding glycan structure-function relationships and the discovery of the knowledge implicit in that information. In GlycO, structural information is modularized at the instance level. That is, structures are composed of canonical building blocks that can be reused in different chemical contexts. Larger structures (e.g.,"glycans") are composed of smaller canonical building blocks (carbohydrate_residues), which are, in turn, composed of even smaller canonical building blocks (carbohydrate_residue_atoms). The building blocks contain specific structural information, such as the absolute configuration, ring form, and anomeric configuration of a carbohydrate residue, and contextual information, such as the location of the residue in the glycan. For collections of glycans, such as N-glycans, that have significant overlap in their biosynthesis, each of the constituent canonical residues embodies contextual information that can be correlated to its interactions with biosynthetic enzymes and other biochemical entities. Thus, simply listing the residues that make up a glycan provides a description of the glycan structure and an implicit description of the biological processes (e.g., biosynthesis, catabolism, signal transduction) that it participates in. In GlycO, the links between canonical residues are expressed as full-fledged nodes. Such promotion of an edge to a node is often referred to as "reification" of a relationship. This schema allows the reified links to be specified in hierarchical terms. That is, the link between two canonical residues (e.g., the proximal core ¿-GlcpNAc residue and an Asn residue) explicitly embodies a more finely grained link between two atoms (i.e., the canonical carbohydrate_residue_C1_atom and the canonical amino_acid_residue_N4_atom). These links are themselves canonical objects that can be reused in a modular fashion. For example, the connection between a particular N-glycan and a given peptide corresponds to a higher level link, which, in turn, embodies the canonical link between the canonical core ¿-GlcNAc residue and the canonical Asn residue. Thus, the schema provides for the complete, unambiguous specification of a complex structure simply by listing its highest level components and links. This not only provides enhanced computability and logical consistency, it provides new classes of objects (links) that can be associated with particular processes. For example, the link described above is biosynthetically formed and hydrolyzed by specific enzymes (ologosaccharyl transferase and PNGase-F, respectively).