HotNet is an algrithom for finding altered subnetworks in a large protein-protein interaction network conatinning a significant number of mutations and copy number alterations (CNAs).
There were 3 significant subnetworks identified in HotNet analysis.
HotNet identifies 3 altered subnetworks based on the matched somatic mutation and copy number alterations data. Table 1 showes the top 10 significant subnetworks.
Table 1. Get Full Table Top 10 out of 3 subnetworks in ranked by p value. The last column of RepeatTimes shows how many times the subnetwork was picked up in differenct delta values.
| Network | No.ofgenes | p-value | RepeatTimes |
|---|---|---|---|
| LAMB1(14),PDE4DIP(10),RINT1(15),GFI1B(10) | 4 | 0.0135 | 1 |
| NRCAM(15),MET(14),PTPRB(6),TNC(11) | 4 | 0.0135 | 1 |
| NEFM(53),TNFRSF10C(54),TNFRSF10A(54),PON1(15),SCARB1(7) | 5 | 0.0160 | 1 |
Somatic mutation data from Mutsig pipeline, copy number alterations derived from GISTIC pipeline and influence matrix derived from Human Protein Reference Database (HPRD) provided by HotNet website [1].
HotNet is an algorithm for de novo identification of significantly altered subnetworks.First, it formulates an influence measure between pairs of genes in the network using a diffusion process defined on the graph. Second, it identifies subnetworks using either a combinatorial model or an enhanced influence model. Finally, it derives a two-stage multiple hypothesis test that mitigates the testing of a large number of hypotheses in subnetwork discovery [2].