Biotechnology is rapidly improving, largely due to our ability to manipulate genetic material and efficiently measure biological processes. This talk will describe advances in computational approaches and experimental platforms to probe and control stochastic biological processes within individual cells under a microscope. First, I will describe a novel platform for interfacing individual cells with computational models of gene expression using optogenetics. Master equation-based Bayesian filters are used to perform state estimation and control the gene expression in each cell independently. Second, I will talk about recent work using Graph Neural Networks to model protein-protein interactions. Overall, I will present how a multitude of methods from statistics, AI/ML, and mechanistic modeling can be used to advance biological understanding.