Paver as a service: What shall I deliver?

We need the funtional categorization and quantitative data.

You can either choose an available system as KEGG, GO, etc. and tailor it with respect to your experiment, since you won't need all the categories, or your own functional categorization.

The structure of gene functional data need to have a certain format. Inside a table several columns should represent different hierarchy levels of a gene functional classification (of your choice):

level_1<tab>level_2<tab>level_3<tab>protein_name<:>protein_ID

Example:

metabolism<tab>carbon metabolism<tab>glycolysis<tab>ENO:IDxyz

The more detailed this protein/gene classification scheme (which you have to provide), the clearer the results will be. Because Paver will analyse homogenity of functional cluster's gene expression, you need at least three genes per functional cluster and at least about 400 genes/proteins in total.

We also need tables of expression data. Expression data can come from any source like arrays, 2D gels, gelfree analyses .... These tables shall look like this:

protein_ID<tab>expression_value[<tab>coefficient_of_variation]

where IDxyz is the reference to its respective entry in the functional categorization. Content in [] is optional.

Example:

IDxyz<tab>1.157<tab>0.2

We can help you with data normalization or rescaling and transformation. Please let us know if this is of interest for you.

As a matter of course your data will be kept confidential.