Biology subjects

Sivak, D. A.

Publications and source records attributed to Sivak, D. A..

4 recordsLinked to original sources

bioRxiv2023

D-SPIN constructs gene regulatory network models from multiplexed scRNA-seq data revealing organizing principles of cellular perturbation response

Gene regulatory networks within cells modulate the expression of the genome in response to signals and changing environmental conditions. Reconstructions of gene regulatory networks can reveal the information processing and control principles used by cells to maintain homeostasis and execute cell-state transitions. Here, we introduce a computational framework, D-SPIN, that generates quantitative models of gene regulatory networks from single-cell mRNA-seq datasets collected across thousands of distinct perturbation conditions. D-SPIN constructs probabilistic models of regulatory interactions between genes or gene-expression programs to fit the cell state distributions under different perturbations. Using large Perturb-seq and drug-response datasets, we demonstrate that D-SPIN models reveal key regulators of cell fate decisions and the coordination of distant cellular pathways in response to gene knockdown perturbations. D-SPIN also dissects gene-level drug response mechanisms in heterogeneous cell populations, elucidating how combinations of immunomodulatory drugs acting on distinct regulators induce novel cell states through additive recruitment of gene expression programs. The D-SPIN model further predicts cell state distributions under drug dosage combinations beyond the training data. D-SPIN provides a computational framework for constructing interpretable models of gene regulatory networks to reveal principles of cellular information processing and physiological control.

bioinformatics↗

bioRxiv2021

A gut commensal niche regulates stable association of a multispecies microbiota

The intestines of animals are typically colonized by a complex, relatively stable microbiota that influences health and fitness, but the underlying mechanisms of colonization remain poorly understood. As a typical animal, the fruit fly, Drosophila melanogaster, is associated with a consistent set of commensal bacterial species, yet the reason for this consistency is unknown. Here, we use gnotobiotic flies, microscopy, and microbial pulse-chase protocols to show that a commensal niche exists within the proventriculus region of the Drosophila foregut that selectively binds bacteria with exquisite strain-level specificity. Primary colonizers saturate the niche and exclude secondary colonizers of the same strain, but initial colonization by Lactobacillus physically remodels the niche to favor secondary colonization by Acetobacter. Our results provide a mechanistic framework for understanding the establishment and stability of an intestinal microbiome. One-Sentence SummaryA strain-specific set of bacteria inhabits a defined spatial region of the Drosophila gut that forms a commensal niche.

microbiology↗

bioRxiv2021

Stochastic microbiome assembly depends on context

Observational studies reveal substantial variability in microbiome composition across individuals. While some of this variability can be explained by external factors like environmental, dietary, and genetic differences between individuals, in this paper we show that for the model organism Drosophila melanogaster the process of microbiome assembly is inherently stochastic and contributes a baseline level of microbiome variability even among organisms that are identically reared, housed, and fed. In germ-free flies fed known combinations of bacterial species, we find that some species colonize more frequently than others even when fed at the same high concentration. We develop a new ecological technique that infers the presence of interactions between bacterial species based on their colonization odds in different contexts, requiring only presence/absence data from two-species experiments. We use a progressive sequence of probabilistic models, in which the colonization of each bacterial species is treated as an independent stochastic process, to reproduce the empirical distributions of colonization outcomes across experiments. We find that incorporating context-dependent interactions substantially improves the performance of the models. Stochastic, context-dependent microbiome assembly underlies clinical therapies like fecal microbiota transplantation and probiotic administration, and should inform the design of synthetic fecal transplants and dosing regimes. Significance StatementIndividuals are constantly exposed to microbial organisms that may or may not colonize their gut microbiome, and newborn individuals assemble their microbiomes through a number of these acquisition events. Since microbiome composition has been shown to influence host physiology, a mechanistic understanding of community assembly has potentially therapeutic applications. In this paper we study microbiome acquisition in a highly-controlled setting using germ-free fruit flies inoculated with specific bacterial species at known abundances. Our approach revealed that acquisition events are stochastic, and the colonization odds of different species in different contexts encode ecological information about interactions. These findings have consequences for microbiome-based therapies like fecal microbiota transplantation that attempt to modify a persons gut microbiome by deliberately introducing foreign microbes.

microbiology↗

bioRxiv2018

Model selection for biological crystallography

Structural biologists have fit increasingly complex model types to protein X-ray crystallographic data, motivated by higher-resolving crystals, greater computational power, and a growing appreciation for protein dynamics. Once fit, a more complex model will generally fit the experimental data better, but it also provides greater capacity to overfit to experimental noise. While refinement progress is normally monitored for a given model type with a fixed number of parameters, comparatively little attention has been paid to the selection among distinct model types where the number of parameters can vary. Using metrics derived in the statistical field of model comparison, we develop a framework for statistically rigorous inference of model complexity. From analysis of simulated data, we find that the resulting information criteria are less likely to prefer an erroneously complex model type and are less sensitive to noise, compared to the crystallographic cross-validation criterion Rfree. Moreover, these information criteria suggest caution in using complex model types and for inferring protein conformational heterogeneity from experimental scattering data.

biophysics↗