Predicting Drug-Induced Liver Injury Using Convolutional Neural Network and Molecular Fingerprint-Embedded Features

Nguyen-Vo, Thanh-Hoang; Nguyen, Loc; Do, Nguyet; Le, Phuc; Nguyễn, Thị Hoài; Nguyen, Binh P.; Le, Ly

doi:10.1021/acsomega.0c03866

Cited by 41 publications

(49 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Li et al developed a model for predicting drug-induced liver injury using time splits and evaluated its predictive performance . The area under the receiver operating characteristic curve (AUC) value in their study was lower than those in other reports about adverse event prediction using the random split. , Although they do not strongly advocate the importance of time split in their paper, their results suggest that the accuracy of the random-split-derived model is overestimated compared to the accuracy of the time-split-derived model. Because their study focused specifically on drug-induced liver injury, used only structure-based features as input, and did not compare to random split under the same settings, it is unclear whether the random split-derived model is overestimated or whether the same claim can be made for other data sets and adverse events.…”

Section: Introductionmentioning

confidence: 89%

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning

Morita

Mizuno

Kusuhara

2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Adverse events are a serious issue in drug development, and many prediction methods using machine learning have been developed. The random split cross-validation is the de facto standard for model building and evaluation in machine learning, but care should be taken in adverse event prediction because this approach does not strictly match the real-world situation. The time split, which uses the time axis, is considered suitable for real-world prediction. However, the differences in model performance obtained using the time and random splits are not clear due to the lack of comparable studies. To understand the differences, we compared the model performance between the time and random splits using nine types of compound information as input, eight adverse events as targets, and six machine learning algorithms. The random split showed higher area under the curve values than did the time split for six of eight targets. The chemical spaces of the training and test datasets of the time split were similar, suggesting that the concept of applicability domain is insufficient to explain the differences derived from the splitting. The area under the curve differences were smaller for the protein interaction than for the other datasets. Subsequent detailed analyses suggested the danger of confounding in the use of knowledge-based information in the time split. These findings indicate the importance of understanding the differences between the time and random splits in adverse event prediction and suggest that appropriate use of the splitting strategies and interpretation of results are necessary for the real-world prediction of adverse events. We provide the analysis code and datasets used in the present study at .

show abstract

Section: Introductionmentioning

confidence: 89%

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning

Morita

Mizuno

Kusuhara

2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Long shortterm memory, a deep learning architecture, belongs to a group of recurrent neural networks which are widely used in natural language processing and machine translation. For a decade, deep learning has been widely implemented to solve multiple issues in diverse fields, including biology [50], chemistry [51][52][53], and biochemistry [54][55][56][57]. Numerous computational approaches were developed using deep learning to address diverse biological issues [58][59][60][61][62][63][64].…”

Section: Introductionmentioning

confidence: 99%

iPromoter-Seqvec: identifying promoters using bidirectional long short-term memory and sequence-embedded features

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background Promoters, non-coding DNA sequences located at upstream regions of the transcription start site of genes/gene clusters, are essential regulatory elements for the initiation and regulation of transcriptional processes. Furthermore, identifying promoters in DNA sequences and genomes significantly contributes to discovering entire structures of genes of interest. Therefore, exploration of promoter regions is one of the most imperative topics in molecular genetics and biology. Besides experimental techniques, computational methods have been developed to predict promoters. In this study, we propose iPromoter-Seqvec – an efficient computational model to predict TATA and non-TATA promoters in human and mouse genomes using bidirectional long short-term memory neural networks in combination with sequence-embedded features extracted from input sequences. The promoter and non-promoter sequences were retrieved from the Eukaryotic Promoter database and then were refined to create four benchmark datasets. Results The area under the receiver operating characteristic curve (AUCROC) and the area under the precision-recall curve (AUCPR) were used as two key metrics to evaluate model performance. Results on independent test sets showed that iPromoter-Seqvec outperformed other state-of-the-art methods with AUCROC values ranging from 0.85 to 0.99 and AUCPR values ranging from 0.86 to 0.99. Models predicting TATA promoters in both species had slightly higher predictive power compared to those predicting non-TATA promoters. With a novel idea of constructing artificial non-promoter sequences based on promoter sequences, our models were able to learn highly specific characteristics discriminating promoters from non-promoters to improve predictive efficiency. Conclusions iPromoter-Seqvec is a stable and robust model for predicting both TATA and non-TATA promoters in human and mouse genomes. Our proposed method was also deployed as an online web server with a user-friendly interface to support research communities. Links to our source codes and web server are available at https://github.com/mldlproject/2022-iPromoter-Seqvec.

