Pankaj Mehta

The paper integrates ecological interactions into population genetics using dynamical mean-field theory to predict mutation outcomes in complex communities.

The paper integrates genotype-to-phenotype maps into ecological models to develop an ecology of metagenomes, revealing gene-level dynamics and metagenomic hitchhiking.

The paper develops a framework to simulate metabolic evolution, identifies evolutionary collective modes as key drivers of simplicity in metabolic network dynamics, and validates these modes using E. coli data.

This paper applies attention-based models to predict phenotype from genotype, showing improved performance in epistatic regimes and enabling transfer learning across environments.

This paper shows simple linear models can achieve state-of-the-art results on single-cell RNA sequencing benchmarks, outperforming complex foundation models in out-of-distribution tasks.

The paper establishes a human iPSC-derived co-culture model of pulmonary fibrosis that reveals bidirectional fibrogenic epithelial-mesenchymal crosstalk and identifies potential druggable pathways.

The paper introduces CLASSIC, a platform that maps sequence-to-function relationships in synthetic biology by combining long- and short-read sequencing to analyze gene circuits.

The paper develops a general theory of ecological invasions to predict outcomes and shows that parent-mutant coexistence is common due to ecological feedbacks.

The paper introduces a model to identify decision-making classes in cell fate transitions using single-cell RNA-seq data, showing examples in lung and hematopoietic differentiation.

The paper introduces a theoretical framework to calculate ecosystem responses to environmental perturbations using geometric transformations in consumer-resource models.

This paper introduces a differentiable Gillespie algorithm for simulating chemical kinetics, parameter estimation, and designing synthetic biological circuits, demonstrating its effectiveness with E. coli promoter models.

This paper engineers synthetic phosphorylation signaling networks in human cells to enable dynamic sensing and suppression of activated T cells.

The paper normalizes five signal relay network models of Dictyostelium discoideum to assess their ability to reproduce observed population and single-cell dynamics, finding that single-cell excitable dynamics and fold-change detection are essential for robust coordination.

This paper introduces a novel hierarchical density clustering algorithm, HAL-x, which efficiently generates multiple cluster labels for single-cell data and achieves high accuracy in classifying clinical statuses.

The paper develops a framework linking protein-protein interaction strength to information transmission and dynamical response in signaling networks, finding a speed-information trade-off.