Causal Network Discovery from Interventional Count Data with Latent Linear DAGs

arXiv:2603.25838 · stat.ME · Submitted 2026-03-26 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Statistics. The day in statistics: methods, theory and the inference under them.

Tom: We're looking at the paper: "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs".

Jane: Here's the gist of the paper. The paper studies identification and estimation in latent linear structural causal models for interventional count data using a latent linear Gaussian directed acyclic graph (DAG) model with Poisson measurement error.

Tom: This work is motivated by the challenges of analyzing Perturb-seq data, which are count-valued and subject to substantial measurement error and latent state heterogeneity.

Jane: Model Formulation The authors propose a latent linear Gaussian DAG model with Poisson measurement error that explicitly separates the latent causal structure from the observed counts.

Tom: The observed gene expression count Xpmqj of gene j in environment m is modeled as a Poisson random variable conditional on the library size L, cell-level covariates C, and the latent true expression level Zpmqj: XpmqjL, C, Zpmqj Poisson(L exp(s pC + Zpmqj), j = 1,..., p.

Title and authors: Tom: So this paper is called "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs," and it’s looking at how to find those causal links in gene networks when you have counts and a lot of messy measurement error.

Jane: Right, so the core problem they are tackling is that we usually use observational data, but that just gives us a Markov equivalence class, not the actual directed graph structure because we don't know which edges are truly causal Lu which is exactly right; observational data alone isn't enough to uniquely identify the causal directions of edges Jane and this paper suggests using interventional data for gene perturbation studies to fix that problem.

Tom: But they’re not just using any interventional data; they are dealing with count-valued data, which means we have Poisson measurement error, and also latent state heterogeneity, which is like there are hidden cellular states influencing things Meng which is a big headache in biology because it introduces confounding that existing methods often don't handle well.

Jane: So the paper proposes a specific model—a latent linear Gaussian directed acyclic graph model with Poisson measurement error—that tries to separate the true causal structure from all that noise and hidden stuff Lu which is interesting because they’re explicitly modeling both the count process and the underlying causal mechanism at once.

Tom: The key idea for identification, according to them, is using a mean-shift intervention design where every gene gets perturbed at least once, and they establish population-level identifiability based on the mean difference between environments Lu which means if you perturb a gene in one setting versus another, that shift should reveal information about the causal mechanism.

Jane: That sounds like a practical test for any new method; so how do they actually estimate this structure, and what are the results of their estimation procedure?

Tom: Well, their estimation involves four steps: first removing those measurement layer effects using a generalized linear model with a log link and an offset related to the library size Meng then mapping observed moments to latent means using a Poisson–lognormal model through empirical moments Lu and finally recovering the columns of that mean shift vector by looking at cross-environment mean shifts Jane.

Lu: And for the final step, they use an optimization problem solved by the alternating direction method of multipliers, or ADMM, to estimate the causal matrix A while respecting sparsity and acyclicity constraints Tom which is a smart way to handle those complex structural requirements in high-dimensional settings.

Title and authors: Jane: So if we take all that together for a moment, what does this mean practically for someone trying to actually build a gene regulatory network from real Perturb-seq data?

Tom: It means they’ve derived non-asymptotic estimation error bounds for the causal coefficient matrix A under a specific condition called beta-min type conditions on the interventional strength and signal strength Lu which gives them finite sample error bounds for estimating A under interventional Poisson count data, as stated in theorem five point two.

Meng: But I always wonder about how robust this is when we move to real traffic at scale; what are the computational costs of running this ADMM optimization on a truly large network?

Tom: Meng, the paper shows that performance improves systematically with stronger mean shifts and lower graph density, and larger interventional sample sizes Lu which suggests that getting good data quality through better interventions helps stabilize the estimation procedure.

Jane: What about the biological interpretation of these results? Does this method just give us a list of edges, or does it help us understand the actual biology behind those connections?

Lu: They do more than just list edges because they provide finite-sample recovery for the latent causal DAG via thresholding the estimated coefficient matrix A under theorem five point three Tom which means you get a concrete, sparse graph structure that they claim is recoverable.

