A toolbox of code snippets that can be used to feed other statistical and data mining analyses, all conducted in a social way within github. Main advantages include: (1) huge repository of previous analyses, with the people who conducted literally one email/g+ post away, (2) clinical researchers do not need to wait for statisticians/analysts to conduct small modifications or test new approaches, although they keep working side by side with clinical researchers for more complex analyses, (3) no need to copy and paste from result logs into a manuscript table -- tables will be generated automatically over time we will automate most of our analyses, (4) all analyses will start with you receiving a script template from a mentor, when clinical researchers can then see one and then do one, (5) all of our analyses are fully reproducible, with scripts being available within manuscripts, something that reviewers will definitely appreciate
rpietro
commented
12 years ago
function to upload many packages and post as gist
lapply(c("ggplot2", "psych", "RCurl", "irr", "car","Hmisc", "gmodels","qpcR"), library, character.only=T)