Pathway inference methods allow the mapping of biochemical networks, the discovery of signaling components, and the assignment of functions to understudied proteins and genes. Literature and automated text mining have been successfully used to reconstruct metabolic and signaling circuits, while gene regulatory networks may be inferred from gene expression data. As an alternative approach to map members of proliferative pathways, functional pathway inference analysis (FPIA) is based on the premise that genes producing similar phenotypes following perturbation across multiple cell lines belong to a common pathway. We have demonstrated this concept with the use of gene dependency datasets that allow the provision of probabilistic values of pathway membership for thousands of genes. Here, we provide a detailed protocol for the implementation of FPIA in the `cordial` R package. As an illustration of how FPIA may be used to identify new pathway members, we present a step-by-step description of its use for the investigation of genes functionally associated to PI3K and TP53.