Shabanova, Ekaterina (2009) Patterns of genetic recombination and variation in the human genome. PhD thesis, Universität zu Köln.
Genetic recombination plays an important role in shaping genome variation. It enhances haplotype diversity, helps to maintain genome integrity and ensures the proper segregation of the chromosomes. While participating in DNA rearrangement, recombination is not an �independent player�. It is tightly connected and influenced by other genomic features, such as nucleotide diversity. A correlation between nucleotide diversity and recombination rate was observed in the human genome, as well as in the genomes of other organisms (Arabidopsis, Drosophila, etc). The traditional view that selection has contributed to shape this pattern was questioned by the view that it may solely be due to a mutagenic effect. Extensive analysis of the data from dbSNP and broad scale recombination maps revealed a high degree of uncertainty in the inferred correlation coefficients. One goal of this study was to re-assess the magnitude and disentangle the possible reasons for this observation. The results show that there is no evidence for a presence of strong correlation between nucleotide diversity and recombination rate. In fact, the observed effect can be due to insufficient data or poor data quality in earlier studies. Analysis of the more recent data shows that the correlation between diversity and recombination may be due to sequence composition, such as sequence composition. While looking at the fine scale it became clear that recombination hotspots are very ephemeral structures, which do not have influence on long-term molecular evolution. Only detailed experimental studies can reveal, whether the apparent correlation between diversity and recombination rate has a causal connection. In order to obtain high resolution recombination rates, a single sperm typing approach was applied to 2.5 Mb sequence consisting of four Encode regions on human chromosome 11. The regions were selected not on the basis of classical linkage studies, but in unbiased fashion. The outcome revealed a cross over rate of 0.12 cM/Mb, which was much lower compared to the expected 1.875 cM/Mb according to high resolution recombination maps. We confirm an increased gene conversion rate compared to cross over. Out of 10 recombinants, 7 had conversion haplotypes. We assume that such a low cross over rate can be due to the individual or sex-specific variation. Our data did not support genetic characteristics known to be associated with gene conversion or recombination in general. For instance, identified conversion tracts did not have a high GC content. No association between converted or cross over regions and a 13-mer degenerate motif (predicted to be associated with recombination hotspots) was observed. The observed CpG fraction was so low compared to the expected, that the possible explanation could be a extensive methylation of the converted regions. Finally, we hypothesize, that there is a difference in regulation mechanism as well as in frequency of double-strand breaks resolved as conversion events in short and long tract conversions. The frequency of the breaks can also be a reason to differentiate hotspots-associated conversions and the ones occurring in coding regions.
Actions (login required)