Tom: And their simulation studies show that their proposed method consistently outperforms methods like NOTEARS, LiNGAM, and MRS-PoissonSEM across nearly all regimes in terms of both the F1 score and SHD Lu which is a direct comparison showing where the practical advantage lies.

Jane: I'm curious about what they say about biological plausibility; can this method tell us if the inferred network actually looks like something we’d expect to see in real biology?

Lu: They show that the inferred gene regulatory network supports biological plausibility by demonstrating that hub genes like UBTF and GATA1 have edges supported by ChIP-seq data Jane which is a strong piece of evidence linking the model back to existing experimental data.

Tom: So, to wrap up on the mechanics, they essentially propose a way to disentangle latent confounding using mean shifts in count data models and then use an ADMM optimization with sparsity constraints to recover the causal structure A Jane which is a solid framework for tackling that measurement error issue we talked about earlier.

Title and authors: Meng: From my side, the main thing I worry about is that the performance depends on having enough interventional samples, and if you can't get those, this method might just be too unstable for production deployment Tom.

Jane: That’s a fair point; they do mention that stability is tied to the sample size of per-intervention data Lu, which means practical application requires careful experimental design.

Tom: Anyway, looking at what they conclude with the paper "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs," it seems like the main implication is that we can now identify latent causal structures in complex gene networks using interventional count data even when we have significant measurement error and hidden confounding Lu.

Jane: Exactly, they show a path forward for analyzing Perturb-seq data by providing a method to separate the latent expression levels from the Poisson measurement process Tom.

Lu: And their work on estimation, particularly using ADMM and those finite sample error bounds, gives us confidence that this is not just a theoretical exercise but something with actual recovery guarantees Jane.

Tom: So we've covered the identification strategy, the four-step estimation procedure, and the simulation results showing it beats other methods like NOTEARS in terms of F1 score Lu.

Meng: It sounds like they are promising a high-precision tool for analyzing these count data types across various thresholds Jane, but I'm still waiting to see how well it handles scenarios where the latent state heterogeneity is extremely complex.

Lu: They do flag that their framework accommodates latent confounding without explicitly modeling the confounders by exploiting those interventional mean shifts Tom, which is a way of handling uncertainty without needing an explicit model for every hidden variable.

Jane: And to wrap this up, the paper "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs" gives us a concrete framework for identifying causal links in gene networks by leveraging interventional count data and latent linear models Lu.

Tom: It’s a lot of machinery—the Poisson process, the Gaussian structural model, the ADMM optimization—but the results show it can recover the true directed acyclic graph A under certain conditions Lu.

Meng: For those of us who have to deploy this in a real system, I think I'll keep an eye on those finite sample bounds and see if they hold up when we try to run it on massive datasets Tom.

Jane: We’ll definitely be following their work as they move toward applying these methods to other count models like zero-inflated or dropout-aware Poisson models in the future Lu.

The paper's summary: Tom: So, to recap this paper, they’re taking messy count data from something like Perturb-seq experiments and using this latent linear model structure—a directed acyclic graph with Poisson measurement error—to figure out the true underlying causal connections between genes.

Jane: That's a lot of noise there. What they're really doing is separating the signal of the gene expression itself from the noise introduced by how we measure those counts, and then using controlled interventions to untangle all that hidden stuff, like latent confounding factors.

Lu: Exactly. The paper proves that you can identify this entire directed acyclic graph structure even when you have unmeasured confounding variables lurking in the background because they exploit the mean shift between different experimental setups—the control versus the intervention environment.

Meng: Wait, so if we're using these interventional count data, what are they actually proving? Are they just guessing a direction or is there a formal guarantee that this latent structure is identifiable?

Tom: They establish population-level identifiability under a mean-shift intervention design where every gene gets perturbed at least once. This means if you perturb each node in the graph even once, you can figure out the true causal mechanism A from both your observational data and your interventional data simultaneously.

Jane: That’s a big deal because it moves us past just looking at correlations in the data and gives us a way to map out the actual regulatory pathways as they are supposed to function.

Lu: And their estimation procedure is quite rigorous, involving four steps—cleaning up those measurement layer effects, mapping observed moments to latent means with a Poisson lognormal model, recovering those mean shift vectors using cross-environment contrasts, and finally estimating the causal matrix A using an optimization method called ADMM that respects sparsity and acyclicity constraints.

