Gene Regulatory Network Analysis

This paper identifies design principles for perfect adaptation in biological networks using nonlinear systems theory, showing that only two network architectures can achieve local adaptation across the state space.

This study shows how E. coli populations adjust sensory diversity based on environmental ligand prevalence by measuring kinase activity and using mathematical modeling.

SCRIPro infers gene regulatory networks from single-cell and spatial multi-omics data, outperforming existing methods in reconstructing cell type-specific GRNs.

This paper shows that bow-tie architecture emerges in evolving feedforward networks due to weak initial molecular interactions, not adaptive advantages.

The paper constructs a genome-scale transcriptional regulatory network of the human placenta using genomic and transcriptomic data, identifying key regulators of preterm birth.

AttentionPert predicts gene expression under multiplexed genetic perturbations and generalizes to unseen conditions, outperforming existing methods in accuracy and handling diverse perturbations.

The paper optimizes data integration for GRN inference in Arabidopsis thaliana by introducing DIOgene, which improves prediction accuracy and identifies gene-specific integration intensities.

The paper compares deterministic and stochastic models of cellular event timing, showing significant differences in predicted trigger times under certain conditions.

This paper classifies network topologies capable of generating infinitesimal homeostasis in multiple-input parameter networks and identifies three distinct mechanisms.

This study uses information theory to show that spatial gradient-sensing becomes more beneficial for larger, slower, and less persistent chemotactic agents, under physical noise constraints.

This study models how transcription factors regulate gene expression by altering transcription steps, revealing how promoter strength and TF mechanisms affect steady-state regulation, noise, and response time.

The paper identifies and validates novel regulators of vascular smooth muscle cell proliferation using network analysis and experimental validation, highlighting RUNX1 and the TIMP1-CD74 axis as potential therapeutic targets.

This study uses machine learning to predict cell division from ERK and Akt activity dynamics in single cells, finding ERK dynamics more predictive than Akt.

The paper introduces a statistical method to infer hidden hierarchical structures in biological networks using correlations of observable components.

This study develops a theoretical framework to evaluate phosphorylation-driven protein switches and demonstrates two novel switches with fast, reversible responses to kinase activity.

The paper introduces parallel implementations of the continuous time Boolean model simulator for large CPU clusters and GPU accelerators to handle increased model complexity.

The paper shows that Hill functions are the universal limit for sharpness of input-output responses in thermodynamic equilibrium models.

The paper develops a stochastic agent-based model to study how gene expression noise and population dynamics interact to enable cellular decision-making, using Escherichia coli data to demonstrate its application.

The paper designs synthetic inhibitory circuits to study multicellular self-organization and shows they can generate patterned structures and morphological oscillations.

This paper develops a mathematical framework to measure and tune tempo in developmental gene regulatory networks by analyzing perturbations that preserve dynamical orbits.

PHOENIX uses biologically informed neural ODEs to infer genome-wide regulatory dynamics, demonstrating scalability and interpretability in simulated and real data.

The paper designs synthetic genetic circuits in plants using bacterial transcription factors to sense and integrate phenylpropanoid metabolite levels, enabling Boolean logic operations for metabolic pathway control.

The paper introduces PyDyNo, a framework to identify dominant reaction paths in biochemical networks using Bayesian parameter optimization, revealing distinct signal execution modes in apoptosis and JAK2/STAT5 networks.

The paper introduces a general vesicle modeling tool integrated into the STEPS software for simulating vesicle dynamics in cell models with realistic mesh reconstructions.

GRouNdGAN simulates single-cell RNA-seq data using causal generative adversarial networks guided by gene regulatory networks, outperforming existing simulators and enabling in-silico perturbation experiments.

The paper develops an algorithm to identify buffering structures in chemical reaction networks by leveraging an inverse theorem of the law of localization, enabling efficient detection of modular regulatory units.

This paper introduces neural network approximations for solving steady-state distributions in multi-species transcriptional models, enabling efficient inference of biophysical parameters from transcriptomic data.

Velorama infers causal gene regulatory networks using RNA velocity data, revealing TF interaction speeds and their links to disease.

The paper introduces BiSDL, a modeling language for multicellular synthetic biology, enabling spatial and multi-level cellular dynamics representation and simulation.

The paper develops a machine learning framework to create organ-specific gene expression datasets and infer gene regulatory networks, achieving high accuracy in predicting transcription factors regulating organ-specific processes.