Rates and spectra of de novo structural mutations in Chlamydomonas reinhardtii.

Genetic variation originates from several types of spontaneous mutation, including single-nucleotide substitutions, short insertions and deletions (indels), and larger structural changes. Structural mutations (SMs) drive genome evolution and are thought to play major roles in evolutionary adaptation, speciation, and genetic disease, including cancers. Sequencing of mutation accumulation (MA) lines has provided estimates of rates and spectra of single-nucleotide and indel mutations in many species, yet the rate of new SMs is largely unknown. Here, we use long-read sequencing to determine the full mutation spectrum in MA lines derived from two strains (CC-1952 and CC-2931) of the green alga Chlamydomonas reinhardtii The SM rate is highly variable between strains and between MA lines, and SMs represent a substantial proportion of all mutations in both strains (CC-1952 6%; CC-2931 12%). The SM spectra differ considerably between the two strains, with almost all inversions and translocations occurring in CC-2931 MA lines. This variation is associated with heterogeneity in the number and type of active transposable elements (TEs), which comprise major proportions of SMs in both strains (CC-1952 22%; CC-2931 38%). In CC-2931, a Crypton and a previously undescribed type of DNA element have caused 71% of chromosomal rearrangements, whereas in CC-1952, a Dualen LINE is associated with 87% of duplications. Other SMs, notably large duplications in CC-2931, are likely products of various double-strand break repair pathways. Our results show that diverse types of SMs occur at substantial rates, and support prominent roles for SMs and TEs in evolution.

Variation in mutation, recombination, and transposition rates in Drosophila melanogaster and Drosophila simulans.

The rates of mutation, recombination, and transposition are core parameters in models of evolution. They impact genetic diversity, responses to ongoing selection, and levels of genetic load. However, even for key evolutionary model species such as Drosophila melanogaster and Drosophila simulans, few estimates of these parameters are available, and we have little idea of how rates vary between individuals, sexes, or populations. Knowledge of this variation is fundamental for parameterizing models of genome evolution. Here, we provide direct estimates of mutation, recombination, and transposition rates and their variation in a West African and a European population of D. melanogaster and a European population of D. simulans Across 89 flies, we observe 58 single-nucleotide mutations, 286 crossovers, and 89 transposable element (TE) insertions. Compared to the European D. melanogaster, we find the West African population has a lower mutation rate (1.67 × 10-9 site-1 gen-1 vs. 4.86 × 10-9 site-1 gen-1) and a lower transposition rate (8.99 × 10-5 copy-1 gen-1 vs. 23.36 × 10-5 copy-1 gen-1), but a higher recombination rate (3.44 cM/Mb vs. 2.06 cM/Mb). The European D. simulans population has a similar mutation rate to European D. melanogaster, but a significantly higher recombination rate and a lower, but not significantly different, transposition rate. Overall, we find paternal-derived mutations are more frequent than maternal ones in both species. Our study quantifies the variation in rates of mutation, recombination, and transposition among different populations and sexes, and our direct estimates of these parameters in D. melanogaster and D. simulans will benefit future studies in population and evolutionary genetics.

Variation in the Spectrum of New Mutations among Inbred Strains of Mice.

The mouse serves as a mammalian model for understanding the nature of variation from new mutations, a question that has both evolutionary and medical significance. Previous studies suggest that the rate of single-nucleotide mutations (SNMs) in mice is ∼50% of that in humans. However, information largely comes from studies involving the C57BL/6 strain, and there is little information from other mouse strains. Here, we study the mutations that accumulated in 59 mouse lines derived from four inbred strains that are commonly used in genetics and clinical research (BALB/cAnNRj, C57BL/6JRj, C3H/HeNRj, and FVB/NRj), maintained for eight to nine generations by brother-sister mating. By analyzing Illumina whole-genome sequencing data, we estimate that the average rate of new SNMs in mice is ∼µ = 6.7 × 10-9. However, there is substantial variation in the spectrum of SNMs among strains, so the burden from new mutations also varies among strains. For example, the FVB strain has a spectrum that is markedly skewed toward C→A transversions and is likely to experience a higher deleterious load than other strains, due to an increased frequency of nonsense mutations in glutamic acid codons. Finally, we observe substantial variation in the rate of new SNMs among DNA sequence contexts, CpG sites, and their adjacent nucleotides playing an important role.

Recommendations for improving statistical inference in population genomics.