Meng: ADMM sounds computationally heavy. What’s the practical cost here? If I have thousands of genes, will this run in a reasonable time on our production hardware?

Tom: The paper shows it's feasible because they derive non-asymptotic estimation error bounds for the causal coefficient matrix A under beta-min type conditions, which gives them finite sample error bounds. Plus, their simulation studies show performance improves with stronger mean shifts and lower graph density.

Jane: It seems like this is a very practical tool because they compare it against methods like NOTEARS and LiNGAM and consistently show better results in terms of the F1 score and SHD metrics across almost every test case they ran.

Lu: They also provide a way to measure how well the model recovers the true edges by thresholding their estimated coefficient matrix A up to some threshold tau, which gives you a direct measure of edge-level accuracy on your network.

Meng: And for someone just looking at this for real biology, what does it mean that they can link their inferred networks back to ChIP-seq data?

Tom: They show the inferred gene regulatory network supports biological plausibility by demonstrating that hub genes like UBTF and GATA1 have edges supported by existing ChIP-seq data, which gives those connections real biological weight.

Jane: So, this isn't just a math exercise; it’s a framework that lets researchers go from raw Perturb-seq counts to biologically plausible gene regulatory networks with measurable accuracy.

Lu: The implication is that we can now tackle complex systems with latent confounding and non-Gaussian noise by explicitly modeling the latent causal structure through interventional data.

Tom: So, while there are still caveats about needing strong interventions for identification, this method gives us a solid path forward for analyzing these count data types.

Jane: We’ll need to see how they extend this framework to other complex count models in the future, like zero-inflated Poisson models or dropout-aware ones.

The paper's improvements: Tom: So, they're not just giving us a model; they're suggesting improvements to how we actually use this framework for real network discovery. What are these suggested enhancements?

Jane: They focus on making the method more robust when dealing with different types of count data, specifically suggesting extensions to handle zero-inflated or dropout-aware Poisson models.

Lu: That’s right, they acknowledge that their current setup is tailored for a standard Poisson process, but they explore how the underlying latent linear Gaussian structure can be adapted to fit more realistic biological noise patterns.

Meng: If you're shipping this to production, what does "adapting" actually mean in terms of deployment complexity? Does it just add another layer of model tuning or something more significant?

Tom: The authors suggest that the framework accommodates latent confounding without needing an explicit model for every single confounder by leveraging those interventional mean shifts, which is a way to handle uncertainty without needing a massive set of assumptions upfront.

Jane: So they’re trying to build this tool so it works even when we don't know every single hidden variable in the system.

Lu: Yes, they also discuss variance-shift interventions as a potential way to identify the structure using second order contrasts, though those typically require larger sample sizes for that identification to hold up.

Tom: It’s a trade-off, I guess. You get more identification power with variance shifts but you need a bigger budget for the intervention data itself.

Jane: And they are also pushing the idea of using permutation-based causal structure learning with unknown intervention targets, which is another way to test if the identified DAG structure is consistent across different ways of perturbing the system.

Meng: Permutation testing sounds like it adds computational cost because you have to run many iterations just to check for consistency, right?

Tom: It does, but they show that this approach helps in finding causal structures when we don't know the intervention targets beforehand.

Lu: And they’re linking their work with methods from other areas, like score-based causal representation learning, which suggests there's a broader direction for how we represent these complex systems mathematically.

Jane: It sounds like the paper isn't just a single solution but part of a larger toolbox for building more reliable and comprehensive causal models in biology.

Conclusion: Tom: So we've covered how they use latent linear models to disentangle gene expression counts from measurement error using targeted interventions to find causal structures in Perturb-seq data.

Jane: Exactly, and the main takeaway is that this approach gives us a concrete way to move beyond just finding correlations and actually map out the directed regulatory pathways.

Lu: The paper "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs" establishes population-level identifiability under mean-shift intervention designs, which means we can recover the true underlying causal mechanism A from observational and interventional data together.

Meng: So, if I'm deploying this to a production system, what is the real catch? What breaks when you try to run this on massive datasets with millions of genes?

