Search

For large libraries of small molecules, exhaustive combinatorial chemical screens become infeasible to perform when considering a range of disease models, assay conditions, and dose ranges. Deep learning models have achieved state-of-the-art results in silicofor the prediction of synergy scores. However, databases of drug combinations are biased toward synergistic agents and results do not generalize out of distribution. During 5 rounds of experimentation, we employ sequential model optimization with a deep learning model to select drug combinations increasingly enriched for synergism and active against a cancer cell line—evaluating only ∼5% of the total search space. Moreover, we find that learned drug embeddings (using structural information) begin to reflect biological mechanisms. In silicobenchmarking suggests search queries are ∼5–10×enriched for highly synergistic drug combinations by using sequential rounds of evaluation when compared with random selection or ∼3×when using a pretrained model.

In the present manuscript, the dynamics of the non-spherical test particle in the planar version of spatial collinear restricted four-body problem is presented. Using the numerical methods, the parametric variation of the position of libration points (LPs) is illustrated when the oblateness/prolateness parameters vary in the pre-assumed intervals. Moreover, it is also unveiled how these parameters affect the stability of the LPs and the regions of motion. For all the LPs, we have shown their nature by classifying them not only in two category i.e., as linearly stable and unstable but also as minima, index-1, and index-2 saddles.The performed numerical investigations strongly suggest that the oblateness/prolateness parameters are indeed very influential factors in this dynamical model.