Thursday, July 21, 2011

Changing partners in life: recombination biology

Two papers have appeared this week about the nature of recombination in humans, one in Nature Genetics, by Wegmann et al. ("Recombination rates in admixed individuals identified by ancestry-based inference"), and the other in Nature by Hinch et al. ("The landscape of recombination in African Americans").  Do they add anything new to what we know about recombination or human evolution?

When sperm or egg cells are about to be produced, all our 23 pairs of chromosomes line up against their corresponding partners in the center of the cells, before being pulled apart into the separate daughter cells that will eventually be the sperm or egg cell, in which one copy of each will be present.

One might expect that a sperm or egg cell would have either the copy of, say chromosome 3, that the parent got from his or her parent, and is thus transmitted intact to the future grandchild.  But this is not what happens.  Instead, the lined-up 'homologous' chromosomes stick to each other here and there, and swap pieces.  This is called recombination and means that the grandchild receives a mix of pieces of each grandparent's chromosomes.

Recombination is treated generally as a random process: the chromosomes stick to each other in random places along their length.  But this is inaccurate.  The sticking points happen more frequently in some places than in others.  This differs from chromosome to chromosome and even to some extent between males and females.  Why is that so?

One reason is that recombination is effected by active mechanisms that involve proteins (that is, that are coded for by genes somewhere in the genome) and the DNA sequences that these proteins stick to in bringing recombination about.  Variation in those binding sequences can affect the frequency with which recombination will occur.  The results show up statistically as relative hot-spots of recombination.

Recombination generates new variation (new combinations of genetic variants) as a resource with effects on evolution, and this is because these combinations can have effects on the organism's traits.

These two papers are companion pieces.  As Wegmann et al. report,

By comparing the AfAdm [African admixture] map to existing maps, we were able to make several observations: (i) there is evidence for subtle population differences in recombination rates between African and European populations, (ii) African-European admixed individuals appear to have recombination rates that are, on average, intermediate between the African and European rates, and (iii) the degree to which the rates are intermediate is predictable from the average ancestry coefficient (~80% African and ~20% European) in our sample. Further, in admixed individuals, recombinations appear to be concentrated at hotspots in a manner correlated with ancestry: individuals with more African ancestry have recombinations at hotspots found in the HapMapYRI map, and individuals with more European ancestry have recombinations at hotspots found in the HapMapCEU map. These observations are consistent with the differentiation between populations for fine-scale recombination rates1–5 and with the European-African differentiation at PRDM9, the only known major locus affecting fine-scale recombination rates.

Whether this is important, or worth whatever its costs were to discover, is open to question.  Could the same not already be, or have been, known from work on other species where it would be much less costly?  And if it has effects on disease--always the self-interested rationale of investigators by which such stories are touted in the news--such effects would be detectable by the current methods (such as GWAS) that hunt for them.

So these papers provide what might be useful and important information about a basic aspect of cell biology and the generation of variation.  The value will prove out in the future, but certainly our repertoire of basic knowledge about variation, and in this case human variation, is increased thereby.

No comments: