Welcome to SYNPRED: Prediction of Drug Combination Effects in Cancer using Different Synergy Metrics and Ensemble Learning. This webserver allows the user to input two drugs, in the form of two distinct SMILE files and attain the prediction of their synergistic effects on specific cancer cell lines. Please cite A.J. Preto et al. 2022, full paper available at https://doi.org/10.1093/gigascience/giac087.
Due to the high volume of the plots, these are only loaded upon click on the respective button, please feel free to explore our results and hover over the plots to discover additional information. On the button below, select a plot associated to dataset description.
Sankey diagrams showing the results of permutation importance. Click on the dropdown menu to select the Full-agreement model in which the feature contribution was determined.
Background: In cancer research, high-throughput screening technologies produce large amounts of multiomics data from different populations and cell types. However, analysis of such data encounters difficulties due to disease heterogeneity, further exacerbated by human biological complexity and genomic variability. The specific profile of cancer as a disease (or, more realistically, a set of diseases) urges the development of approaches that maximize the effect while minimizing the dosage of drugs. Now is the time to redefine the approach to drug discovery, bringing an Artificial Intelligence-powered informational view that integrates the relevant scientific fields and explores new territories.
Results: Here, we show SYNPRED, an interdisciplinary approach that leverages specifically designed ensembles of AI algorithms, links omics and biophysical traits to predict anticancer drug synergy. It uses five reference models (Bliss, Highest Single Agent, Loewe, Zero Interaction Potency and Combination Sensitivity Score), which, coupled with Machine Learning algorithms, allowed us to attain the ones with the best predictive performance and pinpoint the most appropriate reference model for synergy prediction, often overlooked in similar studies. By using an independent test set, SYNPRED exhibits state-of-the-art performance metrics either in the classification (accuracy – 0.85, precision – 0.91, recall – 0.90, AUROC – 0.80, and F1-score - 0.91) or in the regression models, mainly when using the Combination Sensitivity Score synergy reference model (RMSE – 11.07, MSE – 122.61, Pearson – 0.86, MAE – 7.43, Spearman – 0.87). Moreover, data interpretability was achieved by deploying the most current and robust feature importance approaches. A simple web-based application was constructed, allowing easy access by non-expert researchers.
Conclusions: The performance of SYNPRED rivals that of the existing methods that tackle the same problem, yielding unbiased results trained with one of the most comprehensive datasets available (NCI-ALMANAC). The leveraging of different reference models allowed deeper insights into which of them can be more appropriately used for synergy prediction. The Combination Sensitivity Score clearly stood out with improved performance among the full scope of surveyed approaches and synergy reference models. Furthermore, SYNPRED takes a particular focus on data interpretability, which has been in the spotlight lately when using the most advanced AI techniques. Full paper available at https://doi.org/10.1093/gigascience/giac087.
To access our stand-alone version, you can access our SynPred GitHub repository. To use this version of SynPred, it will be required the download and unpacking of our models, which is available in the button bellow. Please follow the instructions for environment setup and model development as indicated in the GitHub page as it is the proper way to get SynPred running on the stand-alone version.