Agreed. Transform with Pipe.jl really comes to life. I want to mention a new framework for some syntactic sugar around the basic transform/combine commands in DataFrame:
https://github.com/matthieugomez/DataFramesMacros.jl
Hope this is helpful to someone.