Inferring Gene Networks for Strains of Dehalococcoides Highlights Conserved Relationships between Genes Encoding Core Catabolic and Cell-Wall Structural Proteins

The interpretation of high-throughput gene expression data for non-model microorganisms remains obscured because of the high fraction of hypothetical genes and the limited number of methods for the robust inference of gene networks. Therefore, to elucidate gene-gene and gene-condition linkages in th...

Full description

Saved in:

Bibliographic Details
Published in:	PloS one 2016-11, Vol.11 (11), p.e0166234-e0166234
Main Authors:	Mansfeldt, Cresten B, Heavner, Gretchen W, Rowe, Annette R, Hayete, Boris, Church, Bruce W, Richardson, Ruth E
Format:	Article
Language:	eng
Subjects:	Analysis Bayesian analysis Biology and life sciences Biomarkers Bioremediation Cell culture Cell Wall - genetics Cell Wall Skeleton - genetics Cell walls Chloroflexi - genetics Chloroflexi - metabolism Computer and Information Sciences Computer simulation Consensus Sequence - genetics Datasets Dehalococcoides DNA microarrays Engineering Environmental engineering Environmental science Enzymes Formate dehydrogenase Gene expression Gene Regulatory Networks - genetics Genes Genetic aspects Genomes Genomics Hydrogenase Hypotheses Laboratories Linkages Metabolism Metabolism - genetics Metabolites Microorganisms Network topologies Networks Nickel Oligonucleotide Array Sequence Analysis Oxidases Proteins Random variables Research and Analysis Methods Respiration Reverse engineering Statistical inference Strains (organisms) Structural proteins Studies Transcription
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The interpretation of high-throughput gene expression data for non-model microorganisms remains obscured because of the high fraction of hypothetical genes and the limited number of methods for the robust inference of gene networks. Therefore, to elucidate gene-gene and gene-condition linkages in the bioremediation-important genus Dehalococcoides, we applied a Bayesian inference strategy called Reverse Engineering/Forward Simulation (REFS™) on transcriptomic data collected from two organohalide-respiring communities containing different Dehalococcoides mccartyi strains: the Cornell University mixed community D2 and the commercially available KB-1® bioaugmentation culture. In total, 49 and 24 microarray datasets were included in the REFS™ analysis to generate an ensemble of 1,000 networks for the Dehalococcoides population in the Cornell D2 and KB-1® culture, respectively. Considering only linkages that appeared in the consensus network for each culture (exceeding the determined frequency cutoff of ≥ 60%), the resulting Cornell D2 and KB-1® consensus networks maintained 1,105 nodes (genes or conditions) with 974 edges and 1,714 nodes with 1,455 edges, respectively. These consensus networks captured multiple strong and biologically informative relationships. One of the main highlighted relationships shared between these two cultures was a direct edge between the transcript encoding for the major reductive dehalogenase (tceA (D2) or vcrA (KB-1®)) and the transcript for the putative S-layer cell wall protein (DET1407 (D2) or KB1_1396 (KB-1®)). Additionally, transcripts for two key oxidoreductases (a [Ni Fe] hydrogenase, Hup, and a protein with similarity to a formate dehydrogenase, "Fdh") were strongly linked, generalizing a strong relationship noted previously for Dehalococcoides mccartyi strain 195 to multiple strains of Dehalococcoides. Notably, the pangenome array utilized when monitoring the KB-1® culture was capable of resolving signals from multiple strains, and the network inference engine was able to reconstruct gene networks in the distinct strain populations.
ISSN:	1932-6203 1932-6203