Tom: The caveat is that identification relies on each gene being subject to at least one non-vanishing one-sparse mean shift intervention, which means we need careful experimental design upfront.

Jane: And the estimation procedure involves four steps, from cleaning up measurement effects with a generalized linear model to using an optimization method called ADMM to estimate the causal matrix A while respecting sparsity constraints.

Lu: They also show that performance improves systematically with stronger mean shifts and lower graph density, which suggests that data quality through better interventions really helps stabilize the estimation procedure.

Meng: That means for real-world use, the quality of your perturbation data is a major bottleneck for this method's success.

Tom: And on the biological side, they show that their inferred networks support plausibility by linking hub genes like UBTF and GATA1 to existing ChIP-seq data, which gives those connections real biological weight.

Jane: It feels like this paper provides a solid foundation for analyzing gene networks with count data where we usually get lost in the noise of measurement error and hidden variables.

Lu: They are also exploring how to extend this framework to handle more complex noise patterns, like zero-inflated or dropout-aware Poisson models, which opens up more biological scenarios.

Meng: So there’s a clear path for scaling this idea into a broader set of tools for different types of count data problems.

Tom: Exactly. And the paper "Causal Network Discovery from Interventional Count Data with Latent Linear DAGs" provides a robust framework that merges structural causal models with count data analysis.

Jane: It’s promising because it offers a practical method to uncover true gene regulatory networks without needing every single hidden variable in your system modeled explicitly.

Lu: The next step is to see how well this framework integrates with other representation learning techniques, which the authors touch on in their discussion of score-based causal representation learning.

Meng: I’m still waiting for the numbers on how much more intervention data you actually need to get those robust finite sample error bounds they talk about.

Tom: That's fair. But overall, this work is a strong demonstration that we can leverage controlled perturbations to gain causal insights in high-dimensional count data settings.

Yijiao Zhang, Hongzhe Li

University of Pennsylvania

stat.ME

Submitted: 2026-03-26

Updated: 2026-11-09

Comments: 35 pages, 4 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 88/100

The gist: The paper studies identification and estimation in latent linear structural causal models for interventional count data using a latent linear Gaussian directed acyclic graph (DAG) model with Poisson…

Key concepts

Latent Linear Gaussian DAG Model
This is a mathematical framework where the underlying true gene expression levels are hidden (latent) but follow a linear Gaussian structure. The model assumes these latent variables interact according to a directed acyclic graph (DAG), which represents the causal relationships between genes, allowing researchers to separate the true biology from measurement noise.
Poisson Measurement Error
This refers to the realistic scenario where gene expression counts are not perfectly observed; they are subject to Poisson noise. This noise is explicitly modeled in the equation, meaning the observed counts are related to the true latent expression levels through a Poisson distribution, accounting for common experimental inaccuracies.
Mean-Shift Intervention Design
This strategy involves intentionally perturbing (intervening on) genes in different experimental environments. By comparing how gene means change between these perturbed and control groups, researchers can isolate and identify the causal effects of specific genes on the system's latent state, which is crucial for identifying the DAG structure.
Alternating Direction Method of Multipliers (ADMM)
This is a specific mathematical optimization technique used in the estimation procedure. It helps solve complex problems involving estimating parameters like the causal coefficient matrix A while simultaneously enforcing constraints, such as sparsity and acyclicity, making it computationally feasible for high-dimensional count data analysis.

Terminology

Summary

The paper studies identification and estimation in latent linear structural causal models for interventional count data using a latent linear Gaussian directed acyclic graph (DAG) model with Poisson measurement error. p. 0 This work is motivated by the challenges of analyzing Perturb-seq data, which are count-valued and subject to substantial measurement error and latent state heterogeneity. p. 0

Model Formulation

The authors propose a latent linear Gaussian DAG model with Poisson measurement error that explicitly separates the latent causal structure from the observed counts. The observed gene expression count Xpmqj of gene j in environment m is modeled as a Poisson random variable conditional on the library size L, cell-level covariates C, and the latent true expression level Zpmqj: XpmqjL, C, Zpmqj Poisson(L exp(s pC + Zpmqj), j = 1,. p. 5.., p. The latent expression vector Zpmq is characterized by a linear Gaussian structural causal model (SCM) where Zpmq AZpmq ηpmq εpmq, with A being the causal coefficient matrix and εpmq being Gaussian noise. p. 6 The mean shift between interventional and control environments is captured by the contrast of the intercepts ηpmq - ηp0q. p. 6

