AI-Driven Multi-Scale Investigation of the RAS/RAF Activation Lifecycle (ADMIRRAL)

As part of the NCI-DOE Collaboration, the ADMIRRAL project provides a computational complement to the ongoing NCI RAS Initiative, which focuses on one of cancer’s hardest problemsidentifying targeted treatment for oncogenic RAS proteins.

The ADMIRRAL project team works to develop a predictive molecular-scale understanding of RAS-RAF activation. Oncogenic mutations in RAS genes that maintain RAS proteins in an active conformation that promotes cell growth drive more than 20% of all human cancers. RAS mutations are prevalent in lung, colon, and pancreatic cancers, resulting in poor prognosis and high mortality.

Biological data from advances in genomics, imaging, biophysics, and structural biology clarifies parts of the biological, mechanistic underpinnings of RAS-driven cancers. Experts generally believe that activation of the RAS-RAF complex involves substantial re-organization of RAF and rearrangement of RAS and RAF on the membrane. However, their ability to evaluate this behavior at the appropriate fine length and time scales limits experiments. Conversely, time and length scales that are too short to describe relevant motion limit traditional computational approaches.

The ADMIRRAL project will build a novel computing infrastructure to bridge this gap, using machine learning (informed by experiment) to guide a massive ensemble of simulations that capture biologically relevant time and length scales with unprecedented resolution. This capability will paint the most accurate picture of key RAS-RAF interactions, including with the cellular membrane. This work will also enable the generation of powerful, new hypotheses for RAS-RAF-driven cancers.

The ADMIRRAL project is co-led by:

  • Dwight Nissley, NCI’s Frederick National Laboratory for Cancer Research  
  • Fred Streitz, Lawrence Livermore National Laboratory, Department of Energy

Aims of the Project

  • Characterize reorganization of RAF protein upon disengagement with the 14-3-3 protein
  • Delineate large-scale domain rearrangement of the RAS-RAF protein complex on the membrane
  • Describe the engagement and dimerization of the RAF protein kinase domains

You can find ADMIRRAL resources (and previous “Pilot 2—Molecular Level Pilot” resources) on the NCI-DOE Collaboration AI/ML Resources page, on GitHub, and through the NCI Predictive Oncology Model and Data Clearinghouse.


Vote below about this page’s helpfulness.