New psoriasis gene would have been missed by testing linkage only

A recent study published in nature genetics study highlights a ß-defensin gene where increased copy numbers are related to psoriasis. The normal range is 2 – 12 copies with the risk increasing with each extra copy. I don´t know if the given explanation is really correct: more genomic copies lead to more RNA, lead to more antimicrobial peptide, lead to skin inflammation. Maybe there is an unknown intermediate step: more antimicrobial peptide -> induce other inflammtory interleukins -> skin inflammation.
In any case, I agree with the editor that this is a careful done study. Most remarkable, however, is another fact – this gene would have been missed by linkage. Linkage studies have been the mainstay of complex disease genetics for the last decade while the authors now find

We have identified a psoriasis susceptibility locus not previously identified by linkage analysis … We simulated pedigrees … Only a minority of simulated studies involving 500 affected sibling pairs detect linkage at even ‘suggestive’ levels of significance (P < 0.05) diploid copy number may not have a simple relationship with inheritance of parental alleles ...undetectable by even large linkage studies and may only be discovered by candidate-gene association studies.

Linkage is something different from association – it tests if recombination between two marker (a physical marker and a disease locus) is different from 0.5 – the expected recombination fraction of two loci segregating independently according to Mendel’s second law. The conclusion in the new paper is in full agreement what we already found in 2001. My doubt was not so much about Mendel’s second law but the recombination differences being really detectable.

It may be that high-density SNP association analysis in combination with functional genomic data may prove to be necessary to detect susceptibility loci (which may be of small effect) for many complex human diseases (Risch 2000). In the meantime, like true love, true linkage remains hard to find.

With “high-density SNP association analysis” we thought of SNPs prevalent in the particular population (which is not the case with the recent GWA panels). Furthermore we wanted functional evidence (not provided by any GWA). So, in contrast to some more enthusiastic comments about the recent progress in human genetics, I would repeat that true association remains hard to find.