Dr. Chakra Chennubhotla

"Large-scale Analysis of the Evolution of Organellar Networks” Accurately characterizing the spatio-temporal evolution of the organellar networks in a large collection of time-lapse videos will provide significant insights into the host response to specific stimuli. Thus, we will design and implement a high-throughput, large-scale computational pipeline to quantify changes in organellar shape, number, and spatial distribution over large sequences of Z-stack microscope images and digital videos, to improve our understanding of cellular mechanisms as a response to environment.  Training plan: Weeks 1-3: Learn open source tools (Scikit-learn, CellProfiler) to extract cellular components from videos. Weeks 4-6: Build network interaction models (e.g., discrete-time finite Markov chains) for cellular component changes. Weeks 4-7: Summarize spatio-temporal evolution of organellar networks (e.g. frequency domain characterization). Weeks 8-9: Test, validate, and disseminate tools to bioimaging community. This project will be jointly supervised by Drs. Chennubhotla and Shannon Quinn at the University of Georgia and supported by the NSF grant: DBI 1458766.

Dr. Anne-Ruxandra Carvunis

Develop an algorithm to date the emergence of new genes in evolutionary history.

Dr. Nathan Clark

“Identifying species-specific, functional, genomic adaptations” The Clark lab is concerned with the evolution of species and their genomes as they adapt to new environments. We have recently developed new computational approaches to compare genome sequences to characterize functional changes between species. The student(s) will apply phylogenetic and genomic techniques to screen for particular genetic pathways that have experienced adaptive evolutionary change in a particular species. Such adaptive changes are identified using classical sequenced-based signatures as well as novel methods developed in our lab. Students will then correlate these instances of adaptive evolution with the life histories and organismal traits of the species involved. This project will train students to use genomics to gain insight in to organismal biology and teach them fundamentals of molecular evolution, population genetics, genomics, and phylogenetics.

Dr. Rob Coalson

"Computational Studies of Molecular Transport through the Nuclear Pore Complex” The Nuclear Pore Complex (NPC) controls the flow of large molecules into and out of the nucleus via a passive diffusion mechanism, involving unstructured protein filaments, that is not fully understood. The Coalson group has developed coarse-grained models and Molecular Dynamics (MD) simulation codes based on the elements of this complex system. Students will gain familiarity with all facets of the model, learn to run the simulation codes using the LAMPPS simulation software, and gain an appreciation for the microscopic forces that control these dynamics. In these coarse-grained simulations, the purpose of the interaction forces utilized is to encode basic physico-chemical properties such as particle sizes, electrostatics and inter-particle binding energies. Through their simulations, students will develop a core understanding of the importance of geometric effects and basic energetic interactions on the dynamics of macromolecular systems relevant to molecular biology.

Dr. Vaughn Cooper

Dynamics of adaptation in bacterial biofilmsThe Cooper laboratory studies how bacterial populations evolve from clones to form complex communities while growing on surfaces, and uses their own sequencers (Illumina NextSeq, Oxford Nanopore MinION) to identify the mutations associated with these evolutionary transitions. Students will participate in all steps of this process, from production of sequencing libraries from DNA samples, to management of raw output, to the inference of the dynamics within experimentally evolved or naturally occurring bacterial populations. This research will develop broad bioinformatics skills and encourage students to make linkages between inferred mutations and their functional consequences, ultimately contributing to publications from the laboratory.

Dr. Lance Davidson

Modeling 3D multicellular arrays for investigating emergent properties of tissue self-assembly” Tissue architecture is established through the action of multiple cells in close association yet little is known about the integration of cell signaling and mechanics. We seek to develop data-driven models of 2D and 3D multicellular arrays to investigate sensing, transmission, and actuation of mechanical events. We use existing frameworks, e.g. Chaste (Oxford University) or Virtual Leaf (CWI-Amsterdam) together with image data collected from 3D developing tissues to develop models integrating mechanics and biochemical signaling networks. Emphasis will be made on model development, isolation of critical experimental phenomena, and predictive modeling of observable phenomena.  Training plan: Weeks 1-4: Learn modeling framework by implementing known models drawn from literature. Weeks 3-6: Use word models to develop mathematical frameworks and computational implementations. Weeks 4-7: Implement code. Weeks 6-9: Investigate parameter spaces for emergent processes.

Dr. Jacob D. Durrant

Lab projects focus on answering these questions: How can we identify new protein pockets that we can target with drug-like molecules? How do protein motions affect drug binding? Can we predict those motions in a computer? What new techniques, perhaps related to machine learning, can we use to better predict if a drug-like molecule binds to a protein? If it does bind, with what strength of binding?

Dr. G. Bard Ermentrout

(1) Modeling of the spatio-temporal dynamics that underlie various pathologies in the nervous system such as epilepsy and visual hallucinations. 
(2) Models for odor localization in complicated environments.

