The Expanded p53 Interactome as a Predictive Model for Cancer Therapy

Abstract

The tumour suppressor gene TP53 is implicated in the majority of all human cancers, thus pivotal to genomic integrity. Even though over 72,000 PubMed publications are linked with the keyword p53 and this number is continuously increasing, due to the complexity of its interactions we are still far from fully elucidating p53's role in tumorigenesis. Computational methodologies are novel tools to depict and dissect complex disease networks. The Boolean PKT206 p53-DNA damage model has previously demonstrated good predictive capability for p53 wildtype and null tumours in various in silico knockouts. Here, we have expanded PKT206 to generate a more clinically robust representation of p53 dynamics. The new PMH260 model incorporates 260 nodes representing genes, with 980 interactions between them representing inhibitions and activations. Additional biological outputs, including angiogenesis, cell cycle arrest and DNA repair were also amalgamated into the model. Three in silico knockouts of highly connected nodes (p53, MDM2 and FGF2) were generated and logical steady state analysis and dependency relationships determined. 71 % of predictions were considered true from superimposition of human osteosarcoma and HCT116 microarray profiles. In silico knockout analysis revealed 98 potential novel predictions, of which 13 were validated by literature; 83 % of them were overlapping with PKT206. Thus the expanded Boolean PMH260 model offers a promising platform for clinical potential in targeted cancer therapeutics.

Insight into glucocorticoid receptor signalling through interactome model analysis

“…Understanding the node connectivity within the model was crucial to the choice of which nodes would be selected for in silico knockout analysis, as previous studies have focussed on in silico knockouts for only the most highly connected nodes [ 18 , 19 ].…”

“…Although such models provide useful insight, they are both time-consuming and resource-expensive to create due to the required biological data. Boolean modelling on the contrary allows for the generation of large-scale models that provide a qualitative overview of the behaviour of an entire network [ 18 , 19 ]. In these cases, interactions and molecular levels are simplified to ON or OFF binary values, removing the need to know exact rate and kinetic equations thus reducing computational demand [ 20 ].…”

“…Correct prediction rates reached 71% for the model, signifying the strength of this approach [ 18 ]. An expanded p53 interactome was later developed to more accurately model the signalling phenomena [ 19 ].…”

“…The aim of this research was to develop a Boolean model for the GR interaction network similar to the p53 interactomes [ 18 , 19 ]. The model (GEB052: G lucocorticoid receptor model by E myr B akker, consisting of 52 nodes) contains 241 interactions.…”

Gene Set to Diseases (GS2D): disease enrichment analysis on human gene sets with literature data

“…The second use case involves 254 human genes interacting with p53, previously used as a predictive model for cancer therapy [22]. Knowing that the tumour suppressor p53 is involved in many human cancers [23], we would expect GS2D to recapitulate this knowledge when analysing p53 interacting genes by returning a list of enriched diseases that includes various types of cancers.…”