Iowa State University Brigham Young University University of Georgia







Overview
Phylogeny
Fiber Evolution


Introgression Populations
Homoeolog-specific Profiling
Genetic Networks & Phenotype
Effects of Selection
Sequence Capture

Genetic and Physical mapping resources
Comparative BAC Sequencing
Genome Sequence Resources
EST D-genome map
EST Resources
Microarray

Web Database
Education and Outreach
Significance for cotton industry
Cotton Literature
Cotton Links
Events
Wendel Lab
PGML (Paterson Lab)
Udall Lab

Lists & protocols
Publications
How to
CEGC Site Search

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
turn explanations on/off

Perturbations in Genetic Networks and Gene Expression


Overview | Procedures | Genetic Networks | cis and trans effects | People | Publications

Procedures

For each of the 40-60 selected NILs, plants will be grown in growth chambers and fibers will be collected from developmental stages representing primary and secondary wall synthesis (~10 and 25 days post-anthesis, respectively). Using methods we have optimized for microarray experiments, fibers will be isolated and RNAs extracted using multiple ovules from multiple flowers of several individuals (the lines will be homozygous). Three biological replicates will be included for each NIL studied. The same procedures will be applied to all four parental lines, as well as the A and D genome diploids G. arboreum and G. raimondii; the diploids are required for analytical reasons associated with the homoeolog-specific design and to polarize expression states in the polyploid relative to their antecedent conditions. Thus, 46-66 biological entities will be examined, three replicates of each using two fiber time-points; hence, we are budgeting for a total of 400 microarrays. cDNA synthesis, labeling reactions, and hybridizations will follow established protocols we and Nimblegen have previously employed.

Given that each NIL is expected to contain, on average, a single introgressed chromosome segment of 16 cM, and using the estimates that the cotton genome contains 4500 cM and 50,000 genes, we anticipate that on average a NIL will include ~180 genes (though this will vary tremendously due to the uneven distribution of genes in the genome). Given the deep EST sampling and our present assembly (http://cottonevolution.info), we anticipate that a high proportion of these genes will be represented on the chips, and that many if not most of these will be expressed during fiber development. Additional EST generation will allow the next microarray design to interrogate most genes in the genome as well as both homoeologs separately for a significant fraction of these; thus, two types of information will be generated, representing differential expression of gene pairs (not distinguished by homoeolog) and differential expression of homoeologs.

Statistical procedures will be modeled after our previous studies where mixed linear models were used to partition sources of expression variance. Hovav et al. analyzed traditional oligonucleotide data and identified genes that had significantly different expression between stages of fiber growth and in different species. In these studies, ‘dye’ and ‘treatment’ were modeled as fixed effects, and ‘microarray’ and ‘replicate’ were modeled as random effects. Flagel et al. used mixed linear models to empirically diagnose expression differences between homoeologs in allotetraploid cotton. Biases in homoeolog expression were determined by comparison of the allotetraploid genome-specific probe signals to a diploid genome-specific probe signals that constituted a known dose response curve. Only genes expressed equally in the diploid A- and D-genome tissues were represented in the dose response curve to insure that homoeolog bias was not confounded with gene expression levels. For this project, similar mixed models will be used to diagnose the contribution of introgressed genome segments into the allotetraploid genome. After calibrating allotetraploid expression levels to a diploid dose response curve, perturbations of homoeolog expressions levels will be diagnosed by comparing expression in NIL lines to expression in the recurrent parent (fixed treatment effects).


We welcome your comments and suggestions.