Using a Spark DataFrame from AWS Glue as input data for Zingg Pipes: compatibility and options

Hey, if I have already read data into a dataframe in a notebook with spark access (aws glue in this case), is there a way to use this as my input data or is that not compatible with zingg pipes?