Streamline DE machinery

[marionilab]

Streamline DE machinery

[LTLA]

Don't forget to add a lfc= option for pairwiseWilcox(); this could probably be co-opted from tol=.

[kevinrue]

• return unsorted output from findMarkers. This should allow you to just do the following for collating results into a single table.

markers.pbmc.up <- findMarkers(sce.pbmc, sce.pbmc$cluster, direction="up", log.p=TRUE)
x <- DataFrame(
  row.names = rownames(sce.pbmc),
  lapply(markers.pbmc.up, "[[", i="log.FDR")
  )