runDEG: Apply DEG analysis to voom normalised data for a specific...
In carlmurie/RNASeqUtilities: Helper functions for RNASeq analysis and presentation

Apply DEG analysis to voom normalised data for a specific model and coefficients.

1 2	runDEG(voomDat, coefs, doRandomEffect = FALSE, blocker = NULL, labels = NULL, doWrite = FALSE, doRead = FALSE, rdaPath = NULL)

`voomDat`	Eset of voom normalised data which must include the design. This is the output of voom function in limma.
`coefs`	Vector of coefficients to test. They must map to a coefficient in the design matrix in 'voomDat'.
`doRandomEffect`	Boolean indicating whether there is a random effect in the model. If so then 'duplicateCorrelation' function is run based on 'blocker'. The blocking factor must be identified by 'blocker' parameter.
`blocker`	Vector showing factor of random effect. It is only used if doRandomEffect is TRUE.
`labels`	Vector of character strings that labels each coefficient being tested. This allows for more human readable labels than the lmFit coefficients. Labels must map one to one to 'coefs'.
`doWrite`	Boolean indicating whether to write the random correlation and fitted model to file defined in 'rdaPath'. Useful because computation time can be prohibitive when fitting model.
`doRead`	Boolean indicating whether to read random corrleation and fitted model from file defined in 'rdaPath'. Data must have been previously saved with 'doWrite' parameter.
`rdaPath`	Character string defining full path where rda object will be stored or read. Must be defined if 'doWrite' or 'doRead' are TRUE.