The increasing availability of large-scale single-cell atlases has enabled the detailed description of cell states. In parallel, advances in deep learning allow rapid analysis of newly generated query datasets by mapping them into reference atlases. However, existing data transformations learned to map query data are not easily explainable using biologically known concepts such as genes or pathways. Here we propose expiMap, a biologically informed deep-learning architecture that enables single-cell reference mapping. ExpiMap learns to map cells into biologically understandable components representing known ‘gene programs’. The activity of each cell for a gene program is learned while simultaneously refining them and learning de novo programs. We show that expiMap compares favourably to existing methods while bringing an additional layer of interpretability to integrative single-cell analysis. Furthermore, we demonstrate its applicability to analyse single-cell perturbation responses in different tissues and species and resolve responses of patients who have coronavirus disease 2019 to different treatments across cell types.

Background Single-cell metabolic studies bring new insights into cellular function, which can often not be captured on other omics layers. Metabolic information has wide applicability, such as for the study of cellular heterogeneity or for the understanding of drug mechanisms and biomarker development. However, metabolic measurements on single-cell level are limited by insufficient scalability and sensitivity, as well as resource intensiveness, and are currently not possible in parallel with measuring transcript state, commonly used to identify cell types. Nevertheless, because omics layers are strongly intertwined, it is possible to make metabolic predictions based on measured data of more easily measurable omics layers together with prior metabolic network knowledge. Scope of Review We summarize the current state of single-cell metabolic measurement and modeling approaches, motivating the use of computational techniques. We review three main classes of computational methods used for prediction of single-cell metabolism: pathway-level analysis, constraint-based modeling, and kinetic modeling. We describe the unique challenges arising when transitioning from bulk to single-cell modeling. Finally, we propose potential model extensions and computational methods that could be leveraged to achieve these goals. Major Conclusions Single-cell metabolic modeling is a rising field that provides a new perspective for understanding cellular functions. The presented modeling approaches vary in terms of input requirements and assumptions, scalability, modeled metabolic layers, and newly gained insights. We believe that the use of prior metabolic knowledge will lead to more robust predictions and will pave the way for mechanistic and interpretable machine-learning models.

Dictyostelium development begins with single-cell starvation and ends with multicellular fruiting bodies. Developmental morphogenesis is accompanied by sweeping transcriptional changes, encompassing nearly half of the 13,000 genes in the genome. We performed time-series RNA-sequencing analyses of the wild type and 20 mutants to explore the relationships between transcription and morphogenesis. These strains show developmental arrest at different stages, accelerated development, or atypical morphologies. Considering eight major morphological transitions, we identified 1371 milestone genes whose expression changes sharply between consecutive transitions. We also identified 1099 genes as members of 21 regulons, which are groups of genes that remain coordinately regulated despite the genetic, temporal, and developmental perturbations. The gene annotations in these groups validate known transitions and reveal new developmental events. For example, DNA replication genes are tightly coregulated with cell division genes, so they are expressed in mid-development although chromosomal DNA is not replicated. Our data set includes 486 transcriptional profiles that can help identify new relationships between transcription and development and improve gene annotations. We show its utility by showing that cycles of aggregation and disaggregation in allorecognition-defective mutants involve dedifferentiation. We also show sensitivity to genetic and developmental conditions in two commonly used actin genes, act6 and act15, and robustness of the coaA gene. Finally, we propose that gpdA is a better mRNA quantitation standard because it is less sensitive to external conditions than commonly used standards. The data set is available for democratized exploration through the web application dictyExpress and the data mining environment Orange.