Identification Strategy

Population-level identifiability of the latent causal DAG is established under a mean-shift intervention design where each gene is perturbed at least once. Identification relies on the mean difference between environment m and the control being ∆µpmq: µpmq - µp0q = αm Belm. p. 8 This implies that ∆µpmq equals the l-th column of B up to a scalar factor. p. 8 The causal mechanism A is identifiable by jointly leveraging observational and interventional data provided each node is subject to at least one non-vanishing one-sparse mean-shift intervention in the latent expression model. p. 9

Estimation Procedure

The estimation procedure involves four main steps: (i) estimating and removing measurement layer effects using a generalized linear model with log link and offset log L. (ii) mapping observed moments to latent means under the Poisson–lognormal model via the transformation of empirical moments. p. 8 (iii) recovering the columns of B with latent mean contrasts ∆µpmq via cross-environment mean shifts. p. 8 (iv) estimating A under sparsity and acyclicity constraints using an optimization problem solved by the alternating direction method of multipliers (ADMM). p. 3

Theoretical Guarantees and Recovery

The authors derive non-asymptotic estimation error bounds for the causal coefficient matrix A under a beta-min type condition on the interventional strength and signal strength. Theorem 5.2 provides finite-sample error bounds for estimating the latent causal coefficient matrix A˚ under interventional Poisson count data. p. 3 Furthermore, Theorem 5. p. 23 establishes finite-sample recovery of the latent causal DAG via thresholding the estimated coefficient matrix. p. 3

Simulation and Application

Simulation studies demonstrate that performance improves systematically with stronger mean shifts, lower graph density, and larger interventional sample sizes. The proposed method consistently outperforms competing methods like NOTEARS, LiNGAM, and MRS-PoissonSEM across nearly all regimes in terms of both F1 score and SHD. Application to Perturb-seq data shows the proposed method achieves the highest precision across the entire range of thresholds. p. 0 The inferred gene regulatory network supports biological plausibility by showing hub genes like UBTF and GATA1 with edges supported by ChIP-seq data. p. 1

Discussion and Future Work

The framework accommodates latent confounding without explicitly modeling the confounding effects by exploiting interventional mean shifts. Extensions are possible to other count models like zero-inflated or dropout-aware Poisson models. p. 4 Variance-shift interventions could enable identification via second-order contrasts but typically require larger sample sizes. p. 8

Data Availability

The data that support the findings of this study are openly available at https://gwps.wi.mit.edu. The research is supported by NIH grants GM129781 and HG013841. The paper includes supplementary materials with additional simulation details and evaluations. The references include works on constraint-based methods, score-based methods, and causal discovery with latent confounding. p. 4

