DAVIDQuery: DAVIDQuery
In DAVIDQuery: Retrieval from the DAVID bioinformatics data resource into R

Description Usage Arguments Details Value Author(s) References See Also Examples

Launch a query against DAVID, the Database for Annotation, Visualization and Integrated Discovery. Return the results into an R object.

1
2

DAVIDQuery(ids = "O00161,O75396", type = "UNIPROT_ACCESSION", annot, tool, URLlengthLimit = 2048,
details = TRUE, verbose = FALSE, writeHTML = FALSE, testMe = FALSE, graphicMenu = FALSE,  formatIt = TRUE)

`ids`	IDs for desired objects, as a character vector or as a single string with ids separated by ",". To be passed to the DAVID website, the format has to be the latter.
`type`	Type of input ids. If missing, a menu is contructed from the R object DAVIDTypeChoices. See http://david.abcc.ncifcrf.gov/content.jsp?file=DAVID_API.html#input_list for up-to-date choices.
`annot`	Type of annotation requested. If missing, a menu is contructed from the R object DAVIDAnnotChoices. See http://david.abcc.ncifcrf.gov/content.jsp?file=DAVID_API.html#input_list for up-to-date choices.
`tool`	Type of gene tool to use. If missing, a menu is contructed from the R object DAVIDToolChoices. See http://david.abcc.ncifcrf.gov/content.jsp?file=DAVID_API.html#input_list for up-to-date choices. As of this writing, the tool choices corresponding to the Functional Annotation tools cannot be handled by this package.
`URLlengthLimit`	Published maximum length of the constructed URL.
`details`	If TRUE (default), a list of intermediate results is returned; otherwise, just the final query result.
`verbose`	If TRUE (default is FALSE), more debugging information is printed.
`writeHTML`	If TRUE (default is FALSE), write the received intermediate HTML to files.
`testMe`	If TRUE (default is FALSE), assign default valuse and run.
`graphicMenu`	If TRUE (default is FALSE), use a GUI window for the pick menus.
`formatIt`	If TRUE (default), try to interpret the returned character table and structure the result. If false, the unadorned character table returned by DAVID.

The API described at http://david.abcc.ncifcrf.gov/content.jsp?file=DAVID_API.html is used. The return is screen-scraped, a new URL is formulated and transmitted, again the return is screen-scraped to find the name of the results file, and finally that file is retrieved into a string matrix.

Obviously this approach is brittle, but it has survived the 2008 DAVID update. A real API would be better, of course.

The return value DAVIDQueryResult is just a character matrix. Its content structure depends on the choices of tool and annotation arguments, so there has been no attempt to manipulate it into, say, a data frame with nice column names.

If detail==FALSE, only DAVIDQueryResult is returned. This a character matrix holding the results of the tab-delimited file returned by DAVID.

If detail==TRUE, a list with contents useful for trouble-shooting:

`firstURL`
`firstStageResult`
`DAVIDaction`
`secondURL`
`secondStageResult`
`hasSessionEnded`
`downloadFileName`
`downloadURL`
`DAVIDQueryResult`

Roger Day

http://david.abcc.ncifcrf.gov http://david.abcc.ncifcrf.gov/content.jsp?file=DAVID_API.html Article: DAVID gene ID conversion tool. Da Wei Huang, Brad T Sherman, Robert Stephens, Michael W Baseler, H Clifford Lane, and Richard A Lempicki http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pubmed&pubmedid=18841237

DAVIDQueryLoop, formatDAVIDResult

result = DAVIDQuery(testMe=TRUE)
print(names(result))
print(result$firstURL)
print(result$secondURL)
print(names(result$DAVIDQueryResult))
print(names(result$DAVIDQueryResult$O00161))
print(result$DAVIDQueryResult$O00161$GENE_SYMBOL)  ### Uses UNIPROT ID's for input.