Analysis of the genome sequences of three Drosophila melanogaster spontaneous mutation accumulation lines

Abstract

We inferred the rate and properties of new spontaneous mutations in Drosophila melanogaster by carrying out whole-genome shotgun sequencing-by-synthesis of three mutation accumulation (MA) lines that had been maintained by close inbreeding for an average of 262 generations. We tested for the presence of new mutations by generating alignments of each MA line to the D. melanogaster reference genome sequence and then compared these alignments base by base. We determined empirically that at least five reads at a site within each line are required for accurate single nucleotide mutation calling. We mapped a total of 174 single-nucleotide mutations, giving a single nucleotide mutation rate of 3.5 3 10 À9 per site per generation. There were no false positives in a random sample of 40 of these mutations checked by Sanger sequencing. Variation in the numbers of mutations among the MA lines was small and nonsignificant. Numbers of transition and transversion mutations were 86 and 88, respectively, implying that transition mutation rate is close to 23 the transversion rate. We observed 1.53 as many G or C ! A or T as A or T ! G or C mutations, implying that the G or C ! A or T mutation rate is close to 23 the A or T ! G or C mutation rate. The base composition of the genome is therefore not at an equilibrium determined solely by mutation. The predicted G + C content at mutational equilibrium (33%) is similar to that observed in transposable element remnants. Nearest-neighbor mutational context dependencies are nonsignificant, suggesting that this is a weak phenomenon in Drosophila. We also saw nonsignificant differences in the mutation rate between transcribed and untranscribed regions, implying that any transcription-coupled repair process is weak. Of seven short indel mutations confirmed, six were deletions, consistent with the deletion bias that is thought to exist in Drosophila.[Supplemental material is available online at www.genome.org.]The rates and properties of spontaneous mutations are important for many questions in evolutionary biology and molecular evolution. For example, under neutrality, the rate of molecular evolution is expected to be equal to the mutation rate, so betweenspecies molecular divergence can be used to date divergence times of species by assuming clock-like molecular evolution. Conversely, the rate of molecular divergence at silent sites between species can be used to estimate the mutation rate. However, this requires the assumption of neutrality, and values for the generation time and divergence dates of the species are also needed.Mutation accumulation (MA) experiments are an alternative way to directly study new mutational variation. MA lines are started by subdividing a homozygous progenitor strain, then allowing spontaneous mutations to accumulate, often for many tens of generations. The lines are maintained by a form of close inbreeding (typically full-sib mating or selfing) that reduces the effectiveness of natural selection, so the rate of fixation of mutations is expected to be close ...

A genome-wide view of Caenorhabditis elegans base-substitution mutation processes

“…An average bs of 2.7 (Ϯ0.4) ϫ 10 Ϫ9 per site per generation was estimated for the 10 MA-line genomes by using this approach, and MA-line-specific mutation rate estimates varied only Ϸ3-fold among genomes with overlapping confidence intervals. This rate is also highly similar to the recently reported D. melanogaster rate, 3.5 ϫ 10 Ϫ9 , based on Solexa analysis of three MA lines (9). The average mutation rate observed in the three C. elegans genomes analyzed by 454, bs ϭ 3.1 (Ϯ1.9) ϫ 10 Ϫ9 , was highly similar to that observed in the seven genomes analyzed by Solexa: bs ϭ 2.5 (Ϯ0.5) ϫ 10 Ϫ9 .…”

“…An analysis of Saccharomyces cerevisiae MA lines applied the power of new high-throughput 454 pyrosequencing technology but was still only able to identify 33 nuclear base substitutions because of the small size of the yeast genome, although it did reveal spectra consistent with previous observations using reporter genes (8). A recent analysis of three D. melanogaster MA lines using Illumina high-throughput sequencing technology identified 174 nuclear DNA point mutations and characterized genome-wide mutation rates and spectra in this species (9).…”

Direct mutation analysis by high-throughput sequencing: From germline to low-abundant, somatic variants

“…The reduced cost and higher throughput of MPS has led to the first genome-wide estimates for germline mutation rates and spectra in yeast[85], worms[86], plants[87], and flies[88]. The experiments, run on both 454 and Illumina machines, used accumulation lines, where inbred populations accumulated mutations over many generations.…”

“…References to data on numbers of germline cell divisions: human[9], fly[88], worm[86], A. thaliana [87]. …”

“…References to data on generational mutation rates: human[91], fly[88], worm[86], A. thaliana [87], yeast[85], E. coli [12]. …”