IQCELL: A platform for predicting the effect of gene perturbations on developmental trajectories using single-cell RNA-seq data

Abstract

The increasing availability of single-cell RNA-sequencing (scRNA-seq) data from various developmental systems provides the opportunity to infer gene regulatory networks (GRNs) directly from data. Herein we describe IQCELL, a platform to infer, simulate, and study executable logical GRNs directly from scRNA-seq data. Such executable GRNs allow simulation of fundamental hypotheses governing developmental programs and help accelerate the design of strategies to control stem cell fate. We first describe the architecture of IQCELL. Next, we apply IQCELL to scRNA-seq datasets from early mouse T-cell and red blood cell development, and show that the platform can infer overall over 74% of causal gene interactions previously reported from decades of research. We will also show that dynamic simulations of the generated GRN qualitatively recapitulate the effects of known gene perturbations. Finally, we implement an IQCELL gene selection pipeline that allows us to identify candidate genes, without prior knowledge. We demonstrate that GRN simulations based on the inferred set yield results similar to the original curated lists. In summary, the IQCELL platform offers a versatile tool to infer, simulate, and study executable GRNs in dynamic biological systems.

Virtual cells in a virtual microenvironment recapitulate early development-like patterns in human pluripotent stem cell colonies

“…Second, as GARMEN simulates the activity of thousands of interacting GRNs, the number of nodes should be kept low. This, however, is not a barrier, as we showed ( Heydari et al., 2022 ) that even for multistep differentiation typically a limited set of genes can play a significant role in differentiation. Third, our RD model is isotropic, yielding centro-symmetric solutions, which means that spontaneous symmetry breaking in 3D is likely to be challenging.…”

“…This could potentially be fixed by allowing boundary conditions to self-organize. Last, GRN creation is currently manual and can be automated via GRN inference tools, e.g., IQCell ( Heydari et al., 2022 ) or GENIE3 ( Huynh-Thu et al., 2010 ). We recommend rigorous pruning of automated networks by considering only those genes that significantly contribute to differentiation to ensure tractability over multiple scales.…”

spliceJAC : transition genes and state‐specific gene regulation from single‐cell transcriptome data

“…Reconstructing interactions between genes from single‐cell transcriptomic data, and scRNA‐seq data specifically, is an important problem that has been tackled with various techniques (Pratapa et al , 2020; Nguyen et al , 2021). Learning causal relations between genes can unlock novel biological insights and even predict the effect of molecular perturbations such as mutations or therapeutic interventions (Heydari et al , 2022). Several available GRN inference methods, such as SCODE (Matsumoto et al , 2017), SCNS (Woodhouse et al , 2018), and GRISLI (Aubin‐Frankowski & Vert, 2020), rely on pseudotime ordering.…”

Computational approaches for direct cell reprogramming: from the bulk omics era to the single cell era

“…As there are only finitely many possible states, one can find steady states or cycles which reflect stable cell states, and the downstream effect of perturbing TFs on these states can then be modeled. These types of approaches have successfully contributed to the modeling of a number of reprogramming and differentiation processes like neuronal differentiation [ 22 ], reprogramming to pluripotency [ 21 ] and T-cell and red blood cell development [ 23 ].…”

“…A variety of trajectory inference methods [ 41 , 42 ] have been developed to infer a pseudotime, constructing an order of the cells along these dynamic processes. Recently, IQCELL [ 23 ] exploits this application of single cell data to bypass the scalability issue of Boolean Networks, estimating the regulatory relationships in a developmental system in an unbiased way. Here, they use mutual information on the scRNA-seq data to establish the interaction network, and a gene hierarchy is built using pseudotime to infer the order of gene regulation ( Figure 2 ).…”