水环境病原菌对人类和水生动物的健康以及水产品生物安全带来了重大威胁，是公共卫生、水产养殖、食品安全等行业的重点监测对象。然而水环境病原菌数据库建设相对滞后，相关数据库分散在临床医学和水产动物病害等领域，且缺乏信息交流与融合，完整性仅限于各自独立的学科，不能满足区域尺度或生态学视角下，大规模水源性病原鉴定及生物安全评价等高通量监测的需求。因此，本研究通过整理人类介水传染病、水生动物、哺乳动物、植物和跨宿主疾病等7大类细菌病原信息，构建多线程可调度通讯模型和全局序列匹配算法，开发了水环境细菌病原数据库 (DPiWE，dayuz.com)。DPiWE收集了14门、27纲、54目、116科、221属、1 097种、9 070株细菌病原的物种分类、16S rRNA基因、宿主 (195种) 和感染类型 (21种) 信息。并在Web端实现信息检索、序列比对和注释结果可视化等功能。案例分析显示，DPiWE构建的系统发育网络，清晰地将养殖环境菌株DS10?D19划分为鳆发光杆菌；用DPiWE对海水混养系统细菌高通量测序结果进行注释，揭示3种养殖动物病原分布具有明显差异，患病组水体有传播人体和鱼类共患病病原的风险。DPiWE及配套分析流程可为水环境生物安全高通量评价、渔业生态健康维护和水产动物病害个性化防治提供新的思路和数据基础。

Pathogenic bacteria in the water environment are mainly monitored in the public health, food safety, aquaculture and other industries due to their major threats to the health of humans and aquatic animals, and the biosafety of aquatic products. However, pathogenic database involved in water environment pathogen is mainly constructed according to independent disciplines, and scattered in the fields of clinical medicine and aquatic animal diseases, which can no longer meet the high-throughput identification and biosafety evaluation of pathogenic bacteria involved in the water environment in the regional scale or ecological perspective. In this study, a database of pathogenic bacteria involved in water environment (DPiWE) was constructed by collecting the taxonomic information of pathogenic bacteria from humans, aquatic animals, mammals, plants, and cross-host comorbidities. A multi-threaded schedulable communication model and a multi-task mode global sequence matching algorithm were developed to construct DPiWE. The database collected 9 070 pathogenic bacteria strains, which belong to 14 phyla, 27 classes, 54 orders, 116 families, 221 genera and 1 097 species. The corresponding 16S rRNA gene sequences, host information and infection types of these strains were also collected in DPiWE. This database was deployed at a website (http://dayuz.com/) with the functions including web user management, pathogenic information retrieval, sequence upload, storage and alignment, and visualization of annotation result. Two examples were used to test the functions of DPiWE. The first example showed that, DPiWE can accurately construct a phylogenetic network of an unidentified bacterium (strain DS10?D19) isolated from cultural seawaters, according to its 16S rRNA gene sequence, and identified it as Photobacterium leiognathid. The result of network also showed that the network structure of strain DS10?D19 was similar to P. leiognathid and P. angustum. The second example showed that the compositions of pathogens in the intestines of three mariculture animals were significantly different through annotating the high-throughput sequencing data using DPiWE, and the rearing water in diseased groups had potential risk of spreading the comorbid pathogenic bacteria of human and fish. The DPiWE and its supporting data analysis process can provide new ideas and data foundations for high-throughput detection of the biosafety of water environment, protecting health of fishery ecology, and controlling diseases of aquatic animals, in the future.

S917.1

国家重点研发计划(2016YFC1402205)；国家自然科学基金(31672658)；宁波市农业重大专项(2017C110001)