The field of population genomics has grown rapidly in response to the recent advent of affordable, large-scale sequencing technologies. As opposed to the situation during the majority of the 20th century, in which the development of theoretical and statistical population genetic insights outpaced the generation of data to which they could be applied, genomic data are now being produced at a far greater rate than they can be meaningfully analyzed and interpreted. With this wealth of data has come a tendency to focus on fitting specific (and often rather idiosyncratic) models to data, at the expense of a careful exploration of the range of possible underlying evolutionary processes. For example, the approach of directly investigating models of adaptive evolution in each newly sequenced population or species often neglects the fact that a thorough characterization of ubiquitous nonadaptive processes is a prerequisite for accurate inference. We here describe the perils of these tendencies, present our consensus views on current best practices in population genomic data analysis, and highlight areas of statistical inference and theory that are in need of further attention. Thereby, we argue for the importance of defining a biologically relevant baseline model tuned to the details of each new analysis, of skepticism and scrutiny in interpreting model fitting results, and of carefully defining addressable hypotheses and underlying uncertainties.

The distribution of fitness effects of spontaneous mutations in Chlamydomonas reinhardtii inferred using frequency changes under experimental evolution.

The distribution of fitness effects (DFE) for new mutations is fundamental for many aspects of population and quantitative genetics. In this study, we have inferred the DFE in the single-celled alga Chlamydomonas reinhardtii by estimating changes in the frequencies of 254 spontaneous mutations under experimental evolution and equating the frequency changes of linked mutations with their selection coefficients. We generated seven populations of recombinant haplotypes by crossing seven independently derived mutation accumulation lines carrying an average of 36 mutations in the haploid state to a mutation-free strain of the same genotype. We then allowed the populations to evolve under natural selection in the laboratory by serial transfer in liquid culture. We observed substantial and repeatable changes in the frequencies of many groups of linked mutations, and, surprisingly, as many mutations were observed to increase as decrease in frequency. Mutation frequencies were highly repeatable among replicates, suggesting that selection was the cause of the observed allele frequency changes. We developed a Bayesian Monte Carlo Markov Chain method to infer the DFE. This computes the likelihood of the observed distribution of changes of frequency, and obtains the posterior distribution of the selective effects of individual mutations, while assuming a two-sided gamma distribution of effects. We infer that the DFE is a highly leptokurtic distribution, and that approximately equal proportions of mutations have positive and negative effects on fitness. This result is consistent with what we have observed in previous work on a different C. reinhardtii strain, and suggests that a high fraction of new spontaneously arisen mutations are advantageous in a simple laboratory environment.

Comparative genomics of Chlamydomonas.

Despite its role as a reference organism in the plant sciences, the green alga Chlamydomonas reinhardtii entirely lacks genomic resources from closely related species. We present highly contiguous and well-annotated genome assemblies for three unicellular C. reinhardtii relatives: Chlamydomonas incerta, Chlamydomonas schloesseri, and the more distantly related Edaphochlamys debaryana. The three Chlamydomonas genomes are highly syntenous with similar gene contents, although the 129.2 Mb C. incerta and 130.2 Mb C. schloesseri assemblies are more repeat-rich than the 111.1 Mb C. reinhardtii genome. We identify the major centromeric repeat in C. reinhardtii as a LINE transposable element homologous to Zepp (the centromeric repeat in Coccomyxa subellipsoidea) and infer that centromere locations and structure are likely conserved in C. incerta and C. schloesseri. We report extensive rearrangements, but limited gene turnover, between the minus mating type loci of these Chlamydomonas species. We produce an eight-species core-Reinhardtinia whole-genome alignment, which we use to identify several hundred false positive and missing genes in the C. reinhardtii annotation and >260,000 evolutionarily conserved elements in the C. reinhardtii genome. In summary, these resources will enable comparative genomics analyses for C. reinhardtii, significantly extending the analytical toolkit for this emerging model system.

De Novo Mutation Rate Variation and Its Determinants in Chlamydomonas.

