CellOracle: Dissecting cell identity via network inference and in silico gene perturbation
Kenji KamimotoiD, Christy M. HoffmanniD, Samantha A. MorrisiD
Abstract
Here, we present CellOracle, a computational tool that integrates single-cell transcriptome and epigenome profiles to infer gene regulatory networks (GRNs), critical regulators of cell identity. Leveraging inferred GRNs, we simulate gene expression changes in response to transcription factor (TF) perturbation, enabling network configurations to be interrogated in silico, facilitating their interpretation. We validate the efficacy of CellOracle to recapitulate known regulatory changes across hematopoiesis, correctly predicting the outcomes of well-characterized TF perturbations. Integrating CellOracle analysis with lineage tracing of direct reprogramming reveals distinct network configurations underlying different reprogramming failure modes. Furthermore, analysis of GRN reconfiguration along successful reprogramming trajectories identifies new factors to enhance target cell yield, uncovering a role for the AP-1 subunit Fos, with the hippo signaling effector, Yap1. Together, these results demonstrate the efficacy of CellOracle to infer and interpret cell-type-specific GRN configurations, at high-resolution, promoting new mechanistic insights into the regulation and reprogramming of cell identity.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 81
- Organism
- Assay
- Research Focus
- Approach
Abstract summary
Claude Opus 5.5 · xhigh effort · 18 Oct 2026 · from the abstract · CC BY-NC-ND 4.0
Gene regulatory networks inferred from single-cell transcriptome and epigenome data are perturbed in silico to predict transcription-factor perturbation outcomes and to nominate factors that raise reprogramming yield.
Gene regulatory networks (GRNs) are presented as critical regulators of cell identity, but a network inferred from single-cell data is hard to interpret on its own, leaving open what a given transcription factor (TF) does. CellOracle couples network inference to a simulation step, so a TF can be perturbed in silico and the resulting gene expression changes examined without running the experiment first.
Key results
SuggestedApplied to hematopoiesis, the simulations reproduced regulatory changes that were already known and correctly predicted the outcomes of well-characterized TF perturbations.
Reported as a validation across hematopoiesis with correct prediction of well-characterized TF perturbations; no accuracy figures, cell numbers or comparison with other GRN inference methods appear in the abstract. · Abstract
SuggestedPairing the inferred networks with lineage tracing of direct reprogramming separated different ways reprogramming fails, each with its own network configuration.
Stated as distinct network configurations underlying distinct failure modes; the failure modes, the reprogramming system and the number of cells or lineages are not specified in the available text. · Abstract
SuggestedTracking how the networks change along trajectories that reprogram successfully pointed to the AP-1 subunit Fos, together with the Hippo pathway effector Yap1, as a way to increase the yield of target cells.
The abstract names Fos and Yap1 as factors identified for enhancing target cell yield but gives no yield measurement, effect size or validation experiment detail. · Abstract
Methods, data and reproducibility
CellOracle combines single-cell transcriptome and epigenome profiles to infer cell-type-specific gene regulatory networks, then uses those networks to simulate the expression changes expected after perturbing a transcription factor, so network states can be probed in silico.
- single-cell transcriptome and epigenome profiles spanning hematopoiesis
- lineage-tracing data from a direct reprogramming experiment, including successful and failed trajectories
- Methods & tools
- CellOracle
Limitations (2)
- The validation is against TF perturbations whose outcomes were already known, with no quantitative accuracy and no comparison against other GRN inference or perturbation-prediction methods in the available text. · noted in this summary
- The gain in target cell yield attributed to Fos and Yap1 is asserted without any number, so the size of the effect and how it was measured cannot be assessed here. · noted in this summary
Who should read:Read this if you infer regulatory networks from single-cell data and want to simulate transcription-factor perturbations to prioritise candidates for reprogramming experiments.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Gene Regulatory Network Reconfiguration in Direct Lineage Reprogramming
Kenji Kamimoto, Mohd Tayyab Adil, … Samantha A. Morris·Stem Cell ReportsPeer-reviewed·Dec 29, 2022
- Fatecode: Cell fate regulator prediction using classification autoencoder perturbation
Mehrshad Sadria, Anita T. Layton, … Gary D. Bader·bioRxiv·Dec 16, 2022
- Reconstruction of dynamic regulatory networks reveals signaling-induced topology changes associated with germ layer specification
Emily Y. Su, Abby Spangler, … Patrick Cahan·Stem Cell ReportsPeer-reviewed·Jan 27, 2022
- Critical issues found in "Dissecting cell identity via network inference and in silico gene perturbation"
Jalil Nourisa, Antoine Passemiers, Sven Tomforde·bioRxiv·Oct 16, 2024
- CRGEM: Cellular Reprogramming using mechanism-driven Gene Expression Modulation
Avani Mahadik, Nitin Nitin, … Vivek Singh·bioRxiv·May 12, 2023
- CellPolaris: Decoding Cell Fate through Generalization Transfer Learning of Gene Regulatory Networks
Guihai Feng, Xin Qin, … Xin Li·bioRxiv·Sep 25, 2023