show abstract

“…The structure–activity relationships (SARs) allow predicting DILI based on the structural formula of a compound and, thus, are applicable at the early stages of drug development. − Plenty of SAR models for the prediction of DILI have been developed in recent years. − The accuracy of the models built varies from 0.6 to 0.9 depending on the data sets, methods, and descriptors. The ensemble approaches ,,, based on the integration of prediction results from many models and created using various machine learning techniques and descriptors, as well as those ,,,, based on deep learning, usually outperform “classical” ones based on the only machine learning method and the only type of descriptors. Most studies were focused on the prediction of DILI itself, whereas some others , were related to the prediction of particular toxicity endpoints, such as hepatitis, cholestasis, or ALF.…”

Section: Introductionmentioning

confidence: 99%

“…The confidence of drug-DILI associations was determined by checking if these relationships were mentioned in previously published datasets. The DILIrank dataset was used to create and test SAR models in many studies. ,,,− , Only one study was related to the creation of a three-class model, whereas the others were related to the creation of binary models. In some studies, ,, only most- and no-DILI-concern drugs were used to create models, which may restrict their applicability domain and lead to incorrect predictions for compounds causing moderate DILI (less-DILI-concern).…”

Section: Introductionmentioning

confidence: 99%

Relationships between the Structure and Severe Drug-Induced Liver Injury for Low, Medium, and High Doses of Drugs

Ivanov

Lagunin

Филимонов

et al. 2022

Chem. Res. Toxicol.

View full text Add to dashboard Cite

Assessment of structure–activity relationships (SARs) for predicting severe drug-induced liver injury (DILI) is essential since in vivo and in vitro preclinical methods cannot detect many druglike compounds disrupting liver functions. To date, plenty of SAR models for the prediction of DILI have been developed; however, none of them considered the route of drug administration and daily dose, which may introduce significant bias into prediction results. We have created a dataset of 617 drugs with parenteral and oral administration routes and consistent information on DILI severity. We have found a clear relationship between route, dose, and DILI severity. According to SAR, nearly 40% of moderate- and non-DILI-causing drugs would cause severe DILI if they were administered at high oral doses. We have proposed the following approach to predict severe DILI. New compounds recommended to be used at low oral doses (<∼10 mg daily), or parenterally, can be considered not causing severe DILI. DILI for compounds administered at medium oral doses (∼10–100 mg daily; 22.2% of drugs under consideration) can be considered unpredictable because reasonable SAR models were not obtained due to the small size and heterogeneity of the corresponding dataset. The DILI potential of the compounds recommended to be used at high oral doses (more than ∼100 mg daily) can be estimated using SAR modeling. The balanced accuracy of the approach calculated by a 10-fold cross-validation procedure is 0.803. The developed approach can be used to estimate severe DILI for druglike compounds proposed to use at low and high oral doses or parenterally at the early stages of drug development.

show abstract

Predicting Drug-Induced Liver Injury Using Convolutional Neural Network and Molecular Fingerprint-Embedded Features

Cited by 41 publications

References 48 publications

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning

iPromoter-Seqvec: identifying promoters using bidirectional long short-term memory and sequence-embedded features

Relationships between the Structure and Severe Drug-Induced Liver Injury for Low, Medium, and High Doses of Drugs

Contact Info

Product

Resources

About