Evidence on DNA slippage step-length distribution

Simple sequence repeats in Neurospora crassa: distribution, polymorphism and evolutionary inference

Background

Simple sequence repeats (SSRs) have been successfully used for various genetic and evolutionary studies in eukaryotic systems. The eukaryotic model organism Neurospora crassa is an excellent system to study evolution and biological function of SSRs.

Results

We identified and characterized 2749 SSRs of 963 SSR types in the genome of N. crassa. The distribution of tri-nucleotide (nt) SSRs, the most common SSRs in N. crassa, was significantly biased in exons. We further characterized the distribution of 19 abundant SSR types (AST), which account for 71% of total SSRs in the N. crassa genome, using a Poisson log-linear model. We also characterized the size variation of SSRs among natural accessions using Polymorphic Index Content (PIC) and ANOVA analyses and found that there are genome-wide, chromosome-dependent and local-specific variations. Using polymorphic SSRs, we have built linkage maps from three line-cross populations.

Conclusion

Taking our computational, statistical and experimental data together, we conclude that 1) the distributions of the SSRs in the sequenced N. crassa genome differ systematically between chromosomes as well as between SSR types, 2) the size variation of tri-nt SSRs in exons might be an important mechanism in generating functional variation of proteins in N. crassa, 3) there are different levels of evolutionary forces in variation of amino acid repeats, and 4) SSRs are stable molecular markers for genetic studies in N. crassa.

Replication slippage versus point mutation rates in short tandem repeats of the human genome

Short tandem repeats (STRs) are subjected to two kinds of mutational modifications: point mutations and replication slippages. The latter is found to be the more frequent cause of STR modifications, but a satisfactory quantitative measure of the ratio of the two processes has yet to be determined. The comparison of entire genome sequences of closely enough related species enables one to obtain sufficient statistics by counting the differences in the STR regions. We analyzed human-chimpanzee DNA sequence alignments to obtain the counts of point mutations and replication slippage modifications. The results were compared with the results of a computer simulation, and the parameters quantifying the replication slippage probability as well as the probabilities of point mutations within the repeats were determined. It was found that within the STRs with repeated units consisting of one, two or three nucleotides, point mutations occur approximately twice as frequently as one would expect on the basis of the 1.2% difference between the human and chimpanzee genomes. As expected, the replication slippage probability is negligible below a 10-bp threshold and grows above this level. The replication slippage events outnumber the point mutations by one or two orders of magnitude, but are still lower by one order of magnitude relative to the mutability of the markers that are used for genotyping purposes.

Informational structure of two closely related eukaryotic genomes

Attempts to identify a species on the basis of its DNA sequence on purely statistical grounds have been formulated for more than a decade. The most prominent of such genome signatures relies on neighborhood correlations (i.e., dinucleotide frequencies) and, consequently, attributes species identification to mechanisms operating on the dinucleotide level (e.g., neighbor-dependent mutations). For the examples of Mus musculus and Rattus norvegicus we analyze short- and intermediate-range statistical correlations in DNA sequences. These correlation profiles are computed for all chromosomes of the two species. We find that with increasing range of correlations the capacity to distinguish between the species on the basis of this correlation profile is getting better and requires ever shorter sequence segments for obtaining a full species separation. This finding suggests that distinctive traits within the sequence are situated beyond the level of few nucleotides. The large-scale statistical patterning of DNA sequences on which such genome signatures are based is thus substantially determined by mobile elements (e.g., transposons and retrotransposons). The study and interspecies comparison of such correlation profiles can, therefore, reveal features of retrotransposition, segmental duplications, and other processes of genome evolution.