Isaac Newton Institute for Mathematical Sciences

XXVII R.A. Fisher Memorial Lecture

1953: An unrecognized summit in human genetic linkage analysis

4th December 2006 at 17:00

Author: Professor E Thompson (Washington)

Abstract

The foundations of linkage analysis in humans were laid in the 1930s by R. A. Fisher and J.B.S. Haldane, and by the end of that decade many of the issues had been raised. However, the statistical methodology and the statistical properties of procedures remained to be explored. In 1953, at a meeting of the Royal Statistical Society, C.A.B Smith presented a paper which has been much underestimated by the human genetic linkage community. Building on earlier work with Haldane and the log-odds (lods) of Barnard (1949), Smith introduced the lod score to human genetic linkage analysis. Not only was the lod score to be the key tool for human gene mapping of simple genetic traits, but in 1994 a review paper on mapping complex traits described four approaches: lod scores, association-based methods, identity-by-descent methods, and QTL mapping in experimental crosses. Each of these is not only discussed in Smith (1953), but he presents a coherent framework which covers them all.

In computation and informatics on the one hand, and in our ability to sequence DNA and measure gene expression on the other, much has changed since 1953. However, many of those who have proposed "novel'' approaches to the analysis of human genetic data may be surprised to learn that their ideas were suggested over 50 years ago in the RSS read paper of C. A. B. Smith.