+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      DeepSynergy: predicting anti-cancer drug synergy with Deep Learning


      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.



          While drug combination therapies are a well-established concept in cancer treatment, identifying novel synergistic combinations is challenging due to the size of combinatorial space. However, computational approaches have emerged as a time- and cost-efficient way to prioritize combinations to test, based on recently available large-scale combination screening data. Recently, Deep Learning has had an impact in many research areas by achieving new state-of-the-art model performance. However, Deep Learning has not yet been applied to drug synergy prediction, which is the approach we present here, termed DeepSynergy. DeepSynergy uses chemical and genomic information as input information, a normalization strategy to account for input data heterogeneity, and conical layers to model drug synergies.


          DeepSynergy was compared to other machine learning methods such as Gradient Boosting Machines, Random Forests, Support Vector Machines and Elastic Nets on the largest publicly available synergy dataset with respect to mean squared error. DeepSynergy significantly outperformed the other methods with an improvement of 7.2% over the second best method at the prediction of novel drug combinations within the space of explored drugs and cell lines. At this task, the mean Pearson correlation coefficient between the measured and the predicted values of DeepSynergy was 0.73. Applying DeepSynergy for classification of these novel drug combinations resulted in a high predictive performance of an AUC of 0.90. Furthermore, we found that all compared methods exhibit low predictive performance when extrapolating to unexplored drugs or cell lines, which we suggest is due to limitations in the size and diversity of the dataset. We envision that DeepSynergy could be a valuable tool for selecting novel synergistic drug combinations.

          Availability and implementation

          DeepSynergy is available via www.bioinf.jku.at/software/DeepSynergy.

          Supplementary information

          Supplementary data are available at Bioinformatics online.

          Related collections

          Most cited references46

          • Record: found
          • Abstract: found
          • Article: not found

          Learning hierarchical features for scene labeling.

          Scene labeling consists of labeling each pixel in an image with the category of the object it belongs to. We propose a method that uses a multiscale convolutional network trained from raw pixels to extract dense feature vectors that encode regions of multiple sizes centered on each pixel. The method alleviates the need for engineered features, and produces a powerful representation that captures texture, shape, and contextual information. We report results using multiple postprocessing methods to produce the final labeling. Among those, we propose a technique to automatically retrieve, from a pool of segmentation components, an optimal set of components that best explain the scene; these components are arbitrary, for example, they can be taken from a segmentation tree or from any family of oversegmentations. The system yields record accuracies on the SIFT Flow dataset (33 classes) and the Barcelona dataset (170 classes) and near-record accuracy on Stanford background dataset (eight classes), while being an order of magnitude faster than competing approaches, producing a $(320\times 240)$ image labeling in less than a second, including feature extraction.
            • Record: found
            • Abstract: found
            • Article: not found

            Mechanisms of drug combinations: interaction and network perspectives.

            Understanding the molecular mechanisms underlying synergistic, potentiative and antagonistic effects of drug combinations could facilitate the discovery of novel efficacious combinations and multi-targeted agents. In this article, we describe an extensive investigation of the published literature on drug combinations for which the combination effect has been evaluated by rigorous analysis methods and for which relevant molecular interaction profiles of the drugs involved are available. Analysis of the 117 drug combinations identified reveals general and specific modes of action, and highlights the potential value of molecular interaction profiles in the discovery of novel multicomponent therapies.
              • Record: found
              • Abstract: found
              • Article: not found

              Combination therapy for treatment of infections with gram-negative bacteria.

              Combination antibiotic therapy for invasive infections with Gram-negative bacteria is employed in many health care facilities, especially for certain subgroups of patients, including those with neutropenia, those with infections caused by Pseudomonas aeruginosa, those with ventilator-associated pneumonia, and the severely ill. An argument can be made for empiric combination therapy, as we are witnessing a rise in infections caused by multidrug-resistant Gram-negative organisms. The wisdom of continued combination therapy after an organism is isolated and antimicrobial susceptibility data are known, however, is more controversial. The available evidence suggests that the greatest benefit of combination antibiotic therapy stems from the increased likelihood of choosing an effective agent during empiric therapy, rather than exploitation of in vitro synergy or the prevention of resistance during definitive treatment. In this review, we summarize the available data comparing monotherapy versus combination antimicrobial therapy for the treatment of infections with Gram-negative bacteria.

                Author and article information

                Oxford University Press
                01 May 2018
                15 December 2017
                15 December 2017
                : 34
                : 9
                : 1538-1546
                [1 ]Institute of Bioinformatics, Johannes Kepler University, 4040 Linz, Austria
                [2 ]Department of Chemistry, Centre for Molecular Science Informatics, University of Cambridge, Cambridge CB2 1EW, UK
                [3 ]Oncology Innovative Medicines and Early Development, AstraZeneca, Hodgkin Building, Chesterford Research Campus, Saffron Walden, Cambs CB10 1XL, UK
                Author notes
                To whom correspondence should be addressed.
                © The Author(s) 2017. Published by Oxford University Press.

                This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

                : 09 August 2017
                : 06 December 2017
                : 14 December 2017
                Page count
                Pages: 9
                Original Papers
                Data and Text Mining

                Bioinformatics & Computational biology
                Bioinformatics & Computational biology


                Comment on this article