Design of Silicene Nanopore Sensors for DNASequencing Application: Machine learning assisted DFT+NEGF Study

Decoding the genetic code with electronic sensorsrequires devices that combine high signal fidelity with intelligent data interpretation. In this work, we present a firstprinciples quantum transport investigation of a Z-shaped silicene nanoribbon field-effect transistor (FET) incorporating ananopore for single-nucleobase detection. The electronic structureand current–voltage characteristics are computed using densityfunctional theory combined with the nonequilibrium Green’sfunction formalism. The proposed Z-shaped architecture produces enhanced current levels, which are highly advantageous forachieving an improved signal-to-noise ratio in nanoscale sensing.Asymmetric electrode passivation yields substantially highercurrent discrimination among DNA bases than the symmetricall-hydrogen-passivated device. To enable reliable and automatedbase identification from the transport responses, a machinelearning-assisted framework is employed, where a Random Forestclassifier attains 99.2% classification accuracy. SHAP analysis isused to interpret the model and identify the dominant physicaldescriptors governing nucleobase discrimination. The combineddevice–algorithm co-design establishes this platform as a highsensitivity, label-free, and resource-efficient approach for nextgeneration electronic DNA sequencing.