《seuwest.amazonaws.com》由会员分享，可在线阅读，更多相关《seuwest.amazonaws.com（130页珍藏版）》请在金锄头文库上搜索。
1、Protocol S1. Assembly of the supplementary dengue databaseS1.1 OverviewThe dengue database comprises occurrence data linked to point or polygon locations, derived from peer-reviewed literature. Data sources are described in full here. To collate the peer-reviewed database, literature searches were u
2、ndertaken using the major search engines and the resulting articles were manually reviewed. Geo-spatial information for 7,419 occurrences of dengue was extracted using site descriptions within 2,838 of the articles and recorded as specific coordinates or by a polygon definition when this was the onl
3、y information available. All data entries from both data sources were manually checked by the authors and then underwent a series of routine quality control procedures to ensure correct geo-positioning.S1.2 Peer-reviewed Literature SearchPubMed (http:/www.ncbi.nlm.nih.gov/entrez/query.fcgi) 1920 to
4、2009 was searched using term “dengue”. The MESH term technology used in the PubMed citation archive ensured all pseudonyms were automatically included (http:/www.nlm.nih.gov/mesh/2008/MBrowser.html) in the searches. The same process was repeated for ISI Web of Science (http:/wok.mimas.ac.uk) and Pro
5、MED (http:/www.promedmail.org). The searches were last updated on 8th February 2012. No language restrictions were placed on these searches, however, only those citations with a full title and abstract were downloaded into Endnote version 10 (Thomson ResearchSoft, USA, http:/). A total of 5,876 refe
6、rences were retrieved and 2,883 unique articles were identified as potentially containing useable location data. The full articles were obtained for 2,838 or 98.4% and the references are listed in S1.6.In-house language skills allowed processing of all English, French, Portuguese and Spanish article
7、s. Russian articles with data were translated and included in the database. We were unable to extract information from a small number of Turkish, Polish, Hebrew, Italian, German and Chinese articles.S1.3 Disease data and imported casesConfirmed occurrences of dengue virus transmission, dengue haemor
8、rhagic fever or a specific dengue serotype found within the peer reviewed literature were recorded as a dengue occurrence datapoint. A report of multiple cases (i.e. an outbreak) was reported as a single occurrence datapoint, however, later outbreaks at the same location were recorded as new occurre
9、nce datapoints. Reports of autochthonous (locally transmitted) cases or outbreaks were entered as an occurrence within the country. If imported cases were reported with information on the site of contagion, they were recorded as an occurrence from the country of contagion within the database. If imp
10、orted cases were reported with no information about the site of contagion they were not entered into the database. If an imported case led to an outbreak/local transmission within the recipient country and location information was available for the site of initial contagion and the site of the outbr
11、eak, this was recorded as two occurrences; one in the country of contagion and one in the country where the outbreak occurred.Data from both sources was entered into a bespoke PostgreSQL database that links disease data to spatial data. A link to the PDF library was also included in the database.S1.
12、4 Geo-positioning occurrence pointsAll available location information in the peer-reviewed literature was extracted from each article. The site name was used together with all contextual information provided about the study site position, for example the approximate distance from the nearest city or
13、 the district within which the site was located. The site name and contextual information were used together to locate the study site using Microsoft Encarta (Microsoft Corporation, Redmond, WA) and Google Earth 5.1. Place names are often duplicated within a country so the contextual information was
14、 used to ensure the right site was selected. Where the site name was not found, the contextual information was used to scan sites in the approximate area to check for names that had been transliterated in Microsoft Encarta/Google Earth in a different way to the published article (e.g. Imichli and Im
15、ishly). If the site was not located using Microsoft Encarta or Google Earth, a range of online georeferencing resources were tried including Getty Thesaurus (http:/www.getty.edu/research/conducting_research/vocabularies/tgn/index.html), Alexandria (http:/middleware.alexandria.ucsb.edu/client/gaz/adl
16、/index.jsp) and Falling Rain (http:/ If the study site could be georeferenced to a specific place, it was recorded as a point location. In total, 3,286 point locations from the peer-reviewed literature were included in the database at this stage. If the study site could only be identified at an administrative area level (e.g. province or district, etc), it was recorded as a polygon and the centroid (mean centre) was
