Identification of Proteinogenic Amino Acids Using MoS₂ Solid-State Nanopores Assisted by Machine Learning

When and Where

Dec 1, 2022
8:00am - 8:15am

Hynes, Level 2, Room 207

Presenter

Andreina Urquiola Hernandez

Adrien Nicolai

Guyeux Christophe

Senet Patrick

Co-Author(s)

Adrien Nicolai¹,Andreina Urquiola Hernandez¹,Guyeux Christophe¹,Senet Patrick¹

Université de Bourgogne Franche Comté¹

Abstract

Adrien Nicolai¹,Andreina Urquiola Hernandez¹,Guyeux Christophe¹,Senet Patrick¹

Université de Bourgogne Franche Comté¹

Solid-State Nanopores (SSN) made of 2-D materials such as molybdenum disulfide (MoS2) have emerged as one of the most versatile sensors for single-biomolecule detection, which is essential for early disease diagnosis (biomarker detection). One of the most promising applications of SSN is DNA and protein sequencing, at a low cost and faster than the current standard methods. The detection principle relies on measuring the relatively small variations of ionic current as charged biomolecules immersed in an electrolyte traverse the nanopore, in response to an external voltage applied across the membrane. The passage of a biomolecule through the pore yields information about its structure and chemical properties, as demonstrated experimentally particularly for DNA molecules. Indeed, protein sequencing using SSN remains highly challenging since the protein ensemble is far more complex than the DNA ensemble [1]. In this work, we performed extensive unbiased all-atom classical Molecular Dynamics simulations to produce data of translocation of biological peptides through single-layer MoS2 nanopores (D = 1.5 nm). Peptide made of the 20 different amino acids from the different families (non polar/hydrophobic, polar/neutral, basic and acidic) were chemically linked to a short polycationic charge carrier [2]. First, ionic current time series were computed from MD and peptide-induced blockade events were extracted and characterized using structural break detection. Second, clustering (unsupervised learning) of ionic current drops and duration using Gaussian Mixture Model was applied. Using this technique, we demonstrate that each amino acid presents a large diversity of ionic current characteristics, however, charged amino acids were distinguished from the others. Finally, classification (supervised learning) was also implemented to identify residue motifs inside the pore, using XGBoost classifiers. We show that the addition of ionic current drops and duration to time series as input variables of the model greatly improve its performance, leading to the identification of the twentyproteinogenic amino acid. These promising findings may offer a route toward protein sequencing using MoS2 solid-state nanopores. References [1] A. Nicolai and P. Senet. Challenges in Protein Sequencing using MoS2 Nanopores, 2022. In Bowen, W., Vollmer, F., Gordon, R. (eds) Single Molecule Sensing Beyond Fluorescence. Nanostructure Science and Technology. Springer, Cham. https://doi.org/10.1007/978-3-030-90339-8_11 [2] M. D. Barrios Perez, A. Nicolai, P. Delarue, V. Meunier, M. Drndic, and P. Senet. Improved model of ionic transport in 2-D MoS2 membranes with sub-5 nm pores. Appl. Phys. Lett., 2019, 114(2):023107. [3] A. Urquiola Hernandez, A. Nicolai, C. Guyeux, and P. Senet. Identification of the twenty proteinogenic amino acids using MoS2 nanopores assisted by Machine Learning. In preparation.