neXtA5 services

Document triage service

The triage function is an essential element of the curation system directly integrated into neXtA5, but also available through a special API developed on a HTTP POST protocol. This service provides a ranked list of documents for a gene-centric curation request.

The base URL is :

http://candy.hesge.ch/neXtA5/webservice/ranking/json?

Parameters

gene parameter is mandatory as it specifies the name of the researched entity.

The following parameters are available :

• axis : to define the prioritization dimension for the curation (DISEASE|GO_BP|GO_CC|GO_MF|INTERACTION), default is DISEASE
• mode : to define the retrieval mode of the search engine (Vectorial|Boolean), default is Vectorial
• synonyms : to list the synonyms of the searched gene (e.g. "AGMX1,ATK,BPK" for NX_Q06187:BTK)
• dateMin : the publication dates prior to this year will be excluded from the search, default is 2000
• dateMax : the publication dates next to this year will be excluded from the search, default is 2019
• excludedPublication : to list the identifiers of non-relevant publications to filter (e.g. "17360955,20105294" as PMIDs)
• excludedConcept : to list the identifiers of the non-relevant concept to filter out of the ranking strategy (e.g. "C2916,C9305" respectively to exclude the occurrences of "Carcinoma" and "Cancer" from the scoring)
• requested : to define the maximum number of retrieved citations, default is 50 and maximum is 1000

Example

The following URL provides a list of publications corresponding to the kinase IRAK4, ranked for their relevance to the Biological Process curation.

http://candy.hesge.ch/neXtA5/webservice/ranking/json?gene=IRAK4&axis=GO_BP

Output

The output is provided in JSON format :

[ { "abstract":"TLR signal transduction involves a MyD88-mediated pathway [...].", "title":"Role of IRAK4 and IRF3 in the control of intracellular infection with Chlamydia pneumoniae.", "biomedScore":12.829, "neXtA5Score":17.5461, "pubDate":"2007", "publicationId":"17360955", "statements":6 }, ... ]

Output fields for each document are :

• "publicationId" : the identifier of the retrieved publication
• "title" : the title of the publication
• "abstract" : the abstract of the publication
• "pubDate" : the publication year
• "statements" : the number of annotations found in the text
• "biomedScore" : the score provided by the search engine, depending of the initial query (only with the vector space model)
• "neXtA5Score" : the final score calculated by neXtA5, depending of the curation query

Getting Help

You can contact us for questions specifically relating to this service.