Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints
Abstract Background Drug-induced liver injury (DILI) is a critical issue in drug development because DILI causes failures in clinical trials and the withdrawal of approved drugs from the market. There have been many attempts to predict the risk of DILI based on in vivo and in silico identification o...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2017-05-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12859-017-1638-4 |
| _version_ | 1811277960282374144 |
|---|---|
| author | Eunyoung Kim Hojung Nam |
| author_facet | Eunyoung Kim Hojung Nam |
| author_sort | Eunyoung Kim |
| collection | DOAJ |
| description | Abstract Background Drug-induced liver injury (DILI) is a critical issue in drug development because DILI causes failures in clinical trials and the withdrawal of approved drugs from the market. There have been many attempts to predict the risk of DILI based on in vivo and in silico identification of hepatotoxic compounds. In the current study, we propose the in silico prediction model predicting DILI using weighted molecular fingerprints. Results In this study, we used 881 bits of molecular fingerprint and used as features describing presence or absence of each substructure of compounds. Then, the Bayesian probability of each substructure was calculated and labeled (positive or negative for DILI), and a weighted fingerprint was determined from the ratio of DILI-positive to DILI-negative probability values. Using weighted fingerprint features, the prediction models were trained and evaluated with the Random Forest (RF) and Support Vector Machine (SVM) algorithms. The constructed models yielded accuracies of 73.8% and 72.6%, AUCs of 0.791 and 0.768 in cross-validation. In independent tests, models achieved accuracies of 60.1% and 61.1% for RF and SVM, respectively. The results validated that weighted features helped increase overall performance of prediction models. The constructed models were further applied to the prediction of natural compounds in herbs to identify DILI potential, and 13,996 unique herbal compounds were predicted as DILI-positive with the SVM model. Conclusions The prediction models with weighted features increased the performance compared to non-weighted models. Moreover, we predicted the DILI potential of herbs with the best performed model, and the prediction results suggest that many herbal compounds could have potential to be DILI. We can thus infer that taking natural products without detailed references about the relevant pathways may be dangerous. Considering the frequency of use of compounds in natural herbs and their increased application in drug development, DILI labeling would be very important. |
| first_indexed | 2024-04-13T00:27:18Z |
| format | Article |
| id | doaj.art-df81ea811cc9445bbe14f0d1ba3b970b |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-04-13T00:27:18Z |
| publishDate | 2017-05-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-df81ea811cc9445bbe14f0d1ba3b970b2022-12-22T03:10:35ZengBMCBMC Bioinformatics1471-21052017-05-0118S7253410.1186/s12859-017-1638-4Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprintsEunyoung Kim0Hojung Nam1School of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST)School of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST)Abstract Background Drug-induced liver injury (DILI) is a critical issue in drug development because DILI causes failures in clinical trials and the withdrawal of approved drugs from the market. There have been many attempts to predict the risk of DILI based on in vivo and in silico identification of hepatotoxic compounds. In the current study, we propose the in silico prediction model predicting DILI using weighted molecular fingerprints. Results In this study, we used 881 bits of molecular fingerprint and used as features describing presence or absence of each substructure of compounds. Then, the Bayesian probability of each substructure was calculated and labeled (positive or negative for DILI), and a weighted fingerprint was determined from the ratio of DILI-positive to DILI-negative probability values. Using weighted fingerprint features, the prediction models were trained and evaluated with the Random Forest (RF) and Support Vector Machine (SVM) algorithms. The constructed models yielded accuracies of 73.8% and 72.6%, AUCs of 0.791 and 0.768 in cross-validation. In independent tests, models achieved accuracies of 60.1% and 61.1% for RF and SVM, respectively. The results validated that weighted features helped increase overall performance of prediction models. The constructed models were further applied to the prediction of natural compounds in herbs to identify DILI potential, and 13,996 unique herbal compounds were predicted as DILI-positive with the SVM model. Conclusions The prediction models with weighted features increased the performance compared to non-weighted models. Moreover, we predicted the DILI potential of herbs with the best performed model, and the prediction results suggest that many herbal compounds could have potential to be DILI. We can thus infer that taking natural products without detailed references about the relevant pathways may be dangerous. Considering the frequency of use of compounds in natural herbs and their increased application in drug development, DILI labeling would be very important.http://link.springer.com/article/10.1186/s12859-017-1638-4Drug toxicity predictionDrug-induced liver injuryMachine learningData mining |
| spellingShingle | Eunyoung Kim Hojung Nam Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints BMC Bioinformatics Drug toxicity prediction Drug-induced liver injury Machine learning Data mining |
| title | Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints |
| title_full | Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints |
| title_fullStr | Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints |
| title_full_unstemmed | Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints |
| title_short | Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints |
| title_sort | prediction models for drug induced hepatotoxicity by using weighted molecular fingerprints |
| topic | Drug toxicity prediction Drug-induced liver injury Machine learning Data mining |
| url | http://link.springer.com/article/10.1186/s12859-017-1638-4 |
| work_keys_str_mv | AT eunyoungkim predictionmodelsfordruginducedhepatotoxicitybyusingweightedmolecularfingerprints AT hojungnam predictionmodelsfordruginducedhepatotoxicitybyusingweightedmolecularfingerprints |