Dr. James Faeder

“Structure-based models of signaling pathways using the BioNetGen software” Student(s) will use the rule-based modeling language BioNetGen to construct detailed models of key components of signaling pathways and test the effects of both quantitative and structural changes to these models on the outcome of signaling. Students will read textbook chapters on signal transduction pathways, which will be pointed out to them, and select a signaling pathway to investigate further. After reading one or two relevant review articles, students will refine the focus to a few key signaling proteins and their interactions, find related primary literature, and use these to determine the functional elements and interactions of the proteins to be modeled. They will encode these in a BioNetGen model and be guided in the identification or estimation of the relevant parameters. The students will then investigate the behavior of the model as key parameters and structural elements are modified, and also compare the results of the model to the experimental literature.

Dr. David Koes

Our group develops novel computational algorithms and interactive online systems that support rapid and inexpensive drug discovery.  In particular, I am interested in developing high-performance algorithms (possibly using GPUs) for screening and scoring drug-protein interactions as well as slower, higher-fidelity algorithms that include a detailed model of protein-ligand dynamics. Undergraduate researchers will further develop their machine learning, programming, and statistical analysis skills while learning about protein structure and function.

Dr. Dennis Kostka

Transcriptional characterization of cellular heterogeneity in the heart” Single cell sequencing approaches have the potential to transform biological knowledge. We generate and analyze transcriptional profiles of primary heart cells to better understand the molecular underpinnings of cellular heterogeneity. TECBio students will learn how to process single cell RNA-seq data, including quality control, adapter trimming, and alignment, all with open source software. Next, students will learn to normalize data, generate expression abundance estimates, and characterize cells based on highly expressed and/or highly variable genes allowing them to focus on biologically important genes and pathways and to group cells into coherent clusters. Students will gain “computer literacy” (Unix-like operating systems, command line / shell, programming), and skills to perform functional genomics analyses with single cell data.

Dr. Bing Liu

Ferroptosis is the most recently recognized form of regulated cell death, which may be crucial to the development of therapeutics for cancer, radiation disease and asthma. This project aims to use cutting-edge systems biology approaches to understand the signaling/metabolic network that underlies ferroptosis, in collaboration with the leading experimentalists in the field.

Dr. Jeffry Madura

“G quadruplex RNA-protein Interactions” The Madura group has been applying molecular modeling methods to investigate the structure of RNA G-quadruplex and their interaction with peptides. Enhanced sampling molecular dynamics (MD) calculations have been used to study the folding/unfolding of an RNA G-quadruplex while conventional MD have been used to investigate peptide interactions with the G-quadruplex. Based on their interests, students will perform calculations to (i) further investigate the stability of the G-quadruplex under specific solution conditions, and/or (ii) study various peptide:G-quadruplex complexes. These projects will provide students with an immersive computational biophysics research project intimately tied to the life sciences. Students will also work closely with an experimental group by providing molecular level models to explain experimental results. They will also develop independent literature research skills required to check and understand the simulation results they obtain.

Dr. Natasa Miskov-Zivanov

Automation of intra- and inter-cellular model design and explanation using engineering techniques such as digital, analog or mixed-signal circuit design, signal processing, automated control. The domain of the project is interplay between immune system and diseases.

Dr. Robert Parker

The Parker lab works in the area of systems medicine.  REU students would work with Prof. Parker and a graduate student or post-doctoral mentor to construct models of disease processes and treatment response in the areas of solid tumor cancer chemotherapy, glucose control in critical care, inflammation/sepsis, thrombocytopenia, or cystic fibrosis.  Many projects are interdisciplinary with a clinical co-advisor.  The objective of building systems-level models is to incorporate them into decision support systems that help clinicians synthesize the data available to them and to make more informed treatment decisions for individual patients.

Dr. John Shaffer

(1) Genetic association of rare variants in craniofacial enhancer regions with orofacial cleft birth defects
(2) Genome-wide association study of dental fluorosis in children
(3) Replication of genetic associations with dental caries
(4) Clustering the primary dentition to develop biologically-informative phenotypes for genetic analysis

Dr. Shikhar Uttam

Develop an algorithm to identify sparse domains that optimally characterize intra-tumor heterogeneity as a tool for predicting/assessing cancer progression.

Dr. Jianhua Xing

“Modeling Cellular Phenotype Transitions” The Xing lab is using integrated quantitative measurements and computational/mathematical modeling to study how a cell undergoes cell phenotype transformation, specifically in regard to epithelial-to-mesenchymal transitions, a process important for proper embryo development and other several processes. For this project, a student will read introductory review papers on cell-to-cell heterogeneity and the importance of single cell studies. Then students will work with a postdoctoral researcher or graduate student to gain experience on acquiring time-lapse images that they will then analyze. With the data the students will learn to use available softwares (e.g. Cellprofiler) to quantify and characterize the data. Meanwhile, the students will be introduced to mathematical modeling of cell regulatory networks and how to use experimental data to constrain model parameters.

Dr. Leming Zhou

Agent-based modeling of cardiovascular disease, liver fibrosis and Osteoporosis; Statistical modeling of cognitive process for people with aphasia; mobile health apps for people with disabilities such as communication disorders and dexterity; a personal genomic data management and analysis system.