De novo mutations are central for evolution, since they provide the raw material for natural selection by regenerating genetic variation. However, studying de novo mutations is challenging and is generally restricted to model species, so we have a limited understanding of the evolution of the mutation rate and spectrum between closely related species. Here, we present a mutation accumulation (MA) experiment to study de novo mutation in the unicellular green alga Chlamydomonas incerta and perform comparative analyses with its closest known relative, Chlamydomonas reinhardtii. Using whole-genome sequencing data, we estimate that the median single nucleotide mutation (SNM) rate in C. incerta is µ = 7.6 × 10-10, and is highly variable between MA lines, ranging from µ = 0.35 × 10-10 to µ = 131.7 × 10-10. The SNM rate is strongly positively correlated with the mutation rate for insertions and deletions between lines (r > 0.97). We infer that the genomic factors associated with variation in the mutation rate are similar to those in C. reinhardtii, allowing for cross-prediction between species. Among these genomic factors, sequence context and complexity are more important than GC content. With the exception of a remarkably high C→T bias, the SNM spectrum differs markedly between the two Chlamydomonas species. Our results suggest that similar genomic and biological characteristics may result in a similar mutation rate in the two species, whereas the SNM spectrum has more freedom to diverge.

Inferring the distribution of fitness effects of spontaneous mutations in Chlamydomonas reinhardtii.

Spontaneous mutations are the source of new genetic variation and are thus central to the evolutionary process. In molecular evolution and quantitative genetics, the nature of genetic variation depends critically on the distribution of effects of mutations on fitness and other quantitative traits. Spontaneous mutation accumulation (MA) experiments have been the principal approach for investigating the overall rate of occurrence and cumulative effect of mutations but have not allowed the phenotypic effects of individual mutations to be studied directly. Here, we crossed MA lines of the green alga Chlamydomonas reinhardtii with its unmutated ancestral strain to create haploid recombinant lines, each carrying an average of 50% of the accumulated mutations in a large number of combinations. With the aid of the genome sequences of the MA lines, we inferred the genotypes of the mutations, assayed their growth rate as a measure of fitness, and inferred the distribution of fitness effects (DFE) using a Bayesian mixture model. We infer that the DFE is highly leptokurtic (L-shaped). Of mutations with absolute fitness effects exceeding 1%, about one-sixth increase fitness in the laboratory environment. The inferred distribution of effects for deleterious mutations is consistent with a strong role for nearly neutral evolution. Specifically, such a distribution predicts that nucleotide variation and genetic variation for quantitative traits will be insensitive to change in the effective population size.

Inbred lab mice are not isogenic: genetic variation within inbred strains used to infer the mutation rate per nucleotide site.

For over a century, inbred mice have been used in many areas of genetics research to gain insight into the genetic variation underlying traits of interest. The generalizability of any genetic research study in inbred mice is dependent upon all individual mice being genetically identical, which in turn is dependent on the breeding designs of companies that supply inbred mice to researchers. Here, we compare whole-genome sequences from individuals of four commonly used inbred strains that were procured from either the colony nucleus or from a production colony (which can be as many as ten generations removed from the nucleus) of a large commercial breeder, in order to investigate the extent and nature of genetic variation within and between individuals. We found that individuals within strains are not isogenic, and there are differences in the levels of genetic variation that are explained by differences in the genetic distance from the colony nucleus. In addition, we employ a novel approach to mutation rate estimation based on the observed genetic variation and the expected site frequency spectrum at equilibrium, given a fully inbred breeding design. We find that it provides a reasonable per nucleotide mutation rate estimate when mice come from the colony nucleus (~7.9 × 10-9 in C3H/HeN), but substantially inflated estimates when mice come from production colonies.

Understanding the Factors That Shape Patterns of Nucleotide Diversity in the House Mouse Genome.

A major goal of population genetics has been to determine the extent by which selection at linked sites influences patterns of neutral nucleotide diversity in the genome. Multiple lines of evidence suggest that diversity is influenced by both positive and negative selection. For example, in many species there are troughs in diversity surrounding functional genomic elements, consistent with the action of either background selection (BGS) or selective sweeps. In this study, we investigated the causes of the diversity troughs that are observed in the wild house mouse genome. Using the unfolded site frequency spectrum, we estimated the strength and frequencies of deleterious and advantageous mutations occurring in different functional elements in the genome. We then used these estimates to parameterize forward-in-time simulations of chromosomes, using realistic distributions of functional elements and recombination rate variation in order to determine whether selection at linked sites can explain the observed patterns of nucleotide diversity. The simulations suggest that BGS alone cannot explain the dips in diversity around either exons or conserved noncoding elements. A combination of BGS and selective sweeps produces deeper dips in diversity than BGS alone, but the inferred parameters of selection cannot fully explain the patterns observed in the genome. Our results provide evidence of sweeps shaping patterns of nucleotide diversity across the mouse genome and also suggest that infrequent, strongly advantageous mutations play an important role in this. The limitations of using the unfolded site frequency spectrum for inferring the frequency and effects of advantageous mutations are discussed.

