Researchers increasingly turn to explainable artificial intelligence (XAI) to analyze omics data and gain insights into the underlying biological processes. Yet, given the interdisciplinary nature of the field, many findings have only been shared in their respective research community. An overview of XAI for omics data is needed to highlight promising approaches and help detect common issues. Toward this end, we conducted a systematic mapping study. To identify relevant literature, we queried Scopus, PubMed, Web of Science, BioRxiv, MedRxiv and arXiv. Based on keywording, we developed a coding scheme with 10 facets regarding the studies' AI methods, explainability methods and omics data. Our mapping study resulted in 405 included papers published between 2010 and 2023. The inspected papers analyze DNA-based (mostly genomic), transcriptomic, proteomic or metabolomic data by means of neural networks, tree-based methods, statistical methods and further AI methods. The preferred post-hoc explainability methods are feature relevance (n = 166) and visual explanation (n = 52), while papers using interpretable approaches often resort to the use of transparent models (n = 83) or architecture modifications (n = 72). With many research gaps still apparent for XAI for omics data, we deduced eight research directions and discuss their potential for the field. We also provide exemplary research questions for each direction. Many problems with the adoption of XAI for omics data in clinical practice are yet to be resolved. This systematic mapping study outlines extant research on the topic and provides research directions for researchers and practitioners.

OBJECTIVE: Rising interests in distributed ledger technology (DLT) and genomics have sparked various interdisciplinary research streams with a proliferating number of scattered publications investigating the application of DLT in genomics. This review aims to uncover the current state of research on DLT in genomics, in terms of focal research themes and directions for future research. MATERIALS AND METHODS: We conducted a scoping review and thematic analysis. To identify the 60 relevant papers, we queried Scopus, Web of Science, PubMed, ACM Digital Library, IEEE Xplore, arXiv, and BiorXiv. RESULTS: Our analysis resulted in 7 focal themes on DLT in genomics discussed in literature, namely: (1) Data economy and sharing; (2) Data management; (3) Data protection; (4) Data storage; (5) Decentralized data analysis; (6) Proof of useful work; and (7) Ethical, legal, and social implications. DISCUSSION: Based on the identified themes, we present 7 future research directions: (1) Investigate opportunities for the application of DLT concepts other than Blockchain; (2) Explore people's attitudes and behaviors regarding the commodification of genetic data through DLT-based genetic data markets; (3) Examine opportunities for joint consent management via DLT; (4) Investigate and evaluate data storage models appropriate for DLT; (5) Research the regulation-compliant use of DLT in healthcare information systems; (6) Investigate alternative consensus mechanisms based on Proof of Useful Work; and (7) Explore DLT-enabled approaches for the protection of genetic data ensuring user privacy. CONCLUSION: While research on DLT in genomics is currently growing, there are many unresolved problems. This literature review outlines extant research and provides future directions for researchers and practitioners.

Image-based phenotypic drug profiling is receiving increasing attention in drug discovery and precision medicine. Compared to classical end-point measurements quantifying drug response, image-based profiling enables both the quantification of drug response and characterization of disease entities and drug-induced cell-death phenotypes. Here, we aim to quantify image-based drug responses in patient-derived 3D spheroid tumor cell cultures, tackling the challenges of a lack of single-cell-segmentation methods and limited patient-derived material. Therefore, we investigate deep transfer learning with patient-by-patient fine-tuning for cell-viability quantification. We fine-tune a convolutional neural network (pre-trained on ImageNet) with 210 control images specific to a single training cell line and 54 additional screen -specific assay control images. This method of image-based drug profiling is validated on 6 cell lines with known drug sensitivities, and further tested with primary patient-derived samples in a medium-throughput setting. Network outputs at different drug concentrations are used for drug-sensitivity scoring, and dense-layer activations are used in t-distributed stochastic neighbor embeddings of drugs to visualize groups of drugs with similar cell-death phenotypes. Image-based cell-line experiments show strong correlation to metabolic results ( R ≈ 0.7 ) and confirm expected hits, indicating the predictive power of deep learning to identify drug-hit candidates for individual patients. In patient-derived samples, combining drug sensitivity scoring with phenotypic analysis may provide opportunities for complementary combination treatments. Deep transfer learning with patient-by-patient fine-tuning is a promising, segmentation-free image-analysis approach for precision medicine and drug discovery.

UNLABELLED: : Hilbert curves enable high-resolution visualization of genomic data on a chromosome- or genome-wide scale. Here we present the HilbertCurve package that provides an easy-to-use interface for mapping genomic data to Hilbert curves. The package transforms the curve as a virtual axis, thereby hiding the details of the curve construction from the user. HilbertCurve supports multiple-layer overlay that makes it a powerful tool to correlate the spatial distribution of multiple feature types. AVAILABILITY AND IMPLEMENTATION: The HilbertCurve package and documentation are freely available from the Bioconductor project: http://www.bioconductor.org/packages/devel/bioc/html/HilbertCurve.html CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

UNLABELLED: Parallel heatmaps with carefully designed annotation graphics are powerful for efficient visualization of patterns and relationships among high dimensional genomic data. Here we present the ComplexHeatmap package that provides rich functionalities for customizing heatmaps, arranging multiple parallel heatmaps and including user-defined annotation graphics. We demonstrate the power of ComplexHeatmap to easily reveal patterns and correlations among multiple sources of information with four real-world datasets. AVAILABILITY AND IMPLEMENTATION: The ComplexHeatmap package and documentation are freely available from the Bioconductor project: http://www.bioconductor.org/packages/devel/bioc/html/ComplexHeatmap.html CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Trellis graphics are a visualization method that splits data by one or more categorical variables and displays subsets of the data in a grid of panels. Trellis graphics are broadly used in genomic data analysis to compare statistics over different categories in parallel and reveal multivariate relationships. However, current software packages to produce Trellis graphics have not been designed with genomic data in mind and lack some functionality that is required for effective visualization of genomic data. RESULTS: Here we introduce the gtrellis package which provides an efficient and extensible way to visualize genomic data in a Trellis layout. gtrellis provides highly flexible Trellis layouts which allow efficient arrangement of genomic categories on the plot. It supports multiple-track visualization, which makes it straightforward to visualize several properties of genomic data in parallel to explain complex relationships. In addition, gtrellis provides an extensible framework that allows adding user-defined graphics. CONCLUSIONS: The gtrellis package provides an easy and effective way to visualize genomic data and reveal high dimensional relationships on a genome-wide scale. gtrellis can be flexibly extended and thus can also serve as a base package for highly specific purposes. gtrellis makes it easy to produce novel visualizations, which can lead to the discovery of previously unrecognized patterns in genomic data.

SUMMARY: Circular layout is an efficient way for the visualization of huge amounts of genomic information. Here we present the circlize package, which provides an implementation of circular layout generation in R as well as an enhancement of available software. The flexibility of this package is based on the usage of low-level graphics functions such that self-defined high-level graphics can be easily implemented by users for specific purposes. Together with the seamless connection between the powerful computational and visual environment in R, circlize gives users more convenience and freedom to design figures for better understanding genomic patterns behind multi-dimensional data. AVAILABILITY AND IMPLEMENTATION: circlize is available at the Comprehensive R Archive Network (CRAN): http://cran.r-project.org/web/packages/circlize/