Diverse approaches to predicting drug-induced liver injury using gene-expression profiles

Abstract

Background: Drug-induced liver injury (DILI) is a serious concern during drug development and the treatment of human disease. The ability to accurately predict DILI risk could yield significant improvements in drug attrition rates during drug development, in drug withdrawal rates, and in treatment outcomes. In this paper, we outline our approach to predicting DILI risk using gene-expression data from Build 02 of the Connectivity Map (CMap) as part of the 2018 Critical Assessment of Massive Data Analysis CMap Drug Safety Challenge. Results: First, we used seven classification algorithms independently to predict DILI based on gene-expression values for two cell lines. Similar to what other challenge participants observed, none of these algorithms predicted liver injury on a consistent basis with high accuracy. In an attempt to improve accuracy, we aggregated predictions for six of the algorithms (excluding one that had performed exceptionally poorly) using a soft-voting method. This approach also failed to generalize well to the test set. We investigated alternative approaches-including a multisample normalization method, dimensionality-reduction techniques, a class-weighting scheme, and expanding the number of hyperparameter combinations used as inputs to the soft-voting method. We met limited success with each of these solutions. Conclusions: We conclude that alternative methods and/or datasets will be necessary to effectively predict DILI in patients based on RNA expression levels in cell lines. Reviewers: This article was reviewed by Paweł P Labaj and Aleksandra Gruca (both nominated by David P Kreil).

An ensemble learning approach for modeling the systems biology of drug-induced injury

“…However, the accuracy of the best performing classifier is around the 70% mark (minimum 63%, maximum 76%), stating the limitations of predicting DILI. The results are very similar to previous publications [ 8 – 10 , 12 ]. Additionally, we used the comparison of chemical structures as a feature to predict DILI-causing drugs, though this did not improve the accuracy substantially.…”

“…Although the data provided is much more extensive, including gene expression data from more cell lines, the gold standard is still very reduced and unbalanced. The results in terms of accuracy in the training set are very similar to the ones obtained by Sumsion et al [ 12 ], but worse when looking at the independent hold-out test. This is probably due to the fact that the current independent dataset is based on “Ambiguous-DILI” drugs, making the task more challenging.…”

Integration of human cell lines gene expression and chemical properties of drugs for Drug Induced Liver Injury prediction

“…In this case two human cell lines: MCF7 and PC3, were tested. Chierici et al created a deep learning architecture for DILI prediction based on MCF7 and PC3 human cell lines [ 11 ]. The authors obtained results slightly better than random ones - MCC equal 0.19 in the best case.…”

“…The authors obtained results slightly better than random ones - MCC equal 0.19 in the best case. In work [ 12 ] the same problem was solved by 7 various classifiers. Prediction results were similar to the previous one, with accuracy = 0.7 and MCC = 0.20.…”

Comprehensive genome analysis of a pangolin-associatedParaburkholderia fungorumprovides new insights into its secretion systems and virulence

“…The single marker gene approach may not be able to clearly differentiate all closely related members of Burkholderiales (Jin et al, 2020). Therefore, we used a more robust and reliable whole-genome method or the core-genome SNP approach to further confirm the identity of the pangolin Pf (Fig.…”