Interaction prediction in structure-based virtual screening using deep learning

Gonczarek, Adam; Tomczak, Jakub M.; Zaręba, Szymon; Kaczmar, J. W.; Dąbrowski, Piotr; Walczak, Michał

doi:10.1016/j.compbiomed.2017.09.007

Cited by 86 publications

(50 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compared the results of DEEPScreen with five other deep learning based DTI prediction methods that represent the current state-of-the-art (please refer to the Introduction section) by employing the same datasets used in the corresponding studies. For this analysis, we re-trained and tested DEEPScreen using the exact same experimental settings and evaluation metrics that were described in the respective articles [28][29][30][31][32] . A total of 3 different benchmark datasets were employed for this purpose (please refer to the Methods section).…”

Section: State-of-the-art Dnn-based Methods Performance Comparisonmentioning

confidence: 99%

“…The term "deep learning" (DL) is coined for the novel ML techniques that perform significantly better compared to conventional classifiers especially in the fields of computer vision and natural language processing, mainly due to multiple layers of data abstraction 25 32 . The field of the deep learning based DTI prediction is still in its infancy and the studies published so far were mostly focused on the applicability of deep learning algorithms and prototyping 27,28,30,32 . The results of these studies have indicated that deep learning has a great potential to advance the field by identifying unknown DTIs at large-scale [28][29][30][31][32][33] .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DEEPScreen: High Performance Drug-Target Interaction Prediction with Convolutional Neural Networks Using 2-D Structural Compound Representations

Rifaioğlu¹,

Atalay²,

Martin

et al. 2018

Preprint

View full text Add to dashboard Cite

The identification of physical interactions between drug candidate chemical substances and target biomolecules is an important step in the process of drug discovery, where the standard procedure is the systematic screening of chemical compounds against pre-selected target proteins. However, experimental screening procedures are expensive and time consuming, therefore, it is not possible to carry out comprehensive tests. Within the last decade, computational approaches have been developed with the objective of aiding experimental studies by predicting novel drug-target interactions (DTI), via the construction and application of statistical models. In this study, we propose a large-scale DTI interaction prediction system, DEEPScreen, for early stage drug discovery, using convolutional deep neural networks. One of the main advantages of DEEPScreen is employing readily available simple 2-D images of compounds at the input level instead of engineered complex feature vectors that displayed limited performance in DTI prediction tasks previously. DEEPScreen learns complex features inherently from the 2-D molecular representations, thus producing highly accurate predictions.DEEPScreen system was trained for 704 target proteins (using ChEMBL curated bioactivity data) and finalized with rigorous hyper-parameter optimization tests. We compared the performance of DEEPScreen against shallow classifiers such as the random forest, logistic regression and support † A preliminary version of this study have been orally presented at vector machines, to indicate the effectiveness of the proposed deep learning approach. Additionally, we compared DEEPScreen with other deep learning based state-of-the-art DTI predictors on widely used benchmark datasets and showed that DEEPScreen produces better or comparable results to the top performers. The method proposed here can be employed to computationally scan a large portion of the recorded drug candidate compound and protein spaces to aid the experimentalists working in the field of drug discovery and repurposing by providing a preselection of interesting novel DTIs. of correctly predicted non-interacting compound-target pairs, whereas FP (i.e., false positive) represents the number of non-interacting compound target pairs, which are predicted as interacting.

show abstract

Section: State-of-the-art Dnn-based Methods Performance Comparisonmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

DEEPScreen: High Performance Drug-Target Interaction Prediction with Convolutional Neural Networks Using 2-D Structural Compound Representations

Rifaioğlu¹,

Atalay²,

Martin

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…[44] Gonczarek, A., et al develop deep learning architecture for structure-based virtual screening by atom convolution and softmax operation. [45] These studies suggested that DNN were capable of QSAR modeling, and indicated the potential ability of virtual screening. However, DNN's ability to screen large compound libraries has not been tested.…”

Section: Introductionmentioning

confidence: 96%

Development of Ligand‐based Big Data Deep Neural Network Models for Virtual Screening of Large Compound Libraries

Xiao

Chen

et al. 2018

Molecular Informatics

View full text Add to dashboard Cite

High-performance ligand-based virtual screening (VS) models have been developed using various computational methods, including the deep neural network (DNN) method. There are high expectations for exploration of the advanced capabilities of DNN to improve VS performance, and this capability has been optimally achieved using large data training datasets. However, their ability to screen large compound libraries has not been evaluated. There is a need for developing and evaluating ligand-based large data DNN VS models for large compound libraries. In this study, we developed ligand-based large data DNN VS models for inhibitors of six anticancer targets using 0.5 M training compounds. The developed VS models were evaluated by 10-fold cross-validation, achieving 77.9-97.8 % sensitivity, 99.9-100 % specificity, 0.82-0.98 Matthews correlation coefficient and 0.98-0.99 area under the curve, outperforming random forest models. Moreover, DNN VS models developed by pre-2015 inhibitors identified 50 % of post-2015 inhibitors with a 0.01-0.09 % false positive rate in screening 89 M PubChem compounds, also outperforming previous models. Experimental assays of the selected virtual hits of the EGFR inhibitor model led to reasonable novel structures of EGFR inhibitors. Our results confirmed the usefulness of the large data DNN model as a ligand-based VS tool to screen large compound libraries.

show abstract

“…Recently, deep learning has been applying to drug discovery [20], [21]. It has achieved superior performance compared to traditional machine learning techniques in many problems in drug development such as drug visual screening [22], [23], drug-target profiling [24], [25], [26], [27], drug repositioning [28], [29]. Especially in the drug response problem, deep learning is utilized to automatically learn genomic features of cell lines and the structural features of drugs to predict anticancer drug responsiveness [30], [31], [32], [33].…”

Section: Introductionmentioning

confidence: 99%

Graph convolutional networks for drug response prediction

Nguyen¹,

Nguyen

Le³

2020

Preprint

View full text Add to dashboard Cite

Background: Drug response prediction is an important problem in computational personalized medicine. Many machine learning-, especially deep learning-, based methods have been proposed for this task. However, these methods often represented the drugs as strings, which are not a natural way to depict molecules. Also, interpretation has not been considered thoroughly in these methods. Methods: In this study, we propose a novel method, GraphDRP, based on graph convolutional network for the problem. In GraphDRP, drugs are represented in molecular graphs directly capturing the bonds among atoms, meanwhile cell lines are depicted as binary vectors of genomic aberrations. Representative features of drugs and cell lines are learned by convolution layers, then combined to represent for each drug-cell line pair. Finally, the response value of each drug-cell line pair is predicted by a fully-connected neural network. Four variants of graph convolutional networks are used for learning the features of drugs. Results: We find that GraphDRP outperforms tCNN in all performance measures for all experiments. Also, through saliency maps of the resulting GraphDRP models, we discover the contribution of the genomic aberrations to the responses. Conclusion: Representing drugs as graphs are able to improve the performance of drug response prediction. Data and source code can be downloaded at https://github.com/hauldhut/GraphDRP.

show abstract

Interaction prediction in structure-based virtual screening using deep learning

Cited by 86 publications

References 22 publications

DEEPScreen: High Performance Drug-Target Interaction Prediction with Convolutional Neural Networks Using 2-D Structural Compound Representations

DEEPScreen: High Performance Drug-Target Interaction Prediction with Convolutional Neural Networks Using 2-D Structural Compound Representations

Development of Ligand‐based Big Data Deep Neural Network Models for Virtual Screening of Large Compound Libraries

Graph convolutional networks for drug response prediction

Contact Info

Product

Resources

About