Read challenge participant data from Synapse for the purpose of extracting the participant's locations. The locations will be used to construct a map.

clean_participant_data.R

## A script to read participant data out of Synapse and
## (help) clean it up.
##
## J. Christopher Bare
## chris.bare@sagebase.org
## Oct. 24, 2013
############################################################

## read evaluations
ev1 <- synGetEvaluation(1917695)
ev2 <- synGetEvaluation(1917696)

## check if all participants for both challenges are the same
par1 <- synGetParticipants(ev1@properties$id, limit=10000)
par2 <- synGetParticipants(ev2@properties$id, limit=10000)

## note that the only discrepancy is me :)
setdiff(sapply(par1@results, function(par) { par$userId }), sapply(par2@results, function(par) { par$userId }))
synGetUserProfile(377358)

## get user profile objects
user_ids_1 <- sapply(par1@results, function(par) { par$userId })
users <- lapply(user_ids_1, synGetUserProfile)

## a helper function to deal with empty slots
nonempty <- function(x) { if (length(x)==0) { NA } else { x } }

## build a data.frame
user_data <- do.call(rbind, lapply(users, function(user) {
    data.frame(
        id=user$ownerId,
        displayName=user$displayName,
        firstName=user$firstName,
        lastName=user$lastName,
        email=nonempty(user$email),
        position=nonempty(user$position),
        industry=nonempty(user$industry),
        organization=nonempty(user$company),
        location=nonempty(user$location),
        team=nonempty(user$teamName)
    )}
))

write.csv(user_data, file='tox_challenge_user_data.csv', row.names=F)

## insert lots of manual scrubbing in the text editor here
## write.csv(participants, file='Desktop/tox_challenge_participants_scrubbed.csv', row.names=F)

## we now have nicely regularized locations
participants <- read.csv('Desktop/tox_challenge_participants_scrubbed.csv', header=T, stringsAsFactors=F)

## split into country, city and, for US locations, state
split_locations <- strsplit(participants$Location.inferred, ", *")
participants$country <- sapply(split_locations, function(loc) { loc[length(loc)] })
participants$city <- sapply(split_locations, function(loc) { if (length(loc) > 1) { loc[1]} else { NA } })
participants$state <- sapply(split_locations, function(loc) { if (length(loc) ==3 && loc[3]=='USA') { loc[2]} else { NA } })

## what an international crew we have!
table(participants$country, useNA='ifany')