Estimating the per-base-pair mutation rate in the yeast Saccharomyces cerevisiae

Abstract

Although mutation rates are a key determinant of the rate of evolution they are difficult to measure precisely and global mutations rates (mutations per genome per generation) are often extrapolated from the per-base-pair mutation rate assuming that mutation rate is uniform across the genome. Using budding yeast, we describe an improved method for the accurate calculation of mutation rates based on the fluctuation assay. Our analysis suggests that the per-base-pair mutation rates at two genes differ significantly (3.80x10(-10) at URA3 and 6.44x10(-10) at CAN1) and we propose a definition for the effective target size of genes (the probability that a mutation inactivates the gene) that acknowledges that the mutation rate is nonuniform across the genome.

Figure 1.—

Results from three 72-tube fluctuation assays using GIL104 clone A plated onto 1× canavanine, 5-FOA, 10× canavanine, and α-factor. Solid circles show the cumulative distribution of the data. Solid curves indicate the cumulative Luria–Delbrück (one-parameter model) distributions fit to the data with parameter m = 4.80, 1.31, 2.82, and 1.97 for 1× canavanine, 5-FOA, 10× canavanine, and α-factor, respectively. The thick shaded curve is the two-parameter model of postplating growth fit to the data with m = 2.31, d = 2.62; m = 2.39, d = 0.37; and m = 1.10 and d = 1.42 for 1× canavanine and 10× canavanine, respectively. The one-parameter and two-parameter models are the same for 5-FOA. Using Akaike's information criterion (Akaike 1974), the 5-FOA and 10× canavanine fluctuation assays are best described by the one-parameter model; whereas, the 1× canavanine and α-factor fluctuation assays are best described by the two-parameter model.

Figure 2.—

Mutational spectrum of 207 5-FOA-resistant mutations at the URA3 locus. Solid text represents missense mutations; open text on a solid background represents nonsense mutations. A horizontal line separates different mutations at the same codon. Complex mutational events such as large duplications and multiple mutations in the same strain are indicated below the nucleotide sequence.

Figure 3.—

Mutational spectrum of 227 10× canavanine mutations at the CAN1 locus. Solid text represents missense mutations; open text on a solid background represents nonsense mutations. A horizontal line separates different mutations at the same codon. Complex mutational events such as large duplications and multiple mutations in the same strain are indicated below the nucleotide sequence.

Figure 3.—

Mutational spectrum of 227 10× canavanine mutations at the CAN1 locus. Solid text represents missense mutations; open text on a solid background represents nonsense mutations. A horizontal line separates different mutations at the same codon. Complex mutational events such as large duplications and multiple mutations in the same strain are indicated below the nucleotide sequence.