Ramon Grima

PRIME infers mechanistic cell states from multimodal single-cell data by embedding counts in a probability generating function space, enabling robust kinetic analysis and clustering.

This paper introduces a probability generating function-based method for parameter inference in stochastic biochemical reactions and shows it outperforms existing methods in efficiency, accuracy, and robustness.

This paper provides a unified Hilbert space framework for the spectral theory of stochastic gene expression and shows that deterministic models fail to capture relaxation rates under strong autoregulation.

DART uses deep learning to infer promoter states from live-cell imaging data, revealing a linear coupling between activation and inactivation rates.

The paper derives an exact steady-state solution for the joint distribution of nuclear and cytoplasmic mRNA levels and uses it to quantify bursty gene expression in human cells.

This study quantifies phyA dynamics in Arabidopsis under canopy shade conditions and identifies its role in promoting seedling adaptation to moderate shade.

This paper evaluates Negative Binomial models and parameter estimation in scRNA-seq data, finding that NB fits occur in diverse regimes and that burst parameters are reliably informative.

The paper estimates burst frequency and size parameters of gene expression in G1 and G2/M phases using mechanistic models on single-cell RNA-seq data from mouse cells.

The paper shows that power-law behavior in mRNA counts after induction can estimate the number of regulatory steps in gene expression.

Holimap approximates gene network dynamics using a simpler reaction system, showing computational advantages over conventional methods in predicting stochastic time-dependent dynamics.

This paper develops a graph neural network method to predict spatio-temporal stochastic gene expression efficiently, achieving faster computation and extrapolation capabilities.

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

This study provides an exact time-dependent solution for the CME of a gene circuit with autoregulation and bursty expression using spectral decomposition.

The paper develops stochastic models to show that licensing and competition for niche access together regulate stem cell homeostasis and prevent extinction.

This paper develops a theory to quantify and correct bias in transcriptional parameter inference from single-cell data by analyzing the effects of extrinsic noise.

The paper uses single-molecule long-read DNA methylation data to show that bulk-level metrics underestimate heterogeneity in the methylome, particularly in heterochromatic regions.

The paper shows that some fission yeast cell division genes exhibit sub-Poissonian mRNA noise, below the classical Poissonian limit, due to specific transcription and degradation mechanisms.

The paper derives analytical solutions for an extended telegraph model that incorporates cell cycle and growth effects on mRNA fluctuations, showing a linear relationship between mRNA and cell volume fluctuations.

The paper models gene expression and cell size dynamics to analyze concentration fluctuations and homeostasis in E. coli, fission yeast, and budding yeast.

This paper uses neural networks to approximate solutions of the Chemical Master Equation, enabling faster stochastic simulation and inference.

DelaySSAToolkit.jl provides a Julia package for simulating reaction systems with time delays using exact formulations of the delay stochastic simulation algorithm.

The paper introduces a synthetic model-based approach for uncertainty quantification in stochastic gene expression models, showing improved performance over existing methods.

The paper develops a stochastic model to analyze how time-dependent stimuli affect mRNA fluctuations in different sub-cellular locations and derives analytical distributions for mRNA numbers at various stages.

This paper introduces a fluctuation-based method to infer cell-state switching kinetics and topology from a single time point measurement, applicable when repeated measurements are not feasible.

This paper develops a stochastic model to explain non-exponential growth and bimodal cell size distributions in fission yeast, showing how growth phases and control strategies shape these distributions.

The paper develops a mean-field theory to model mRNA copy number distributions across its life cycle and shows it accurately predicts steady-state distributions across various conditions.

This paper shows that a cluster mean-field model accurately predicts statistical properties of large-scale DNA methylation patterns and can infer model parameters from data.

The paper compares two-state (telegraph-like) and mechanistic models of mammalian gene expression, showing they can produce similar transcript number distributions but differ in waiting time distributions.

The paper develops a theoretical framework to analyze mRNA and protein fluctuations in cell lineages and validates it using E. coli data, showing how power spectrum features reveal gene expression mechanisms.