High-content phenotypic screens provide a powerful strategy for studying biological systems, but the scale of possible perturbations and cell states makes exhaustive experiments unfeasible. Computational models that are trained on existing data and extrapolate to correctly predict outcomes in unseen contexts have the potential to accelerate biological discovery. Here, we present CellFlow, a flexible framework based on flow matching that can model single cell phenotypes induced by complex perturbations. We apply CellFlow to various phenotypic screens, accurately predicting expression responses to a wide range of perturbations, including cytokine stimulation, drug treatments and gene knockouts. CellFlow successfully modeled developmental perturbations at the whole-embryo scale and guided cell fate and organoid engineering by predicting heterogeneous cell populations arising from combinatorial morphogen treatments and by performing a virtual organoid protocol screen. Taken together, CellFlow has the potential to accelerate discovery from phenotypic screens by learning from existing data and generating phenotypes induced by unseen conditions.
Across biological systems, cells undergo coordinated changes in gene expression, resulting in transcriptome dynamics that unfold within a low-dimensional manifold. While low-dimensional dynamics can be extracted using RNA velocity, these algorithms can be fragile and rely on heuristics lacking statistical control. Moreover, the estimated vector field is not dynamically consistent with the traversed gene expression manifold. To address these challenges, we introduce a Bayesian model of RNA velocity that couples velocity field and manifold estimation in a reformulated, unified framework, identifying the parameters of an explicit dynamical system. Focusing on the cell cycle, we implement VeloCycle to study gene regulation dynamics on one-dimensional periodic manifolds and validate its ability to infer cell cycle periods using live imaging. We also apply VeloCycle to reveal speed differences in regionally defined progenitors and Perturb-seq gene knockdowns. Overall, VeloCycle expands the single-cell RNA sequencing analysis toolkit with a modular and statistically consistent RNA velocity inference framework.
The maintenance of plasma pH is critical for life in all organisms. The kidney plays a critical role in acid–base regulation in vertebrates by controlling the plasma concentration of bicarbonate. The receptor tyrosine kinase IRR (insulin receptor-related receptor) is expressed in renal β-intercalated cells and is involved in alkali sensing due to its ability to autophosphorylate under alkalization of extracellular medium (pH > 7.9). In mice with a knockout of the insrr gene, which encodes for IRR, urinary bicarbonate secretion in response to alkali loading is impaired. The specific regulatory mechanisms in the kidney that are under the control of IRR remain unknown. To address this issue, we analyzed and compared the kidney transcriptomes of wild-type and insrr knockout mice under basal or bicarbonate-loaded conditions. Transcriptomic analyses revealed a differential regulation of a number of genes in the kidney. Using TaqMan real-time PCR, we confirmed different expressions of the slc26a4, rps7, slc5a2, aqp6, plcd1, gapdh, rny3, kcnk5, slc6a6 and atp6v1g3 genes in IRR knockout mice. Also, we found that the expression of the kcnk5 gene is increased in wild-type mice after bicarbonate loading but not in knockout mice. Gene set enrichment analysis between the IRR knockout and wild-type samples identified that insrr knockout causes alterations in expression of genes related mostly to the ATP metabolic and electron transport chain processes.
Glucose and lipid metabolism are crucial functional systems in eukaryotes. A large number of experimental studies both in animal models and humans have shown that long non-coding RNAs (lncRNAs) play an important role in glucose and lipid metabolism. Previously, human lncRNA DEANR1/linc00261 was described as a tumor suppressor that regulates a variety of biological processes such as cell proliferation, apoptosis, glucose metabolism and tumorigenesis. Here we report that murine lncRNA Falcor/LL35, a proposed functional analog of human DEANR1/linc00261, is predominantly expressed in murine normal hepatocytes and downregulated in HCC and after partial hepatectomy. The application of high-throughput approaches such as RNA-seq, LC-MS proteomics, lipidomics and metabolomics analysis allowed changes to be found in the transcriptome, proteome, lipidome and metabolome of hepatocytes after LL35 depletion. We revealed that LL35 is involved in the regulation of glycolysis and lipid biosynthesis in vitro and in vivo. Moreover, LL35 affects Notch and NF-κB signaling pathways in normal hepatocytes. All observed changes result in the decrease in the proliferation and migration of hepatocytes. We demonstrated similar phenotype changes between murine LL35 and human linc00261 depletion in vitro and in vivo that opens the opportunity to translate results for LL35 from a liver murine model to possible functions of human lncRNA linc00261.
Motivation: Epistasis, the context-dependence of the contribution of an amino acid substitution to fitness, is common in evolution. To detect epistasis, fitness must be measured for at least four genotypes: the reference genotype, two different single mutants and a double mutant with both of the single mutations. For higher-order epistasis of the order n, fitness has to be measured for all 2 n genotypes of an n-dimensional hypercube in genotype space forming a “combinatorially complete dataset”. So far, only a handful of such datasets have been produced by manual curation. Concurrently, random mutagenesis experiments have produced measurements of fitness and other phenotypes in a high-throughput manner, potentially containing a number of combinatorially complete datasets. Results: We present an effective recursive algorithm for finding all hypercube structures in random mutagenesis experimental data. To test the algorithm, we applied it to the data from a recent HIS3 protein dataset and found all 199,847,053 unique combinatorially complete genotype combinations of dimensionality ranging from two to twelve. The algorithm may be useful for researchers looking for higher-order epistasis in their high-throughput experimental data. Availability: https://github.com/ivankovlab/HypercubeME.git .