Patterns of population structure and complex haplotype sharing among field isolates of the green alga Chlamydomonas reinhardtii.

The nature of population structure in microbial eukaryotes has long been debated. Competing models have argued that microbial species are either ubiquitous, with high dispersal and low rates of speciation, or that for many species gene flow between populations is limited, resulting in evolutionary histories similar to those of macroorganisms. However, population genomic approaches have seldom been applied to this question. Here, we analyse whole-genome resequencing data for all 36 confirmed field isolates of the green alga Chlamydomonas reinhardtii. At a continental scale, we report evidence for putative allopatric divergence, between both North American and Japanese isolates, and two highly differentiated lineages within N. America. Conversely, at a local scale within the most densely sampled lineage, we find little evidence for either spatial or temporal structure. Taken together with evidence for ongoing admixture between the two N. American lineages, this lack of structure supports a role for substantial dispersal in C. reinhardtii and implies that between-lineage differentiation may be maintained by reproductive isolation and/or local adaptation. Our results therefore support a role for allopatric divergence in microbial eukaryotes, while also indicating that species may be ubiquitous at local scales. Despite the high genetic diversity observed within the most well-sampled lineage, we find that pairs of isolates share on average ~9% of their genomes in long haplotypes, even when isolates were sampled decades apart and from different locations. This proportion is several orders of magnitude higher than the Wright-Fisher expectation, raising many further questions concerning the evolutionary genetics of C. reinhardtii and microbial eukaryotes generally.

Mitochondrial Mutation Rate, Spectrum and Heteroplasmy in Caenorhabditis elegans Spontaneous Mutation Accumulation Lines of Differing Population Size.

Mitochondrial genomes of metazoans, given their elevated rates of evolution, have served as pivotal markers for phylogeographic studies and recent phylogenetic events. In order to determine the dynamics of spontaneous mitochondrial mutations in small populations in the absence and presence of selection, we evolved mutation accumulation (MA) lines of Caenorhabditis elegans in parallel over 409 consecutive generations at three varying population sizes of N = 1, 10, and 100 hermaphrodites. The N =1 populations should have a minimal influence of natural selection to provide the spontaneous mutation rate and the expected rate of neutral evolution, whereas larger population sizes should experience increasing intensity of selection. New mutations were identified by Illumina paired-end sequencing of 86 mtDNA genomes across 35 experimental lines and compared with published genomes of natural isolates. The spontaneous mitochondrial mutation rate was estimated at 1.05 × 10-7/site/generation. A strong G/C→A/T mutational bias was observed in both the MA lines and the natural isolates. This suggests that the low G + C content at synonymous sites is the product of mutation bias rather than selection as previously proposed. The mitochondrial effective population size per worm generation was estimated to be 62. Although it was previously concluded that heteroplasmy was rare in C. elegans, the vast majority of mutations in this study were heteroplasmic despite an experimental regime exceeding 400 generations. The frequencies of frameshift and nonsynonymous mutations were negatively correlated with population size, which suggests their deleterious effects on fitness and a potent role for selection in their eradication.

Extensive de novo mutation rate variation between individuals and across the genome of Chlamydomonas reinhardtii.

Describing the process of spontaneous mutation is fundamental for understanding the genetic basis of disease, the threat posed by declining population size in conservation biology, and much of evolutionary biology. Directly studying spontaneous mutation has been difficult, however, because new mutations are rare. Mutation accumulation (MA) experiments overcome this by allowing mutations to build up over many generations in the near absence of natural selection. Here, we sequenced the genomes of 85 MA lines derived from six genetically diverse strains of the green alga Chlamydomonas reinhardtii. We identified 6843 new mutations, more than any other study of spontaneous mutation. We observed sevenfold variation in the mutation rate among strains and that mutator genotypes arose, increasing the mutation rate approximately eightfold in some replicates. We also found evidence for fine-scale heterogeneity in the mutation rate, with certain sequence motifs mutating at much higher rates, and clusters of multiple mutations occurring at closely linked sites. There was little evidence, however, for mutation rate heterogeneity between chromosomes or over large genomic regions of 200 kbp. We generated a predictive model of the mutability of sites based on their genomic properties, including local GC content, gene expression level, and local sequence context. Our model accurately predicted the average mutation rate and natural levels of genetic diversity of sites across the genome. Notably, trinucleotides vary 17-fold in rate between the most and least mutable sites. Our results uncover a rich heterogeneity in the process of spontaneous mutation both among individuals and across the genome.

