here is a rather fitting citation link between the papers at #6 and #7 in the latest list of Biology Hot Papers. Reich et al.’s first citation is to Sachidanandam et al., thus cementing the relationship between the raw data and one of its most useful applications to date.

The International SNP Map Working Group published in the same genome edition of Nature as the public genome sequence that still sits at #1. The SNP map is essentially an index to the book of life, a collection of 1.42 million markers that researchers can use, among other things, to guide them to the genes they are interested in. SNPs are single nucleotide polymorphisms; that is, places on the DNA where two individuals differ by a single, known letter of the genetic code.

Unlike the full sequence data, the SNP map has taken much longer to become a Hot Paper. Team leader David Bentley, of the Sanger Centre in the U.K., tells Science Watch that he was neither surprised nor disappointed by this. "The sequence papers … contained extensive analysis and new biological information of value to a wide variety of biologists," Bentley says. "By contrast, the SNP paper … is of benefit to a more-focused group and covers a much newer area of human genetics."

In fact, the paper at #6 is from one of the first of those more-focused groups to publish. David Reich and Eric Lander, of the Whitehead Institute/MIT Center for Genome Research, lead a group that used the SNP map not to home in on genes but to look at a larger-scale phenomenon, linkage disequilibrium. Mendel’s famous laws of inheritance state that characters are inherited independently, that the chances of having one gene (more strictly, one version of a gene) are not influenced by the presence of another gene. In reality genes are often linked—that is, inherited together more often than expected.

Individuals often inherit rather long chunks of DNA from one parent or the other. The chunk is known as a haplotype, and some haplotypes themselves may also be inherited as a group. This is called linkage disequilibrium (LD), and it is of huge theoretical and practical importance. Theoretical because molecular biologists want to know how large the blocks that are inherited together might be. Practical because haplotypes are often associated with reasonably common diseases that have complex genetic origins. The easier it is to identify particular haplotypes, the easier it might be to link them to specific diseases.

The value of the SNP map is that it provides a vast set of markers that are much easier to use than the actual genes that Mendel postulated. Lander’s group used the SNP map to look in detail at the DNA surrounding a set of 19 randomly chosen regions of the genome, comparing the sequence around the anchor points in several human populations. They detected significant LD on average much farther from the anchor points than had been theorized. And stretches of DNA inherited as a block tended to be longer than had been expected.

The study suggests new ways to look for genetic diseases (a promise already fulfilled by the discovery of the genetic basis of resistance to malaria—see P.C. Sabeti, et al., " Detecting recent positive selection in the human genome from haplotype structure," Nature, 419[6909]: 832-7, 24 October 2002) and offers an insight into human evolutionary history. Haplotypes tend to be established by factors that affect the breeding population. A bottleneck, for example, will favor one haplotype at the expense of others. Recombination then breaks apart the haplotype. So the extent of LD around the anchor is a measure of how many generations have taken place since the event that created the haplotype. The team compared the LD patterns of people from northern Europe and Yoruba people from Nigeria. Close to the markers, the LD pattern was very similar. This reflects the common ancestry of the two groups more than 100,000 years ago. Farther from the markers, however, the Yorubans exhibit much lower LD than the Europeans. The implication is that something happened to the ancestors of northern Europeans after they had split from the ancestors of Yorubans.

That event could be the recolonization of Europe after the previous Ice Age by a small band of founders. Or it could have occurred much earlier, a severe bottleneck during the founding of the European population or during the emigration of modern people out of Africa. Either way, the LD data allow an estimation of the extent of the bottleneck. The effective breeding population, by this estimate, could have been as small as 50 individuals for 20 generations, 1,000 individuals for 400 generations, or any other combination with the same ratio.