Machine learning assisted protein identification with highly sensitive MoS2 nanopores

Narayana Aluru, University of Texas at Austin

Usage Details

Sequencing a chain of human amino acids can enable breakthrough advances in early diagnosis of diseases and the health status of the human body. Many diseases including cancer, diabetes and digestive disorders are caused by malfunctioning of ribosomes leading to defective proteins. Therefore, sequencing an amino acid chain helps diagnose diseases at early stages. In this study, using extensive molecular simulations, we intend to study the ability of a nanoporous single-layer molybdenum disulfide (MoS₂) in detecting individual amino acids in a polypeptide chain. This includes featurizing and characterizing all the 20 human amino acids based on their ionic currents and residence times using machine learning techniques. We believe that our findings can help implement medical devices for disease diagnostic purposes. To carry out our study, extensive molecular dynamics simulations (up to 30 μs of simulation time) need to be performed. Therefore, the Blue Waters machine is needed to carry out these expensive computations.