Editing existing models and adding new ones

[MahShaaban]

I would like to ask if there is a way to inspect and edit the models currently implemented in the package? Also, is there a way to transform the data that I have in the form of a graph to build a model and then score it using NPA? Thanks

[sgubianpm]

@MahShaaban , thanks for your interest in the NPA package. There is currently nothing implemented to inspect directly the models, however, when you have scored it using compute_npa, you can look at it in a graphical way with each node showing NPA scores by using plot function. It will display a html page with javascript/D3.js graph.

In order to score the network with your data, the object has to be a R list structured like the COPD1 example provided, which mean each slot is named with contrast/comparison name (which is usually control vs treatmentX) and each slot contains a data.frame object, with row names being gene names (gene symbols) and col names being foldChange and t for statistical value. I just updated the README.md file in the NPA GitHub page so that the instructions given in the vignette are also provided on the main page. I hope this helps.

[MahShaaban]

Thanks @sgubianpm I edited the second question to clarify what I meant. The package contains several models, but I am interested in pathways that are not available in it. I am asking is there a way to transform a network into a model like the ones included in the package? The goal is to use gene expression data, like you describe, to score the model using compute_NPA.

[MahShaaban]

I am still struggling to understand how I can build networks other than the ones provided in the paper or from CBN. In particular, how to use transcriptomic data to build the second layer and integrate it with the functional layer to score the network.