Development of QSAR Model for Estimating the Toxicity of Ionic Liquids towards Vibrio Fischeri Marine Bacteria

ILs are liquids composed entirely of ions where they are salts with melting point
lower than 100°C. Recently, Ionic Liquids (ILs) have received much attention from
the scientific community. The use of ILs has been increasing rapidly in broad range
of application segment such as chemical reaction, separation process, and many
more. Despite their high interest and extraordinary properties, the toxicity of ILs
towards aquatic enviromnent needs to be evaluated. This is because, ILs can easily
dissolve in water, and this may lead to aquatic pollution and related risks. Toxicity
of ILs can be measured experimentally. Indeed, the biological effects of a number of
ILs have recently been reported on various species such as bacterium, aquatic
organisms and cells. However, due to the large number of ILs, it would be very
costly and time consuming to measure the ILs' toxicity using experimental work.
Thus, it is necessary to develop mathematical models that can be used to estimate
the toxicity of ILs by means of Quantitative Structure Activity Relationship
(QSAR). In this work, toxicity of ILs towards Vibrio Fischeri marine bacteria is
estimated by developing a new QSAR model. A dataset was constructed using
effective concentration (ECso) for Vibrio Fischeri marine bacteria collected from
different literature and published data. The data was then analysed using summed
contributions from the cations, alkyl substitution, and anions. The model employed
Multiple Linear Regression (MLR) and polynomial models and the code is written
using MATLAB programming language. The result shows that the model is capable
of predicting the ILs toxicities accurately, where R2=96.83% with an average
absolute deviation error 4.96%. The model describes and predicts the considered
ECso values better than other published models and has the potential to be used as an
alternative to experimental measurement in the determination of ECso values for a
wide range of ILs.