140x140
140x140

Multiscale Modeling of Biological Complexities

Bridging Cellular Biochemistry to Cell Population Dynamics and Tissue-Scale Heterogeneity

ParCell Team:

Dipak Barua
Assistant Professor
Dept. of Chemical and Biochemical Engineering
210 Bertelsmeyer Hall
baruad@mst.edu

Mohammad Aminul Islam
Ph.D. Student
Dept. of Chemical and Biochemical Engineering
215 Bertelsmeyer Hall
Phone: 573-341-7732
mixvc@mst.edu

Collaborators:

Sajal K. Das
Professor
Dept. of Computer Science
325B Computer Science Building
sdas@mst.edu

Sutapa Barua
Assistant Professor
Dept. of Chemical and Biochemical Engineering
210G Bertelsmeyer Hall
baruas@mst.edu

William S. Hlavacek
Scientist
Theoretical Division
Los Alamos National Laboratory
wish@lanl.gov

Project 1: ParCell: A Parallel Framework for Multiscale Modeling (NSF-CBET#1609642)

This National Science Foundation-supported (CBET#1609642) multidisciplinary project is being jointly executed by the research teams of Dr. D. Barua (Chemical and Biochemical Engineering) and Dr. Sajal K. Das (Dept. of Computer Science). Under this project, a multiscale modeling framework, called ParCell, is being developed for cell population modeling. The ParCell framework is aimed to provide a unique capability to create models that will link subcellular molecular mechanisms of signaling and gene transcription to cellular fate decisions and population dynamics. The goal is to enable systematic development of computationally-efficient and scalable multicellular models that will bridge molecular and cellular processes occurring at vastly different time and spatial scales.

80x80

The ParCell framework is founded upon a novel numerical algorithm based on the message passing interface (MPI) parallelism. In an innovative approach, the algorithm systematically transforms a single-cell biochemical network model into a cell population model. It launches parallel simulations on a single-cell stochastic model (on signaling or gene transcription) and treats each parallel simulation process as an individual cell in a population. Cellular communications are mediated via MPI calls in a server-client fashion. Cell death is simulated by terminating a parallel thread, and cell division is simulated by spawning new parallel processes (daughter cells) from an existing process (mother cell). The current version of the ParCell framework supports non-spatial multicellular models. The future versions of the framework will support spatiotemporal models with locally evolving growth environment (tissue conditions).

Project 2: Single-Cell Nanoparticle Uptake

The goal of this project is to quantitatively investigate the control mechanism of nanoparticle uptake at the single-cell level. This project is inspired by drug-delivery and tissue transport point of views. A biological tissue can be considered as a heterogeneous reaction-diffusion system. In such a system, tissue dispersion and penetration of drug nanocarriers depend on many factors. In addition to the transport barriers, the rate of particle uptake plays an important role in determining how far particles can penetrate from the vascular capillary into the tissuue interstitum. Here we use single-cell techniques, such as confocal microscopy and flow cytometer measurements, to quantify the parameters and variables that govern the rate of cellular uptake of drug nanoparticles. Traditionally, such studies focus on how physical attributes of a particle, such as size, shape, or surface propoerties, determine the uptake rate by cells. In this project, our goal is to understand the variables or parameters associated with specific cancer cells and the extracellular environment (transport barriers of the medium). Using a mechanistic reaction-diffusion model, we have quantitatively characterize the uptake characteristics of MDA-MB 231 breast cancer cells under reaction and diffusion controlled systems. Future focus of this work will be directed to distinct endocytic pathways and their contributions in nanoparticle uptake under reaction and diffusion controlled regimes. Another direction will be to quantify cellular heterogeneity in the uptake behavior based on the distribution of the physical attributes of the same or distinct cell types. Quantitative elucidation of the parameters and variables will allow the development of predictive reaction-diffusion models for interrogating tissue transport, dispersion and penetration behavior of nanoparticles.

Project 3: 80x80 Spatiotemporal Modeling of Multivalent Protein-Protein Interaction

We have developed a preliminary agent-based approach for modeling of reaction-diffusion systems considering the site-specific details of multivalent species (molecules, complexes, and ligand-conjugated nanoparticles). The agent-based approach implements an efficient multiscale Brownian Dynamics (BD) simulation algorithm. A time-adaptive feature of the BD algorithm enables highly efficient computation while capturing site-specific binding, dissociaion, and transformation of multivalent species. A snapshot of our preliminary simulations using the BD algorithm is shown. The figure shows assembly of multivalent ligand (green) and receptor (red) molecules in a two-dimensional domain (cell membrane). Models created using the algorithm can be used to interrogate steric effects and diffusion restrictions in the macromolecular assembly of signaling protein molecules and multivalent nanoparticles.

Project 4: 80x80 Multiscale Spatiotemporal Modeling of Nanoparticle Transport in Biological Tissues

The Brownian Dynamics algorithm (Project 2) is integrated with the Method of Regularized Stokeslets (MRS) to simulate transport behavior of nanoparticles and macromolecules in a heterogeneous porous medium under both advection and diffusion. Using the algorithm, we have performed simulation and analysis of nanoparticle size effects on their efficacy of tissue distribution and penetration. We are currently extending this integrated algorithm to account for the shape effects of nanoparticles. To create this capability, spatial graphs are being used to describe the geometry and surface attributes of a nanoparticle. Our future goal is to integrate this Brownian Dynamics algorithm with the MPI parallel framework (Project 1) to enable highly scalable and mechanistic modeling of nutrients, drug molecules and nanoparticle transport in biological tissues.

140x140

Funding Sources (Current and Past)

80x80 80x80 80x80 80x80
140x140