References

Bello K., B Aragam, and P K Ravikumar (2022) Dagma: Learning directed acyclic graphs with gradient-based methods. p. 4 In Proceedings of the 39th International Conference on Machine Learning (ICML), pp. p. 2 1777–1793 PMLR. Bhattacharya R. p. 4, T Nagarajan, D Malinsky, and I Shpitser (2021) Differentiable causal discovery under unmeasured confounding. p. 2 In International Conference on Artificial Intelligence and Statistics, pp. 2314–2322 PMLR. Cai T., W Liu, and X Luo (2011) Constrained l1 minimization for sparse precision matrix estimation. p. 0 Journal of the American Statistical Association 106 (494), 594–607. p. 5 Chevalley M., Y H Roohani, A Mehrjou, J Leskovec, and P Schwab (2025) A large-scale benchmark for network inference from single-cell perturbation data. p. 1 Communications Biology 8 (1), 412. p. 1 Chickering D M. p. 3 (2002) Optimal structure identification with greedy search. p. 3 Journal of machine learning research 3 (Nov), 507–554. p. 2 Choi J. and Y Ni (2023) Model-based causal discovery for zero-inflated count data. p. 4 Journal of Machine Learning Research 24 (200), 1–32. p. 2 Colombo D., M H Maathuis, M Kalisch, and T S Richardson (2012) Learning highdimensional directed acyclic graphs with latent and selection variables. p. 1 The Annals of Statistics 294–321. Dai H., P Spirtes, and K Zhang (2022) Independence testing-based approach to causal discovery under measurement error and linear non-gaussian models. p. 4 Advances in Neural Information Processing Systems 35, 27524–27536. p. 1 Ferreira R., K Ohneda, M Yamamoto, and S Philipsen (2005) Gata1 function, a paradigm for transcription factors in hematopoiesis. p. 3 Molecular and Cellular Biology 25 (4), 1215–1227. p. 1 Heinze-Deml C. p. 3, M H Maathuis, and N Meinshausen (2018) Causal structure learning. p. 1 Annual Review of Statistics and Its Application 5 (1), 371–391. p. 3 Li J., B B Chu, I F Scheller, J Gagneur, and M H Maathuis (2025) Root cause discovery via permutations and cholesky decomposition. p. 3 Journal of the Royal Statistical Society Series B: Statistical Methodology qkaf066. p. 5 Liu E., J Zhang, and C Uhler (2025) Learning genetic perturbation effects with variational causal inference. p. 1 bioRxiv, 2025–06. p. 3 Pal S., J O’quinn, K Aryan, H Pua, J P Long, and A Asiaee (2025) Dag decoration: Continuous optimization for structure learning under hidden confounding. p. 3 arXiv preprint arXiv:2510. p. 002117. Park G. and S Park (2019) High-dimensional poisson structural equation model learning via l1-regularized regression. p. 4 Journal of Machine Learning Research 20 (95), 1–41. p. 2 Park G. and G Raskutti (2015) Learning large-scale poisson dag models based on overdispersion scoring. p. 4 Advances in neural information processing systems 28. p. 1 Pearl J. (2009) Causality: Models, Reasoning, and Inference (2 ed. p. 1). Cambridge: Cambridge University Press. p. 0 Peters J. and P B¨uhlmann (2014) Identifiability of gaussian structural equation models with equal error variances. p. 3 Biometrika 101 (1), 219–228. Replogle J M. p. 1, R A Saunders, A N Pogson, J A Hussmann, A Lenail, A Guna, L Mascibroda, E J Wagner, K Adelman, G Lithwick-Yanai et al. (2022) Mapping information-rich genotype-phenotype landscapes with genome-scale perturb-seq. p. 1 Cell 185 (14), 2559–2575. p. 1 Saeed B., A Belyaeva, Y Wang, and C Uhler (2020) Anchored causal inference in the presence of measurement error. p. 1 In Conference on uncertainty in artificial intelligence, pp. 619–628 PMLR. Sanij E., J Diesch, A Lesmana, G Poortinga, N Hein, G Lidgerwood, D P Cameron, J Ellul, G Goodall, L H Wong et al. (2015) A novel role for the pol i transcription factor ubtf in maintaining genome stability through the regulation of highly transcribed pol ii genes. Genome Research 25 (2), 201–212. Sarkar A. and M Stephens (2021) Separating measurement and expression models clarifies confusion in single-cell rna sequencing analysis. p. 5 Nature genetics 53 (6), 770–777. Schölkopf B., F Locatello, S Bauer, N R Ke, N Kalchbrenner, A Goyal, and Y Bengio (2021) Toward causal representation learning. p. 7 Proceedings of the IEEE 109 (5), 612–634. Shen X., P B¨uhlmann, and A Taeb (2025) Causality-oriented robustness: exploiting general noise interventions in linear structural causal models. p. 0 Journal of the American Statistical Association (just-accepted), 1–20. p. 5 Shimizu S., P O Hoyer, A Hyv¨arinen, and A Kerminen (2006) A linear non-gaussian acyclic model for causal discovery. p. 0 Journal of Machine