Hi All, I'm working on building some galaxy tools which can be used together in a workflow and one data structure that seems to show up a lot is a data matrix where both the data rows and data columns have annotations which I would like to preserve through the workflow. Here is an example that hopefully make this more concrete: we have X genotyping probes in a microarray that we have run on Y samples which results in a data matrix with X rows and Y columns. Each of the probes (rows) has annotation data like allele (A vs B allele), sequence, SNP ID, etc and each of the columns (samples) has it's own annotation data like strain, date etc... So what do you think is the best way to represent this kind of structure? Does galaxy have a mechanism that allows you to associate files as a logical group which would allow my intensity data and meta data to stay together as a "dataset" without having to be in the same file? Thank you Keith