Single-Cell Techniques and Deep Learning in Predicting Drug Response

A comprehensive understanding of heterogeneous tumor subpopulations will benefit drug sensitivity prediction and combination drug treatment design. Deep learning models are powerful and extensively used in drug sensitivity prediction and in inferring drug–target interactions. Single-cell sequencing techniques offer precise and accurate profiling of tumor subpopulations and reveal subtle differences in their response to drug treatments. Applying deep transfer learning to predict drug sensitivity allows us to not only take advantage of prior knowledge obtained from massive bulk sequencing data but also utilize the heterogeneous landscapes generated by single-cell sequencing techniques. The integration of single-cell multi-omic data for drug sensitivity prediction using transfer learning methods poses a special challenge. Rapidly developing single-cell sequencing analyses produce more comprehensive profiles of the genomic, transcriptomic, and epigenomic heterogeneity of tumor subpopulations than do traditional bulk sequencing analyses. Moreover, single-cell techniques allow the response of a tumor to drug exposure to be more thoroughly investigated. Deep learning (DL) models have successfully extracted features from complex bulk sequence data to predict drug responses. We review recent innovations in single-cell technologies and DL-based approaches related to drug sensitivity predictions. We believe that, by using insights from bulk sequence data, deep transfer learning (DTL) can facilitate the use of single-cell data for training superior DL-based drug prediction models. Rapidly developing single-cell sequencing analyses produce more comprehensive profiles of the genomic, transcriptomic, and epigenomic heterogeneity of tumor subpopulations than do traditional bulk sequencing analyses. Moreover, single-cell techniques allow the response of a tumor to drug exposure to be more thoroughly investigated. Deep learning (DL) models have successfully extracted features from complex bulk sequence data to predict drug responses. We review recent innovations in single-cell technologies and DL-based approaches related to drug sensitivity predictions. We believe that, by using insights from bulk sequence data, deep transfer learning (DTL) can facilitate the use of single-cell data for training superior DL-based drug prediction models. a technique to determine chromatin accessibility across the genome. external factors associated with experiments can influence the data produced and result in inaccurate conclusions. This effect represents the systematic technical differences when samples are processed and measured in different batches. examines the sequence information of bulk samples, usually containing multiple cells. the full complement of transcriptional targets that are regulated by a protein. These can include either direct physical targets, transcription factors and cofactors, or indirect targets for signal transduction. chromatin immunoprecipitation with high-throughput sequencing, a technique to identify genome-wide binding sites in DNA for transcription factors and other proteins. the use of more than one drug to treat a disease; this usually reduces the development of drug resistance. a network constructed on several layers of restricted Boltzmann machines. an artificial intelligence function that mimics the workings of the human brain in processing unstructured data through many layers of neural networks. cells express efflux pumps that are able to move drugs out of the cell. cancer cells may express enzymes to break down or modify drugs, leading to their loss of function. a reduction in the effectiveness of a medication, such as an antimicrobial or an antineoplastic, in treating disease. the pharmacodynamic (PD) response to a drug; this includes all the effects of the drug on any physiological and/or pathological processes. the concentration of a drug that inhibits cell growth. a molecular technique that uses fluorescent probes that can specifically bind to DNA/RNA/proteins to visualize the location of those targets. immune checkpoints are accessory molecules that regulate the activation and silencing of T cells. ICB treatment can release inherent limits on the activation and maintenance of T cell effector function by inhibiting the immune checkpoints. the small number of cancer cells that survive drug treatment and usually result in relapse. a genomic method for analyzing B/T cell receptors that are uniquely expressed on the B/T cell surface. The diverse range of BCRs/TCRs expressed by the total B/T cell population of an individual is termed the B/T cell receptor repertoire. examines sequence information from individual cells with optimized next-generation sequencing technologies, providing higher resolution of cellular diversity. cancer cells may modify or downregulate the expression of proteins that are targeted by drugs. an analytical method that can simulate not only system function but also system structure.