Detecting positive selection in the genome.

Population geneticists have long sought to understand the contribution of natural selection to molecular evolution. A variety of approaches have been proposed that use population genetics theory to quantify the rate and strength of positive selection acting in a species' genome. In this review we discuss methods that use patterns of between-species nucleotide divergence and within-species diversity to estimate positive selection parameters from population genomic data. We also discuss recently proposed methods to detect positive selection from a population's haplotype structure. The application of these tests has resulted in the detection of pervasive adaptive molecular evolution in multiple species.

Direct Estimate of the Spontaneous Mutation Rate Uncovers the Effects of Drift and Recombination in the Chlamydomonas reinhardtii Plastid Genome.

Plastids perform crucial cellular functions, including photosynthesis, across a wide variety of eukaryotes. Since endosymbiosis, plastids have maintained independent genomes that now display a wide diversity of gene content, genome structure, gene regulation mechanisms, and transmission modes. The evolution of plastid genomes depends on an input of de novo mutation, but our knowledge of mutation in the plastid is limited to indirect inference from patterns of DNA divergence between species. Here, we use a mutation accumulation experiment, where selection acting on mutations is rendered ineffective, combined with whole-plastid genome sequencing to directly characterize de novo mutation in Chlamydomonas reinhardtii. We show that the mutation rates of the plastid and nuclear genomes are similar, but that the base spectra of mutations differ significantly. We integrate our measure of the mutation rate with a population genomic data set of 20 individuals, and show that the plastid genome is subject to substantially stronger genetic drift than the nuclear genome. We also show that high levels of linkage disequilibrium in the plastid genome are not due to restricted recombination, but are instead a consequence of increased genetic drift. One likely explanation for increased drift in the plastid genome is that there are stronger effects of genetic hitchhiking. The presence of recombination in the plastid is consistent with laboratory studies in C. reinhardtii and demonstrates that although the plastid genome is thought to be uniparentally inherited, it recombines in nature at a rate similar to the nuclear genome.

Recent Evolution in Rattus norvegicus Is Shaped by Declining Effective Population Size.

The brown rat, Rattus norvegicus, is both a notorious pest and a frequently used model in biomedical research. By analyzing genome sequences of 12 wild-caught brown rats from their presumed ancestral range in NE China, along with the sequence of a black rat, Rattus rattus, we investigate the selective and demographic forces shaping variation in the genome. We estimate that the recent effective population size (Ne) of this species = [Formula: see text], based on silent site diversity. We compare patterns of diversity in these genomes with patterns in multiple genome sequences of the house mouse (Mus musculus castaneus), which has a much larger Ne. This reveals an important role for variation in the strength of genetic drift in mammalian genome evolution. By a Pairwise Sequentially Markovian Coalescent analysis of demographic history, we infer that there has been a recent population size bottleneck in wild rats, which we date to approximately 20,000 years ago. Consistent with this, wild rat populations have experienced an increased flux of mildly deleterious mutations, which segregate at higher frequencies in protein-coding genes and conserved noncoding elements. This leads to negative estimates of the rate of adaptive evolution (α) in proteins and conserved noncoding elements, a result which we discuss in relation to the strongly positive estimates observed in wild house mice. As a consequence of the population bottleneck, wild rats also show a markedly slower decay of linkage disequilibrium with physical distance than wild house mice.

Estimation of the spontaneous mutation rate in Heliconius melpomene.

We estimated the spontaneous mutation rate in Heliconius melpomene by genome sequencing of a pair of parents and 30 of their offspring, based on the ratio of number of de novo heterozygotes to the number of callable site-individuals. We detected nine new mutations, each one affecting a single site in a single offspring. This yields an estimated mutation rate of 2.9 × 10(-9) (95% confidence interval, 1.3 × 10(-9)-5.5 × 10(-9)), which is similar to recent estimates in Drosophila melanogaster, the only other insect species in which the mutation rate has been directly estimated. We infer that recent effective population size of H. melpomene is about 2 million, a substantially lower value than its census size, suggesting a role for natural selection reducing diversity. We estimate that H. melpomene diverged from its Müllerian comimic H. erato about 6 Ma, a somewhat later date than estimates based on a local molecular clock.

