Gene Regulatory Network Analysis

This paper develops a deep learning model to estimate gene regulatory networks and RNA kinetics from single-cell RNA-seq data, showing improved recovery of known regulatory relationships and predictive power.

The paper introduces the CASwitch, a synthetic gene circuit that reduces leakiness while maintaining high inducible gene expression using mathematical modeling and experimental validation.

The paper develops an analytical model for noise propagation in gene expression cascades, showing how cascade structure, length, and other factors influence intrinsic and global noise.

This paper analyzes the frequency response and sensitivity of biochemical cascades with varying phosphorylation layers and states.

The paper investigates size regulation mechanisms in slow-growing E. coli cells, finding that different models explain size control under varying media conditions.

The paper investigates the trade-off between upstream and downstream noise in signaling networks, showing that longer timescales buffer upstream noise but accumulate downstream noise, confirmed in biological systems.

The paper introduces scMeMo, a mechanistic modeling framework that infers unobserved variables in single cells using diverse measurements, demonstrating its ability to predict protein levels and cell cycle behavior from limited data.

This paper introduces RENGE, a method for inferring gene regulatory networks from time-series scCRISPR data, which outperforms existing methods and produces GRNs consistent with known databases.

The paper estimates single-cell protein decay rates of degron GFPs in yeast and finds substantial cell-to-cell variability in decay rates, suggesting the need for complex decay dynamics models in gene expression noise estimation.

The paper investigates transcriptional adaptation in mammalian cells using computational analysis, mathematical modeling, and machine learning, identifying widespread prevalence and single-cell manifestations of compensatory gene networks.

LINGER improves GRN inference from single-cell multiome data by integrating atlas-scale bulk data and TF motif knowledge, achieving 2-3 fold higher accuracy.

This study benchmarks expression forecasting methods using 11 perturbation datasets, finding that simple baselines often outperform complex models using mean squared error.

The paper identifies a conserved minimum transcription level of one message per cell cycle for essential genes across E. coli, yeast, and human, linking it to noise robustness.

This study identifies unifying design principles of endocrine gland mass and regulatory circuits across human hormone systems using mathematical analysis.

The paper identifies synthetic gene network-induced phenotypes by analyzing gene expression in E. coli and validates them through antibiotic resistance experiments.

The paper identifies phenotype-relevant genes by analyzing dose response data without using models, finding genes with spikes in transcriptional sensitivity that likely regulate phenotypic transitions.

The paper uses Bayesian methods to analyze kinetic transcription parameters in yeast under stress, revealing fast bursting dynamics within slower cycles.

The paper demonstrates a modular bacterial system that functions as a full subtractor or adder by mixing engineered cells, modeled as artificial neural network components.

The paper introduces a dataset of 210+ Boolean network models with validation to ensure logical consistency and integrity.

This study investigates how growth feedback affects gene circuit dynamics, identifying failure mechanisms and resilient topologies.

The paper introduces a method to analyze the role of specific genes in reachability properties of GRNs using the Arabidopsis thaliana flower development GRN, finding 79 inducible attractor transitions with external control inputs.

MCNET is a deep learning algorithm that integrates multi-omics data to infer gene regulatory networks from scRNA-seq, outperforming existing methods in GRN inference and visualization.

This study investigates whether bacterial cells retain memory of past environmental conditions across generations, finding non-Markovian memory in growth rates but not cell size.

This study extends the initiator-titration model to explain how bacteria control replication initiation through protein copy-number sensing, addressing two long-standing questions about DnaA overproduction and its active/inactive forms.

This study develops an algorithm to estimate gross transcription rates from RNA level fluctuation data and examines how sampling time intervals affect these estimates.

The paper analyzes optimal feedback strategies for reducing timing noise in biochemical events using stochastic models and shows that mixed feedback minimizes noise when degradation is present.

The paper introduces NS-DIMCORN, a generative model using ODEs to reconstruct gene regulatory networks from scRNA-seq data, outperforming existing methods in accuracy.

The paper introduces a computational workflow for predicting transcription factors to achieve cellular reprogramming, providing mechanistic insights into gene regulatory networks.

This study benchmarks gene regulatory network inference methods on simulated and experimental single-cell E. coli data, finding moderate accuracy and highlighting method-specific performance differences.

This study evaluates regulatory network-based machine learning for gene expression prediction in maize using Arabidopsis thaliana networks, finding that simple models outperform complex ones within species but complex models are better for cross-species predictions.