Data-Driven Search for Drug-Membrane Permeability Equations
Drug efficacy depends on its capacity to permeate across the cell membrane. This capacity is quantified by the drug–membrane permeability coefficient. In this talk, I will present results from our recent work where we considered the prediction of passive permeability coefficient via equations that depend on acidity in addition to the widely recognized hydrophobicity. To discover easily interpretable equations that explain the data well, we used sure-independence screening and sparsifyingoperator (SISSO), an artificial-intelligence technique that combines symbolic regression with compressed sensing. Our study is based on a large in silico dataset of 0.4 million small molecules extracted from coarse-grained simulations. I will then rationalize the equation suggested by SISSO via an analysis of the inhomogeneous solubility–diffusion model in several asymptotic acidity regimes. Together, SISSO and analytically derived asymptotes establish and validate an accurate, broadly applicable, and interpretable equation for passive permeability.
Hosted by Professor Zheng Shi