In silico prediction of mitochondrial toxicity of chemicals using machine learning methods

Abstract

Mitochondria are important organelles in human cells, providing more than 95% of the energy. However, some drugs and environmental chemicals could induce mitochondrial dysfunction, which might cause complex diseases and even worsen the condition of patients with mitochondrial damage. Some drugs have been withdrawn from the market due to their severe mitochondrial toxicity, such as troglitazone. Therefore, there is an urgent need to develop models that could accurately predict the mitochondrial toxicity of chemicals. In this paper, suitable data were obtained from literature and databases first. Then nine types of fingerprints were used to characterize these compounds. Finally, different algorithms were used to build models. Meanwhile, the applicability domain of the prediction models was defined. We have also explored the structural alerts of mitochondrial toxicity, which would be helpful for medicinal chemists to better predict mitochondrial toxicity and further optimize lead compounds.

Mitochondria as the Target of Hepatotoxicity and Drug-Induced Liver Injury: Molecular Mechanisms and Detection Methods

“…Advanced organ-on-a-chip and 3D models replicating the gut-liver axis with added microbiota components could represent the right direction for future interdisciplinary research in the DILI field [ 345 , 346 , 347 , 348 ]. Moreover, the implementation of in silico prediction models, machine learning methods, and the development of comprehensive databases could further assist in selecting better and safer candidates for drug development, and predicting, with high accuracy, potential DILI [ 349 , 350 , 351 , 352 ].…”

Cell morphological profiling enables high-throughput screening for PROteolysis TArgeting Chimera (PROTAC) phenotypic signature

“…Hence, there is a need for descriptors tailored or ‘compatible’ with the bRo5 new data modalities (Ermondi et al, 2021; Lipinski et al, 1997). There have been machine learning approaches for the prediction of drug toxicity by using physiochemical descriptors, structural alerts and high throughput imaging data for small molecules (Hemmerich et al, 2020; Zhang et al, 2009; Zhao et al, 2021). However, computational prediction for new modalities is less investigated.…”

Integrating Cell Morphology with Gene Expression and Chemical Structure to Aid Mitochondrial Toxicity Detection

“…18 However, Zhao et al showed that extrapolation to new structural space is difficult and accuracy inside the models' applicability domain was significantly higher when compared with out-of-domain compounds. 18 Since mitochondrial toxicity can be characterised by a multitude of mechanisms 3 , it has been challenging to assemble sufficient data that can sustain computational methods able to extrapolate to new chemical space. Together with the fact that in vitro assays for mitochondrial toxicity are demanding and with varying degree of reliability, there is a clear need for advancements in the field.…”

“…Previous approaches to computationally predict mitochondrial toxicity have to a large extent been based on predicting mitochondrial membrane depolarisation using chemical structure (Supplementary Figure S1, Supplementary Table S1) and machine learning methods including Support Vector Machines 17 , Random Forest models 18,19 , and naïve Bayes classifier 20 . Using molecular descriptors or structural fingerprints, the best models showed a balanced accuracy between 0.74 to 0.86 as reported by Zhao et.…”