Learning Research 7, 2003–2030. p. 2 Spirtes P., C N Glymour, and R Scheines (2000) Causation, prediction, and search. p. 3 MIT press. Squires C. and C Uhler (2023) Causal structure learning: A combinatorial perspective. p. 3 Foundations of Computational Mathematics 23 (5), 1781–1815. Squires C., Y Wang, and C Uhler (2020) Permutation-based causal structure learning with unknown intervention targets. p. 4 In Conference on Uncertainty in Artificial Intelligence, pp. 1039–1048 PMLR. Steine L A. (2022) Helping gata1 make complex decisions. p. 4 Blood 139 (24), 3457–3459. Uhler C., G Raskutti, P B¨uhlmann, and B Yu (2013) Geometry of the faithfulness assumption in causal inference. p. 3 The Annals of Statistics 436–463. van de Geer S. p. 3 and P B¨uhlmann (2013) l0-penalized maximum likelihood for sparse directed acyclic graphs. p. 1 The Annals of Statistics 41 (2), 536–567. Varici B., E Acart¨urk, K Shanmugam, A Kumar, and A Tajer (2025) Score-based causal representation learning: Linear and general transformations. p. 3 Journal of Machine Learning Research 26 (112), 1–90. p. 2 Wang Y., L Solus, K D Yang, and C Uhler (2017) Permutation-based causal inference algorithms with interventions. p. 2 In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, Red Hook, NY, USA, pp. 5824–5833. Wang Y. S. and M Drton (2023) Causal discovery with unobserved confounding and non-gaussian data. p. 4 Journal of Machine Learning Research 24 (271), 1–61. p. 2 Xiao F., J Tang, H Fang, and R Xi (2022) Estimating graphical models for count data with applications to single-cell gene network. p. 1 Advances in Neural Information Processing Systems 35, 29038–29050. p. 1 Xue A., J Rao, S Sankararaman, and H Pimentel (2025) dotears: Scalable and consistent directed acyclic graph estimation using observational and interventional data. p. 0 iScience 28 (2). Yuan Y., X Shen, W Pan, and Z Wang (2019) Constrained likelihood for reconstructing a directed acyclic gaussian graph. p. 0 Biometrika 106 (1), 109–125. Zhang K., M Gong, J D Ramsey, K Batmanghelich, P Spirtes, and C Glymour (2018) Causal discovery with linear non-gaussian models under measurement error: Structural identifiability results. p. 0 p. 0 In UAI, pp. 1063–1072. Zheng X., B Aragam, P Ravikumar, and E P Xing (2018a) Dags with no tears: continuous optimization for structure learning. p. 3 NIPS’18, Red Hook, NY, USA, pp. 9492–9503. Zheng X., B Aragam, P K Ravikumar, and E P Xing (2018b) Dags with no tears: Continuous optimization for structure learning. p. 3 In Advances in Neural Information Processing Systems (NeurIPS), Volume 31, pp. p. 1 9472–9483. Zhang K., M Gong, J D Ramsey, K Batmanghelich, P Spirtes, and C Glymour (2018) Causal discovery with linear non-gaussian models under measurement error: Structural identifiability results. In UAI, pp. 1063–1072.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Latent linear structural causal models for interventional count data can be used to identify latent causal structures from Perturb-seq data by explicitly separating latent expression levels from the observed Poisson measurement process. This allows for identification of a directed acyclic graph even in the presence of latent confounding and non-Gaussian noise.

  2. The proposed estimation procedure integrates sparse inverse matrix estimation with explicit DAG constraints, solved efficiently using the alternating direction method of multipliers (ADMM). This provides a computationally feasible way to estimate the latent causal coefficient matrix A for high-dimensional count data.

  3. The system can perform population-level identification of the latent causal DAG by leveraging cross-environment mean changes in latent means, provided each node is subject to at least one non-vanishing one-sparse mean shift intervention. This enables recovery of the true underlying causal mechanism A from observational and interventional data simultaneously.

  4. The improved AI system can quantify edge level accuracy using metrics like precision, recall, and F1 score for graph recovery based on thresholding the estimated coefficient matrix Ap to Ap tau. This provides a direct measure of how well the model recovers true regulatory edges in perturbed gene networks.

  5. The system can assess biological plausibility by comparing inferred regulatory networks against physical evidence from ChIP-seq data, highlighting edges that are supported by both perturbation data and existing transcription factor interaction networks.

Sources

Related papers