Hepatotoxicity prediction by systems biology modeling of disturbed metabolic pathways using gene expression data

Carbonell, Pablo; López, Oriol; Amberg, Alexander; Pastor, Manuel; Sanz, Ferrán

doi:10.14573/altex.1602071

Cited by 14 publications

(6 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another strategy is using these scores in more sophisticated modeling approaches, representing better the complexity of the endpoints, most representing a mixture of diverse toxicity mechanisms. An example of this strategy was also published by our group ( Carbonell et al , 2017 ) where we applied a systems biology approach to model in vivo liver toxicity using an earlier version of the scoring method described here. Finally, recent advances in machine learning methods ( Angermueller et al , 2016 ) and the use of deep learning techniques ( Mayr et al , 2016 ) suggest that we can still improve the predictive quality of models obtained using our proposed hazard scoring.…”

Section: Discussionmentioning

confidence: 99%

Generating Modeling Data From Repeat-Dose Toxicity Reports

López-Massaguer

Pinto-Gil

Sanz

et al. 2017

Toxicological Sciences

View full text Add to dashboard Cite

Over the past decades, pharmaceutical companies have conducted a large number of high-quality in vivo repeat-dose toxicity (RDT) studies for regulatory purposes. As part of the eTOX project, a high number of these studies have been compiled and integrated into a database. This valuable resource can be queried directly, but it can be further exploited to build predictive models. As the studies were originally conducted to investigate the properties of individual compounds, the experimental conditions across the studies are highly heterogeneous. Consequently, the original data required normalization/standardization, filtering, categorization and integration to make possible any data analysis (such as building predictive models). Additionally, the primary objectives of the RDT studies were to identify toxicological findings, most of which do not directly translate to in vivo endpoints. This article describes a method to extract datasets containing comparable toxicological properties for a series of compounds amenable for building predictive models. The proposed strategy starts with the normalization of the terms used within the original reports. Then, comparable datasets are extracted from the database by applying filters based on the experimental conditions. Finally, carefully selected profiles of toxicological findings are mapped to endpoints of interest, generating QSAR-like tables. In this work, we describe in detail the strategy and tools used for carrying out these transformations and illustrate its application in a data sample extracted from the eTOX database. The suitability of the resulting tables for developing hazard-predicting models was investigated by building proof-of-concept models for in vivo liver endpoints.

show abstract

Section: Discussionmentioning

confidence: 99%

Generating Modeling Data From Repeat-Dose Toxicity Reports

López-Massaguer

Pinto-Gil

Sanz

et al. 2017

Toxicological Sciences

View full text Add to dashboard Cite

show abstract

“…To evaluate the hepatotoxicity of NSAIDs, we used the method developed by Carbonell et al [ 30 ]. To put it shortly, a metabolic model of human metabolism global reconstruction (Recon 2), MODEL1109130000 [ 31 ], was downloaded from the Biomodels database [ 32 ].…”

Section: Methodsmentioning

confidence: 99%

Potential inhibitors of the main protease of SARS-CoV-2 and modulators of arachidonic acid pathway: Non-steroidal anti-inflammatory drugs against COVID-19

Sisakht

Solhjoo

Mahmoodzadeh

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

The main protease of SARS-CoV-2 is one of the key targets to develop and design antiviral drugs. There is no general agreement on the use of non-steroidal anti-inflammatory drugs (NSAIDs) in COVID-19. In this study, we investigated NSAIDs as potential inhibitors for chymotrypsin-like protease (3CLpro) and the main protease of the SARS-CoV-2 to find out the best candidates, which can act as potent inhibitors against the main protease. We also predicted the effect of NSAIDs on the arachidonic pathway and evaluated the hepatotoxicity of the compounds using systems biology techniques. Molecular docking was conducted via AutoDock Vina to estimate the interactions and binding affinities between selected NSAIDs and the main protease. Molecular docking results showed the presence of 10 NSAIDs based on lower binding energy (kcal/mol) toward the 3CLpro inhibition site compared to the co-crystal native ligand Inhibitor N3 (−6.6 kcal/mol). To validate the docking results, molecular dynamic (MD) simulations on the top inhibitor, Talniflumate, were performed. To obtain differentially-expressed genes under the 27 NSAIDs perturbations, we utilized the L1000 final Z-scores from the NCBI GEO repository (GSE92742). The obtained dataset included gene expression profiling signatures for 27 NSAIDs. The hepatotoxicity of NSAIDs was studied by systems biology modeling of Disturbed Metabolic Pathways. This study highlights the new application of NSAIDs as anti-viral drugs used against COVID-19. NSAIDs may also attenuate the cytokine storm through the downregulation of inflammatory mediators in the arachidonic acid pathway.

show abstract

“…Another complication arises when attempting to integrate or fairly compare approaches because, often, the scope or prediction goals of different methods are not readily comparable. Frequently specialized methods are developed for particular classes of compounds ( 22 ) or specific, carefully defined scenarios ( 23 , 24 ). Although most methods measure their success in terms of their ability to predict all possible types of side effects (i.e., from relatively benign to highly dangerous), others ( 10 , 25 ) consider drug toxicity in terms of drug trail failures or withdrawals from the market—a criteria most similar to the one used in this study.…”

Section: Introductionmentioning

confidence: 99%