GC-SAML (Genetic Construct Sensitity Analysis & Machine Learning) aims towards understanding the complex interactions between different characteristics of genetic constructs through the use of MGDrivE simulations analyzed through the MoNeT pipelines.

General Idea

I’ve been interested in building surrogate models on our simulated datasets for some time so that they can be examined and shared amongst our collaborators and the community. MoNeT_ML, mainly developed by undergraduate students, was born out of these initial ideas (we even developed an initial set of online dashboards); but it was still a bit difficult for us to trust the generalizability of the results. Jared Bennett had previously mentioned the existance of the SALib package and how we should do some more rigorous statistical analysis on the data to understand the main factors that determine the performance of genetic constructs. After some time the idea of doing the statistical sensitivity analysis, and fitting a regression model independently to contrast the results against each other as means of validation.

Code Repos

• pypi package: MoNeT_MGDrivE
• Analysis Repos: Data Analysis, Machine Learning