The first purpose of our examine was to deal with the issue of transcriptomic information complexity by introducing a novel Transcriptomic Response Index (TRI), compressing all the transcriptomic area right into a single variable, and linking it with the inhaled Multiwalled Carbon Nanotubes (MWCNTs) properties. This technique permits us to foretell fold change values of 1000’s of differentially expressed genes (DEGs) utilizing a single variable and a single Quantitative Construction-Exercise Relationship (QSAR) mannequin. Within the context of this work, TRI compressed 5167 DEGs right into a single variable, explaining 99,9% of all the transcriptomic area. Additional TRI was linked to the properties of inhaled MWCNTs utilizing a nano-QSAR mannequin with statistics R2 = 0.83, Q2cv = 0.8, and Q2 = 0.78, which present a excessive stage of goodness-of-fit, robustness, and predictability of the obtained mannequin. By coaching a nano-QSAR mannequin on fold modifications of 1000’s of DEGs utilizing a single variable, our examine considerably contributes not solely to New Strategy Methodologies (NAMs) targeted on lowering animal testing but in addition decreases the quantity of computational assets wanted for work with complicated transcriptomic information. Developed throughout this work, the software program referred to as ChemBioML Platform (https://chembioml.com) presents researchers a robust free-to-use device for coaching regulatory acceptable Machine Studying (ML) fashions and not using a sturdy background in programming. The ChemBioML Platform integrates the ML capabilities of Python with the superior graphical interface of Unreal Engine 5, making a bridge between scientific analysis and the sport improvement business.
