NCI Features PHOENIX application

The National Cancer Institute news recently featured PHOENIX (“Prior-informed Hill-like ODEs to Enhance Neuralnet Integrals with eXplainability”), a new ML application that uses neural ordinary differential equations (ODEs), a type of ML that interprets how a system will behave over time, to model how regulatory proteins (transcription factors) influence their target genes. The original article describing the application was published in May of this year in Genome Biology.

In the profile, lead author, Intekhab Hossain and senior and corresponding author John Quackenbush note that PHOENIX takes into account both the complexity of the system and what’s known about how gene transcription works (for example, which of the transcription factors bind to the regulatory region of each gene).

Using that information as a starting point, PHOENIX learns time-dependent functions to predict how each gene will behave in a particular disease state, as it develops and progresses. The application is particularly exciting as it can examine entire repertoires of genes as well as subtle regulatory changes, ultimately enhancing the biological interpretability of the genes underlying cancer.