Degree Type


Date of Award


Degree Name

Doctor of Philosophy




Ecology and Evolutionary Biology

First Advisor

Jean-Luc Jannink

Second Advisor

John D. Nason

Third Advisor

Phillip Dixon


Phenotypic selection can be tested by comparing the inter-population variation in quantitative traits (QST) against inter-population variation in neutral molecular markers (FST). A trait is considered under divergent selection when QST > FST, under stabilizing selection when QST < FST, and selectively neutral when QST = FST. This approach assumes the form of selection to be consistent across populations and can fail to detect selection mosaics in which this consistency is lacking. The overall objectives of this work are to improve the methodology for comparing QST and FST, and to study the relationship between QST and FST in cultivated and wild barley. First, we develop a methodology to detect situations in which the form of selection is heterogenous. The methodology simulates pair-wise QST under neutrality and focuses on the distribution of QST and FST. Our results indicate that studies employing traditional QST and FST methods are insensitive to mixtures of disruptive and stabilizing selection and that more powerful inference concerning the form of selection may be gained via the methods introduced here. Second, we use overall and pair-wise estimates to study the evolutionary history of quantitative traits in wild barley. We evaluate 14 traits and 56 SSR markers in 280 genotypes from 23 populations of wild barley from the Fertile Crescent. We find that a combination of overall and pair-wise comparisons is better suited than the former alone to understand the evolutionary processes that shape populations. Third, we propose the use of QST and F ST studies in cultivated barley. We use 66 polymorphic SSR and 20 quantitative traits in 353 genotypes of barley from 23 breeding programs distributed worldwide. By simultaneously estimating population structure at morphological traits and neutral molecular markers, we identify compatible breeding programs for germplasm exchange, and breeding programs with unique characteristics worth preserving. However, caution is advised in the use of this methodology for fitness-related traits, traits with important GxE, and germplasm with strong artificially-imposed structure especially in an artificial selection context as in breeding programs. In summary, we develop a methodology that is effective for detecting selection mosaics and we provided ample evidence of the occurrence of this phenomenon in both wild and cultivated barley.



Digital Repository @ Iowa State University,

Copyright Owner

Lucía Gutiérrez



Proquest ID


OCLC Number




File Format


File Size

135 pages