Global population divergence and admixture of the brown rat (Rattus norvegicus).

Native to China and Mongolia, the brown rat (Rattus norvegicus) now enjoys a worldwide distribution. While black rats and the house mouse tracked the regional development of human agricultural settlements, brown rats did not appear in Europe until the 1500s, suggesting their range expansion was a response to relatively recent increases in global trade. We inferred the global phylogeography of brown rats using 32 k SNPs, and detected 13 evolutionary clusters within five expansion routes. One cluster arose following a southward expansion into Southeast Asia. Three additional clusters arose from two independent eastward expansions: one expansion from Russia to the Aleutian Archipelago, and a second to western North America. Westward expansion resulted in the colonization of Europe from which subsequent rapid colonization of Africa, the Americas and Australasia occurred, and multiple evolutionary clusters were detected. An astonishing degree of fine-grained clustering between and within sampling sites underscored the extent to which urban heterogeneity shaped genetic structure of commensal rodents. Surprisingly, few individuals were recent migrants, suggesting that recruitment into established populations is limited. Understanding the global population structure of R. norvegicus offers novel perspectives on the forces driving the spread of zoonotic disease, and aids in development of rat eradication programmes.

Contributions of protein-coding and regulatory change to adaptive molecular evolution in murid rodents.

The contribution of regulatory versus protein change to adaptive evolution has long been controversial. In principle, the rate and strength of adaptation within functional genetic elements can be quantified on the basis of an excess of nucleotide substitutions between species compared to the neutral expectation or from effects of recent substitutions on nucleotide diversity at linked sites. Here, we infer the nature of selective forces acting in proteins, their UTRs and conserved noncoding elements (CNEs) using genome-wide patterns of diversity in wild house mice and divergence to related species. By applying an extension of the McDonald-Kreitman test, we infer that adaptive substitutions are widespread in protein-coding genes, UTRs and CNEs, and we estimate that there are at least four times as many adaptive substitutions in CNEs and UTRs as in proteins. We observe pronounced reductions in mean diversity around nonsynonymous sites (whether or not they have experienced a recent substitution). This can be explained by selection on multiple, linked CNEs and exons. We also observe substantial dips in mean diversity (after controlling for divergence) around protein-coding exons and CNEs, which can also be explained by the combined effects of many linked exons and CNEs. A model of background selection (BGS) can adequately explain the reduction in mean diversity observed around CNEs. However, BGS fails to explain the wide reductions in mean diversity surrounding exons (encompassing ~100 Kb, on average), implying that there is a substantial role for adaptation within exons or closely linked sites. The wide dips in diversity around exons, which are hard to explain by BGS, suggest that the fitness effects of adaptive amino acid substitutions could be substantially larger than substitutions in CNEs. We conclude that although there appear to be many more adaptive noncoding changes, substitutions in proteins may dominate phenotypic evolution.

Direct estimation of per nucleotide and genomic deleterious mutation rates in Drosophila.

Spontaneous mutations are the source of genetic variation required for evolutionary change, and are therefore important for many aspects of evolutionary biology. For example, the divergence between taxa at neutrally evolving sites in the genome is proportional to the per nucleotide mutation rate, u (ref. 1), and this can be used to date speciation events by assuming a molecular clock. The overall rate of occurrence of deleterious mutations in the genome each generation (U) appears in theories of nucleotide divergence and polymorphism, the evolution of sex and recombination, and the evolutionary consequences of inbreeding. However, estimates of U based on changes in allozymes or DNA sequences and fitness traits are discordant. Here we directly estimate u in Drosophila melanogaster by scanning 20 million bases of DNA from three sets of mutation accumulation lines by using denaturing high-performance liquid chromatography. From 37 mutation events that we detected, we obtained a mean estimate for u of 8.4 x 10(-9) per generation. Moreover, we detected significant heterogeneity in u among the three mutation-accumulation-line genotypes. By multiplying u by an estimate of the fraction of mutations that are deleterious in natural populations of Drosophila, we estimate that U is 1.2 per diploid genome. This high rate suggests that selection against deleterious mutations may have a key role in explaining patterns of genetic variation in the genome, and help to maintain recombination and sexual reproduction.