Genetic architecture of chalcone isomerase non-coding regions in common bean (Phaseolus vulgaris L.)

Sequence data for 2 non-coding regions of the chalcone isomerase gene were analyzed to study the genetic architecture of common bean ( Phaseolus vulgaris L.). One region corresponded to the first 596 nucleotides (nt) of the 5′-untranslated region (UTR). The other region was the 710 nt intron 3. Data were collected from 67 genotypes representing both landraces and cultivars from the geographical range of the cultivated form of the species. Variability in the 5′-UTR region was represented by single nucleotide polymorphisms (SNPs), whereas intron 3 variation was due to a collection of SNPs and insertion–deletion events. Diversity was greater in the 5′-UTR (π = 0.0175) than in intron 3 (π = 0.0089). For each region, diversity was greater for genotypes of Middle American than Andean origin. A single recombination event was observed, and the hybridization pattern necessary to derive the recombinant genotypes supported the previous observation of an ancestral gene pool from which modern domesticated genotypes are derived. For both regions, a strongly supported Andean group was observed, whereas the presence of 2 Middle American subgroups was also supported. Although a significantly positive Tajima’s D statistic was observed for the 5′-UTR for all genotypes, we conclude that this is more likely the result of a strong demographic effect and that balancing selection is occurring only among the Middle American genotypes.