Large-scale design automation has been transformative in the development of modern-day electronics like our smartphones. Biological networks that model diseases are similarly complex. But biological systems are unique because of their complexity (both known unknowns and unknown unknowns) and context-dependence (individual functioning modules may not work as expected when composed together). Consequently, existing modeling tools lack large-scale compositionality and reliable validation. So, in this project, we are interested in the following question: can we apply design automation to study diseases like cancer?
For this, we have developed various computational tools for modeling, design, and analysis of biomolecular systems: (1) BioCRNpyler (pronounced bio-compiler) is a tool that can easily create large models of chemical reaction networks, (2) Bioscrape for quick simulations of biological models and for parameter estimation using Bayesian learning, (3) AutoReduce for symbolically deriving low-dimensional representations of the models, and (4) Pacti for large-scale compositional analysis of these models. These Python-based tools are ready to be deployed to new applications and case studies in design automation or system analysis problems. Thus, this project primarily involves the use of Python tools to study biological networks composed of hundreds of interacting biomolecules. Our primary objective is to uncover insights that are experimentally elusive and to computationally investigate a vast array of design possibilities for a chosen biomolecular network.
Collaborators are working on an AFOSUR MURI funded project at Caltech and MIT. Read more here.
Last update: Feb 1, 2025. You can contribute to this page by creating a pull request on GitHub.