Cellrank for Directed Single-Cell Fate Mapping
Summary
The employment of computational trajectory inference facilitates the elucidation of cellular state dynamics derived from single-cell RNA sequencing explorations. Nonetheless, the trajectory inference method is contingent upon the knowledge of a biological process's directionality, consequently restricting its utilization to differentiating systems within the realm of standard development. In this context, the authors introduce CellRank as a means for single-cell fate mapping across various conditions, encompassing regeneration, reprogramming, and disease, where directionality remains nebulous. Their methodology integrates the resilience of trajectory inference with the directional data gleaned from RNA velocity, duly incorporating the progressive and stochastic characteristics of cellular fate resolutions, as well as the ambiguities associated with velocity vectors. Upon examination of pancreatic development datasets, CellRank autonomously identifies the initial, intermediate, and terminal populations, prognosticates potentialities of fate, and illustrates continuous gene expression tendencies along discrete lineage pathways. When implemented in the context of lineage-traced cellular reprogramming datasets, the forecasted fate probabilities accurately reconcile reprogramming outcomes. Furthermore, CellRank postulates a heretofore uncharted dedifferentiation trajectory amid post-injury pulmonary regeneration, revealing hitherto unknown intermediate cellular states, which the authors substantiate through experimental means.
Research Criteria
The scholarly impetus of this treatise lies in the introduction of CellRank, an innovative methodology that amalgamates the steadfastness of similarity-based trajectory inference with directional insights derived from RNA velocity to discern directed, probabilistic state-transition trajectories under standard or perturbed circumstances. Distinct from alternative approaches, CellRank autonomously deduces initial, intermediate, and terminal populations within a single-cell RNA sequencing dataset and calculates fate probabilities, duly considering the stochastic essence of cellular fate determinations and the inherent indeterminacies pertaining to velocity approximations.
Sample Type
They demonstrate CellRank's capabilities on pancreatic endocrine lineage development, cellular reprogramming, and lung regeneration data.
Result—CellRank Combines Cell-Cell Similarity with RNA Velocity to Model Cellular State Transitions
Incorporating RNA velocity and cell-cell similarity, CellRank unites cellular state transition models with RNA velocity to learn directed, probabilistic state change trajectories under normal or perturbed conditions. Distinct from alternative approaches, CellRank autonomously infers initial, intermediate, and terminal cell states from scRNA-seq datasets, calculating fate probabilities whilst considering the stochastic nature of cellular fate decisions and uncertainty in velocity estimations. They employ fate probabilities to uncover potential lineage-driving factors and visualize lineage-specific gene expression trends.
Demonstrating CellRank's capabilities in pancreatic endocrine cell lineage development, they accurately recover initial and terminal states, as well as key driving genes for lineage bias and somatostatin-producing δ-cell differentiation. They show that CellRank transcends normal development, applying it to reprogramming datasets, with predicted fate biases correctly recovering the essential facts derived from lineage tracing. Furthermore, by applying CellRank to lung regeneration, they predict a novel dedifferentiation trajectory and experimentally validate the newly discovered intermediate cell state.
Fig.1 Combining RNA velocity with cell–cell similarity to determine initial and terminal states and compute a global map of cellular fate potential. (Lange, 2022)
Creative Biolabs' Service
Single Cell RNA Sequencing Service
Cell populations are rarely homogeneous and synchronized in their characteristics. Single-cell RNA sequencing aims to uncover the transcriptome diversity in heterogeneous samples. Creative Biolabs offers end-to-end workflows including sample preparation, library construction, and data analysis, maximizing your project flexibility, speed, and data accuracy.
Learn moreCreative Biolabs specializes in providing cutting-edge single-cell RNA sequencing (scRNA-seq) services, tailored to meet the unique demands of our diverse clientele. Leveraging state-of-the-art technologies and leveraging our team of experienced scientists, we strive to deliver unparalleled insights into cellular heterogeneity, gene expression dynamics, and regulatory networks. Our comprehensive scRNA-seq service encompasses experimental design consultation, sample preparation, sequencing, and advanced bioinformatics analysis. By partnering with us, clients can confidently explore cellular complexity and unlock novel discoveries, accelerating their research endeavors and propelling the boundaries of scientific knowledge. Trust our expertise to illuminate the intricacies of the cellular landscape with precision and reliability. For any information, please contact us.
Reference
- Lange, M.; et al. CellRank for directed single-cell fate mapping. Nature Methods. 2022, 19(2): 159-170.
