Owing to the high cost and time associated with developing and validating anti-cancer drugs in clinical trials which is further exacerbated by the 96% failure rate, the development of preclinical computational models that can accurately predict whether a cell line is sensitive or resistant to a particular drug is imperative. The availability of large-scale pharmacogenomics datasets collected via high-throughput screening technologies offers feasible resources to develop robust drug response models and identify the important biomarkers predictive of drug sensitivity. Large language models (LLM), such as the Generative Pre-trained Transformer (GPT-3) from OpenAI, are "taskagnostic models" pre-trained on large textual corpora crawled from the Web that have exhibited unprecedented capabilities on a broad array of NLP tasks.

Anti-cancer drug sensitivity prediction using deep learning models for individual cell line is a significant challenge in personalized medicine. REFINED (REpresentation of Features as Images with NEighborhood Dependencies) CNN (Convolutional Neural Network) based models have shown promising results in drug sensitivity prediction. The primary idea behind REFINED CNN is representing high dimensional vectors as compact images with spatial correlations that can benefit from convolutional neural network architectures. However, the mapping from a vector to a compact 2D image is not unique due to variations in considered distance measures and neighborhoods. In this article, we consider predictions based on ensembles built from such mappings that can improve upon the best single REFINED CNN model prediction. Results illustrated using NCI60 and NCIALMANAC databases shows that the ensemble approaches can provide significant performance improvement as compared to individual models. We further illustrate that a single mapping created from the amalgamation of the different mappings can provide performance similar to stacking ensemble but with significantly lower computational complexity.

In line with recent advances in neural drug design 1.1 Motivation and sensitivity prediction, we propose a novel Discovery of novel compounds with a desired efficacy and architecture for interpretable prediction of anticancer improving existing therapies are key bottlenecks in the pharmaceutical compound sensitivity using a multimodal industry, which fuel the largest R&D business attention-based convolutional encoder. Our model spending of any industry and account for 19% of the total is based on the three key pillars of drug sensitivity: R&D spending worldwide (Petrova, 2014; Goh et al., compounds' structure in the form of a SMILES 2017). Anticancer compounds, in particular, take the lion's sequence, gene expression profiles of tumors and share of drug discovery R&D efforts, with over 34% of all prior knowledge on intracellular interactions from drugs in the global R&D pipeline in 2018 (5,212 of 15,267 protein-protein interaction networks. We demonstrate drugs) (Lloyd et al., 2017). Despite enormous scientific that our multiscale convolutional attentionbased and technological advances in recent years, serendipity still (MCA) encoder significantly outperforms a plays a major role in anticancer drug discovery (Hargrave-baseline model trained on Morgan fingerprints, a Thomas et al., 2012) without a systematic way to accumulate selection of encoders based on SMILES as well and leverage years of R&D to achieve higher success as previously reported state of the art for multimodal rates in drug discovery. On the other hand, there is strong drug sensitivity prediction (R2 0.86 evidence that the response to anticancer therapy is highly dependent and RMSE 0.89).

Motivation: Human genomic datasets often contain sensitive information that limits use and sharing of the data. In particular, simple anonymisation strategies fail to provide sufficient level of protection for genomic data, because the data are inherently identifiable. Differentially private machine learning can help by guaranteeing that the published results do not leak too much information about any individual data point. Recent research has reached promising results on differentially private drug sensitivity prediction using gene expression data. Differentially private learning with genomic data is challenging because it is more difficult to guarantee the privacy in high dimensions. Dimensionality reduction can help, but if the dimension reduction mapping is learned from the data, then it needs to be differentially private too, which can carry a significant privacy cost. Furthermore, the selection of any hyperparameters (such as the target dimensionality) needs to also avoid leaking private information. Results: We study an approach that uses a large public dataset of similar type to learn a compact representation for differentially private learning. We compare three representation learning methods: variational autoencoders, PCA and random projection. We solve two machine learning tasks on gene expression of cancer cell lines: cancer type classification, and drug sensitivity prediction. The experiments demonstrate significant benefit from all representation learning methods with variational autoencoders providing the most accurate predictions most often. Our results significantly improve over previous state-of-the-art in accuracy of differentially private drug sensitivity prediction.

Predicting the efficacy of a drug for a given individual, using high-dimensional genomic measurements, is at the core of precision medicine. However, identifying features on which to base the predictions remains a challenge, especially when the sample size is small. Incorporating expert knowledge offers a promising alternative to improve a prediction model, but collecting such knowledge is laborious to the expert if the number of candidate features is very large. We introduce a probabilistic model that can incorporate expert feedback about the impact of genomic measurements on the sensitivity of a cancer cell for a given drug. We also present two methods to intelligently collect this feedback from the expert, using experimental design and multi-armed bandit models. In a multiple myeloma blood cancer data set (n=51), expert knowledge decreased the prediction error by 8%. Furthermore, the intelligent approaches can be used to reduce the workload of feedback collection to less than 30% on average compared to a naive approach.