Dynamic Modeling of the Iron Deficiency Induced Genetic Interactions in Arabidopsis thaliana

Dr. Cranos Williams, PI; EnBiSys Researchers: Alexandr Koryachko, Selene Schmittling, Samiul Haque

Collaborating PIs: Dr. Terri Long, Dept. of Plant and Microbial Biology; Dr. Joel Ducoste, Dept. of Civil, Construction, and Environmental Engineering; and Dr. James Tuck, Dept. of Electrical and Computer Engineering

This National Science Foundation-funded INSPIRE project is an interdisciplinary effort spanning two colleges (College of Engineering and College of Agricultural and Life Sciences) and three departments (Electrical and Computer Engineering, Plant and Microbial Biology, and Civil, Construction, and Environmental Engineering). This research aims to further our understanding of the underlying molecular processes involved in the model plant, Arabidopsis thaliana’s, response to iron deprivation stress. Knowledge obtained in this study will impact our ability to genetically engineer crops with improved tolerance to low nutrient soil conditions and crops that exhibit increased iron content. Such strategies may provide avenues for fighting anemia and increasing land area suitable for farming.

The primary objectives of this project are as follows: (1) to develop a dynamic gene regulatory network model of the transcriptional response of Arabidopsis under iron deficient conditions, (2) to identify commonly occurring DNA sequences, which represent possible regulatory binding sites, in genetic regions responsible for gene regulation and, (3) to explore cell-specific regulation across an array of cell-types.

To accomplish these goals, our collaborators use experimental techniques such as microarrays, qRT-PCR, and RNA sequencing to quantify the behavior of gene expression over time and under various iron conditions.  These data capture the dynamics of these genes and can be abstracted as digital signals.  We then use computational algorithms derived from signal processing, machine learning, and system identification to identify genes that play a significant role in iron homeostasis, to infer relationships between genes, to define mathematical constructs capable of describing and/or predicting the dynamics of these gene under single gene and multigene perturbations.  

Contributions to Date: We used clustering, detection and estimation, and pattern recognition techniques to develop an algorithm for deciphering a network of regulatory connections from time course gene expression data. We applied our algorithm to whole genome iron deficiency transcriptome samples, and obtained a network that predicted new regulators/influencers of genes known to modulate iron deficiency response in A. thaliana. With our collaborators, we have validated the identified connections between the regulators and the targets through additional experimentation. We are using these results, in aggregate with additional experimental data to develop a descriptive and predictive dynamic model described by a set of ordinary differential equations.

Towards the goal of identifying the sequences in DNA where regulatory proteins bind, we are developing methods to improve computational strategies grouping genes based on gene expression profiles. These grouping strategies aim to identify genes that are regulated by the same protein or proteins. We will use the developed grouping strategy in combination with existing algorithms to search for common DNA sequences which represent the binding site for the regulating protein.

Work supported by the National Science Foundation: MCB-1247427

Publications produced from this project:

  • A. Koryachko, S. Haque, A. Matthiadis, D. Muhammad, J. J. Ducoste, J. Tuck, T. A. Long, and C. Williams. “Dynamic modelling of the iron deficiency modulated transcriptome response in Arabidopsis thaliana roots.” in silico Plants, 2019.
  • D. Muhammad, S. Schmittling, C. Williams, and T. A. Long, “More than meets the eye: Emergent properties of transcription factors networks in Arabidopsis,” Biochimica et Biophysica Acta (BBA) – Gene Regulatory Mechanisms, 2017.
  • A. Koryachko, A. Matthiadis, J. J. Ducoste, J. Tuck, T. A. Long, and C. Williams. Computational approaches to identify regulators of plant stress response using high-throughput gene expression data. Current Plant Biology, 3:20-29, 2015.
  • A. Koryachko, A. Matthiadis, D. Muhammad, J. Foret, S. M. Brady, J. J. Ducoste, J. Tuck, T. A. Long, and C. Williams. Clustering and differential alignment algorithm: Identification of early stage regulators in the Arabidopsis thaliana iron deficiency response. PLoS ONE, 10(8):e0136591, 2015.