|
We created a new algorithm dedicated to biological network reverse-engineering that allows for single or joint modelling of, for instance, genes and proteins. It is designed to work with patterned data. Famous examples of problems related to patterned data are: recovering signals in networks after a stimulation (cascade network reverse engineering) and analysing periodic signals.
The algorithm begins with a step to select the actors that will be used in the reverse engineering upcoming step. An actor can be included in that selection based on its differential effects (for instance gene expression or protein abundance) or on its time course profile.
The actors are then clustered and various time-varying patterns of interaction between the clusters can be specified.
Many inference functions can be used in the inference step of the algorithm to getting specific features for the inferred network such as sparsity, robust links, high confidence links or stable through resampling links.
In order to include biological a priori knowledge in the network inference, we wanted to enable weighted inference and had to extend some known algorithm such as stability selection.
|
