Prediction of drug combination effects with a minimal set of experiments

Abstract

High-throughput drug combination screening provides a systematic strategy to discover unexpected combinatorial synergies in pre-clinical cell models. However, phenotypic combinatorial screening with multi-dose matrix assays is experimentally expensive, especially when the aim is to identify selective combination synergies across a large panel of cell lines or patient samples. Here we implemented DECREASE, an efficient machine learning model that requires only a limited set of pairwise dose-response measurements for accurate prediction of drug combination synergy and antagonism. Using a compendium of 23,595 drug combination matrices tested in various cancer cell lines, and malaria and Ebola infection models, we demonstrate how cost-effective experimental designs with DECREASE capture almost the same degree of information for synergy and antagonism detection as the fully-measured dose-response matrices. Measuring only the diagonal of the matrix provides an accurate and practical option for combinatorial screening. The open-source web-implementation enables applications of DECREASE to both pre-clinical and translational studies.

Adjuvantation of an Influenza Hemagglutinin Antigen with TLR4 and NOD2 Agonists Encapsulated in Poly(D,L-Lactide-Co-Glycolide) Nanoparticles Enhances Immunogenicity and Protection against Lethal Influenza Virus Infection in Mice

“…Contour plot (lower figure) depicts calculated ZIP synergy score with positive and negative values denoting synergy and antagonism, respectively. For estimation of outlier measurements cNMF algorithm [ 44 ] implemented in SynergyFinder was utilized. Yellow square depicts synergy score achieved using combination of TLR4 and NOD2 agonists at doses 1µg/mL and 20µg/mL, respectively, chosen for encapsulation into single dose of PLGA NPs.…”

comboFM: leveraging multi-way interactions for systematic prediction of drug combination effects

“…comboFM strongly benefits from this information due to its capability to interpolate in the space of dose-response matrices through the computation of latent factors representing similarly behaving drug combinations from the response tensor alone (similarly to recommender systems grouping users by the movies they have liked in the past), while the drug and cell line descriptors merely fine-tune the predictions. It is plausible that by careful experimental design, one could minimize the number of monotherapy responses needed for accurate dose-response matrix prediction [42] whilst maintaining the accuracy of the comboFM model, which we leave as an interesting future research topic. However, in a scenario where one would like to perform predictions for completely new molecules with no prior monotherapy or combination response data in any cell line, the computed latent factors are no longer helpful, and none of the methods could perform well with the current design (Supp.…”

Neural signaling modulates metabolism of gastric cancer

“…To assess whether the drug combinations acted synergistically, we calculated Bliss synergy scores for RAD001 + CPI-613 combinations using the SynergyFinder web-application (Ianevski et al, 2017). Synergy scores were quantified as an average excess over expected drug combination effect given by the Bliss reference model (Ianevski et al, 2019). Bliss Independence model was used because the two drugs (i.e.…”