Purushottam D. Dixit

This paper develops a metagenome-scale metabolic model to predict FMT outcomes by assessing donor-recipient metabolic compatibility in IBS patients.

The paper shows that ligand endocytosis enhances directional information in cell-cell attraction by reshaping extracellular gradients.

The paper explains how weak-affinity ligands can drive broader signaling activity than strong-affinity ligands by showing that receptor architecture decouples signal strength and range through kinetic proofreading.

This paper introduces a reaction-diffusion model called LAGS to show that eukaryotic receptors can perform directional sensing through lateral diffusion, basal activity, and active degradation, without relying on downstream signaling pathways.

The paper infers microbial niche dimensionality from abundance data and shows it correlates with environmental complexity, competition, and biodiversity in human gut microbiomes.

The paper quantifies age-related decline in transcriptional information flow using information theory and single-cell RNA-seq data, finding reduced mutual information between transcription factors and target genes.

The paper introduces a non-equilibrium model explaining how signaling receptors achieve ligand specificity through multi-site phosphorylation and degradation, showing that intermediate-affinity ligands maximize network activity.

This paper introduces a generative model based on the consumer/resource framework to design host-associated microbiomes that correspond to desired host phenotypes and vice versa.

This study uses computational modeling and preclinical data to identify optimal dosing strategies for panRAF and MEK inhibitors in melanoma, revealing distinct dose requirements based on NRAS vs BRAF mutations.

The paper quantifies the distribution of sensing abilities in cell populations using an information theoretic framework and shows most cells achieve better input resolution than the average.

SMbiot identifies low-dimensional latent variables that capture host-microbiome associations and predicts missing metadata or community compositions.

SiGMoiD is a generative model for binary data that infers constraints from data and handles large variable counts, enabling clustering and reduced dimensionality.

EMBED is a dimensionality reduction method for microbiome studies that identifies ecological normal modes to capture community and individual taxa dynamics across longitudinal data.

The paper develops and validates a computational model showing mammalian cells use receptor abundance to encode past stimuli for relative sensing in signaling networks.

The paper introduces MERIDIAN, a Maximum Entropy framework to infer signaling network heterogeneity from cell-to-cell variability in network species abundances, and applies it to EGF-induced Akt signaling.

The paper introduces DIVERS, a method to quantify spatiotemporal dynamics and noise in microbiota abundances using replicate sampling and spike-in sequencing, applied to the human gut microbiome.