CellVGAE: an unsupervised scRNA-seq analysis workflow with graph attention networks.

MOTIVATION: Single-cell RNA sequencing allows high-resolution views of individual cells for libraries of up to millions of samples, thus motivating the use of deep learning for analysis. In this study, we introduce the use of graph neural networks for the unsupervised exploration of scRNA-seq data by developing a variational graph autoencoder architecture with graph attention layers that operates directly on the connectivity between cells, focusing on dimensionality reduction and clustering. With the help of several case studies, we show that our model, named CellVGAE, can be effectively used for exploratory analysis even on challenging datasets, by extracting meaningful features from the data and providing the means to visualize and interpret different aspects of the model. RESULTS: We show that CellVGAE is more interpretable than existing scRNA-seq variational architectures by analysing the graph attention coefficients. By drawing parallels with other scRNA-seq studies on interpretability, we assess the validity of the relationships modelled by attention, and furthermore, we show that CellVGAE can intrinsically capture information such as pseudotime and NF-ĸB activation dynamics, the latter being a property that is not generally shared by existing neural alternatives. We then evaluate the dimensionality reduction and clustering performance on 9 difficult and well-annotated datasets by comparing with three leading neural and non-neural techniques, concluding that CellVGAE outperforms competing methods. Finally, we report a decrease in training times of up to × 20 on a dataset of 1.3 million cells compared to existing deep learning architectures. AVAILABILITYAND IMPLEMENTATION: The CellVGAE code is available at https://github.com/davidbuterez/CellVGAE. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Genome-scale screens identify factors regulating tumor cell responses to natural killer cells.

To systematically define molecular features in human tumor cells that determine their degree of sensitivity to human allogeneic natural killer (NK) cells, we quantified the NK cell responsiveness of hundreds of molecularly annotated 'DNA-barcoded' solid tumor cell lines in multiplexed format and applied genome-scale CRISPR-based gene-editing screens in several solid tumor cell lines, to functionally interrogate which genes in tumor cells regulate the response to NK cells. In these orthogonal studies, NK cell-sensitive tumor cells tend to exhibit 'mesenchymal-like' transcriptional programs; high transcriptional signature for chromatin remodeling complexes; high levels of B7-H6 (NCR3LG1); and low levels of HLA-E/antigen presentation genes. Importantly, transcriptional signatures of NK cell-sensitive tumor cells correlate with immune checkpoint inhibitor (ICI) resistance in clinical samples. This study provides a comprehensive map of mechanisms regulating tumor cell responses to NK cells, with implications for future biomarker-driven applications of NK cell immunotherapies.

Variational Autoencoders for Cancer Data Integration: Design Principles and Computational Practice.

International initiatives such as the Molecular Taxonomy of Breast Cancer International Consortium are collecting multiple data sets at different genome-scales with the aim to identify novel cancer bio-markers and predict patient survival. To analyze such data, several machine learning, bioinformatics, and statistical methods have been applied, among them neural networks such as autoencoders. Although these models provide a good statistical learning framework to analyze multi-omic and/or clinical data, there is a distinct lack of work on how to integrate diverse patient data and identify the optimal design best suited to the available data.In this paper, we investigate several autoencoder architectures that integrate a variety of cancer patient data types (e.g., multi-omics and clinical data). We perform extensive analyses of these approaches and provide a clear methodological and computational framework for designing systems that enable clinicians to investigate cancer traits and translate the results into clinical applications. We demonstrate how these networks can be designed, built, and, in particular, applied to tasks of integrative analyses of heterogeneous breast cancer data. The results show that these approaches yield relevant data representations that, in turn, lead to accurate and stable diagnosis.