One option is querying any database for the function of your genes and score the overlap between them using a Jaccard Index: http://en.wikipedia.org/wiki/Jaccard_index

If you don't have any trustable source of function annotation, what about trying a GO term enrichment analysis?

Among others, you can use AmiGO (http://amigo.geneontology.org/cgi-bin/amigo/term_enrichment?session_id=) or FuncAssociate (http://llama.mshri.on.ca/cgi/func/funcassociate). You decide a p-value threshold (maximum 0.05) and annotate your genes with those annotations scoring equal or below your threshold.

When you build-up a network, functional data should be also considered so that network analysis (tools) could make more sense. A recent example - Signaling Network Assessment of Mutations and Copy Number Variations Predict Breast Cancer Subtype-Specific Drug Targets - http://www.cell.com/cell-reports/fulltext/S2211-1247%2813%2900469-5