Correction: The limits of selection during maize domestication

Nature 398 , 236 - 239 (1999 ).

The primers used for PCR of the 3′ portion of tb1 amplified a duplicate locus (tb1 homeologue) in two samples (11 and 16). Inclusion of these homeologous sequences caused π to rise sharply at the 3′ end of the gene in Fig. 1. Using a new 3′ primer (gaggcatcatccagcagacgagaaa), tb1 sequences were re-isolated (Genbank accessions AF340187–AF340209). The redrawn figure (below) still shows π rising on average, but not as sharply. Because of the known problems with PCR, all statistical tests in the paper were based on sequences isolated as λ clones, and the PCR-isolated sequences were used only in Fig. 1. Thus, other than Fig. 1, no statements or conclusions need amendment. An HKA test with the newly isolated 3′ sequences shows no deviation from neutral expectations for maize (χ² = 0.49, P = 0.78), and π(×1,000) for these sequences is 6.7 for teosinte and 5.8 for maize, confirming that selection is not apparent in the 3′ region of tb1.

Figure 1: Predicted structure of teosinte branched1 (ref.25) and sliding-window analysis of polymorphism (π) in maize and teosinte.

For the sliding-window analysis, π was calculated for segments of 300 bp at 50-bp intervals. Sequences used in the analysis were the subset of the λ-cloned sequences for which we isolated the 3′ end by PCR (Table 1). The position of the Hin dIII (H) restriction endonuclease sites used in λ cloning are shown, as are the predicted exons (rectangles) and coding region (stippled).