A graph is a set of edges connecting nodes. In this case, the nodes would be the proteins. You can put an edge between proteins found to co-immunoprecipitate in two ways: - in the spoke model, you put an edge only between the bait (i.e. the protein you pulled down) and the preys (i.e. the proteins you identified as interactors of the bait) - in the matrix model, you put an edge between any pair of proteins (i.e. also between the preys) Now, you can try to use the abundance values to weight the edges. How you do it depends on what you want this weight to represent and what your abundance values are. A possible weighting scheme could be the product of relative abundances which can be loosely interpreted as a probability of interaction.