A computational pipeline for identifying gene targets and signalling pathways in cancer cells to improve lymphocyte infiltration and immune checkpoint therapy efficacy.

BACKGROUND: Tumour-infiltrating lymphocytes (TILs) are crucial for effective immune checkpoint blockade (ICB) therapy in solid tumours. However, ∼70% of these tumours exhibit poor lymphocyte infiltration, rendering ICB therapies less effective. METHODS: We developed a bioinformatics pipeline integrating multiple previously unconsidered factors or datasets, including tumour cell immune-related pathways, copy number variation (CNV), and single tumour cell sequencing data, as well as tumour mRNA-seq data and patient survival data, to identify targets that can potentially improve T cell infiltration and enhance ICB efficacy. Furthermore, we conducted wet-lab experiments and successfully validated one of the top-identified genes. FINDINGS: We applied this pipeline in solid tumours of the Cancer Genome Atlas (TCGA) and identified a set of genes in 18 cancer types that might potentially improve lymphocyte infiltration and ICB efficacy, providing a valuable drug target resource to be further explored. Importantly, we experimentally validated SUN1, which had not been linked to T cell infiltration and ICB therapy previously, but was one of the top-identified gene targets among 3 cancer types based on the pipeline, in a mouse colon cancer syngeneic model. We showed that Sun1 KO could significantly enhance antigen presentation, increase T-cell infiltration, and improve anti-PD1 treatment efficacy. Moreover, with a single-cell multiome analysis, we identified subgene regulatory networks (sub-GRNs) showing Stat proteins play important roles in enhancing the immune-related pathways in Sun1-KO cancer cells. INTERPRETATION: This study not only established a computational pipeline for discovering new gene targets and signalling pathways in cancer cells that block T-cell infiltration, but also provided a gene target pool for further exploration in improving lymphocyte infiltration and ICB efficacy in solid tumours. FUNDING: A full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

The assessment of efficient representation of drug features using deep learning for drug repositioning.

BACKGROUND: De novo drug discovery is a time-consuming and expensive process. Nowadays, drug repositioning is utilized as a common strategy to discover a new drug indication for existing drugs. This strategy is mostly used in cases with a limited number of candidate pairs of drugs and diseases. In other words, they are not scalable to a large number of drugs and diseases. Most of the in-silico methods mainly focus on linear approaches while non-linear models are still scarce for new indication predictions. Therefore, applying non-linear computational approaches can offer an opportunity to predict possible drug repositioning candidates. RESULTS: In this study, we present a non-linear method for drug repositioning. We extract four drug features and two disease features to find the semantic relations between drugs and diseases. We utilize deep learning to extract an efficient representation for each feature. These representations reduce the dimension and heterogeneity of biological data. Then, we assess the performance of different combinations of drug features to introduce a pipeline for drug repositioning. In the available database, there are different numbers of known drug-disease associations corresponding to each combination of drug features. Our assessment shows that as the numbers of drug features increase, the numbers of available drugs decrease. Thus, the proposed method with large numbers of drug features is as accurate as small numbers. CONCLUSION: Our pipeline predicts new indications for existing drugs systematically, in a more cost-effective way and shorter timeline. We assess the pipeline to discover the potential drug-disease associations based on cross-validation experiments and some clinical trial studies.