NEW YORK (GenomeWeb) – A genome-wide analysis of de novo mutations has found that there are a number of mechanisms that influence how such mutations are distributed throughout the human genome.
Researchers from the US and the Netherlands analyzed some 11,000 de novo mutations in 250 families to find that many mutations are linked to paternal age and replication timing in germline cells, while others tend to cluster in functional regions, suggesting the involvement of a new mutagenic mechanism. Based on this and comparisons to chimpanzees, the team lead by Harvard University's Shamil Sunyaev and University Medical Center Utrecht's Paul de Bakker developed a genome-wide mutation rate map for humans, as they reported in Nature Genetics today.
"We describe here the most extensive catalog thus far of de novo germline mutations in healthy individuals, identifying several mechanisms influencing the distribution of mutations along the genome," Sunyaev, de Bakker, and their colleagues wrote in their paper. "In particular, clustered mutations suggest the existence of a new mutagenic mechanism, and the effect of replication timing on germline mutations depends on paternal age."
To examine genome-wide patterns of de novo mutations in humans, the researchers turned to 250 Dutch parent-offspring families, including 231 trios, 11 families with monozygotic twins, and eight families with dizygotic twins. In these families, the researchers uncovered 11,020 de novo mutations, the vast majority of which were germline mutations.
Paternal age, they said, accounts for some 95 percent of the variation seen in global mutation rate in the human population,
Further, replication timing of de novo mutations was associated with paternal age, the researchers found, as mutations popping up in the offspring of younger fathers were enriched in late-replicating genomic regions, while there was no such replication timing bias for mutations in the offspring of older fathers.
This finding, they noted, held even after accounting for maternal age.
Replication timing, Sunyaev, de Bakker, and their colleagues said, correlates with chromatin structure and gene activity. Regions of the genome that are early replicating have a greater density of genes and those genes are expressed at higher levels, as compared to late-replicating genomic regions.
This, they added, means that de novo mutations linked to paternal age are likely to have functional consequences.
The effect of paternal age on the number of de novo mutations is likely due to variations in replication timing or in mutagenesis between the symmetrical mitoses that occur during the formation of paternal germ cells and the asymmetrical mitoses of the spermatogonia during sperm formation, the researchers said.
No matter the paternal age, the researchers noted that the mutation rate was higher among functional regions of the genome as 1.22 percent of de novo mutations were exonic, a 28.7 percent enrichment. Mutation rates were also elevated in regulatory regions marked by DNAse I-hypersensitivity sites. For both of these, the researchers said that the elevated mutation rate appeared to be linked to CpG dinucleotides.
In addition, where de novo mutations were located in the genome appeared to be non-random, as they tended to form clusters. And those clusters harbored their own mutational patterns. Clustered mutations, for instance, contained fewer transitions and more C to G nucleotide shifts, which the researchers said suggested its own specific mechanism.
Sunyaev, de Bakker, and their colleagues noted that previously developed mutation models aren't representative of germline mutation rates as they are influenced by other evolutionary forces like background selection.
To develop a genome-wide mutation rate map, they took into consideration the human-chimpanzee divergence rates, as well as the flanking sequence context, local recombination rates, mutation type, and which strand is transcribed in coding regions. The researchers also calculated gene-level mutation rates and separately estimated the synonymous, missense, and nonsense mutations rates.
Such mutation rate maps, they said, could be applied to population and medical genetics studies to, for instance, examine evolutionary inferences on human mutation rates and help uncover disease-linked genes that harbor recurrent de novo mutations.