In this study, our goal is to build a network around PMA that includes proteins, GO terms, and pathways that are affected by PMA directly or indirectly. Performing BFSP with PMA as the query keyword and pc= 0.5 returned thousands of proteins and interactions. This is not very surprising since many of proteins in PKC family and those regulating them are hub proteins that are MK-1775 Wee1 inhibitor important in many biological processes. However, not all the reported proteins, pathways or GO terms are actually affected by PMA. The reason is that a significant part of the interaction information used by us is obtained from databases and there is no detailed interaction information available such as directions of the interactions. Proteins that are not affected by PMA directly or indirectly can also be returned, which is not desirable. Clearly, without the directionality information, many false positives are produced and the effect of the signal/query can be difficult to infer accurately. We built a smaller network for PMA by requiring the interactions to be either regulatory type or phosphorylation using interactions extracted from literature, which resulted in only 79 proteins and 166 interactions in total. We manually verified the interactions and kept only the correct ones. The resulting directed network is shown in Fig 4a. In Fig 4b, pathways and GO terms associated with those proteins in Fig 4a are also shown. With this directed network, we can infer with more accuracy the pathways and GO terms affected by PMA. Some pathways are indeed found to be affected by PMA. For example, association of PMA with p38 MAPK signaling pathway is confirmed in Ref , and association of PMA with Atypical NF-kappaB pathway is confirmed in Ref . The former was found through protein MAP3K4 and the latter was found through protein CSNK2A1. In both abstracts, there is no mentioning of the proteins, indicating the relationships were discovered indirectly through other literature. In Fig 5 we show the edges that link the pathways and PMA found using most probable path algorithm . Interestingly, the edges between PMA and the pathways do not actually XAV939 explain the associations because the direction between IGHE and SH3KBP1 is the opposite of what one would expect. It is likely that the real mechanism is not through the path found by MPP. By looking at Fig 4, one can identify a few hub proteins and one of them, JUN, directly regulate the two proteins associated with the two pathways.