Nanopore stochastic sensors are an emerging technology with applications ranging from medical diagnostics to biodefense. They work by measuring the ionic current through a pore. When a molecule passes through the pore, the ionic current is temporarily reduced. The amplitude and duration of current reduction is a signature that can help identify the molecule. While we have a general understanding of the physical principles, we lack a quantitative tool to predict this signature for a particular molecule.
The main objective of this IPRO project is to develop an algorithm to quantitatively predict the amplitude and duration of current reduction for different peptides. The algorithm will likely incorporate AlGDock (github.com/ccbatiit/algdock/), an open-source protein-ligand binding affinity prediction program developed by the Minh research group. Additionally, the team may improve and extend a Graphical User Interface (GUI) for AlGDock that was developed by previous IPRO teams. The team may have the opportunity to also write a scientific research article reporting the results of the project.
The IPRO team will explore approaches based on machine learning from physiochemical properties and based on physical modeling of the pore permeation process. It will investigate software packages to build peptide models, and will create a computer model of the alpha-hemolysin nanopore. The team will run simulations of the permeation process in which the analyte is flexible and pore is rigid. The team will also develop ways to analyze the simulations to predict the current reduction signature.