Multiple input file without barcode

[341996]

We downloaded clustered rDNA files from MG-RAST and we would like to annotate samples but the first step "split_libraries.py" is asking for mapping file for barcodes to demultiplex sequences. Since we don't have already demultiplexed sequences we would like to proceed to "pick_otu" step. How can we provide mapping file without using "split_libraries.py"
It was suggested elsewhere to prepare mock mapping file. However pick otuscript does not accept mapping file.
So, what is the proper way to process multiple files from MG-RAST?