BioSalT: a multigene machine learning model for early salinity stress detection in Solanum lycopersicum

Abstract

Salinity stress is a major threat to Solanum lycopersicum (tomato) yields, necessitating tools for early detection. We employed a computational pipeline leveraging Machine Learning (ML) for robust feature selection, followed by Functional Analysis (FA) to confirm the biological relevance of candidate biomarkers. The resulting multigene predictive model was then rigorously validated in the wet lab using the qRT-PCR. This novel, validated approach yields BioSalT, an accurate multigene ML model engineered for the early and timely diagnosis of salinity stress in tomato.

Joy Prokash Debnath

Graduate Research Assistant

My research interests include computational biology and machine learning.