Meta-analysis of linkage studies

Lander and Kruglyak [1995] gave guidelines for interpreting linkage results based on estimating how often a particular threshold for significance woul d be exceeded by chance in a single genome scan. What is unknown is how oft en two or more genome scans would exceed a particular threshold within the same region. We develop theoretical estimates of these values and compare t hese with the empirical estimates derived from the GAW11 data. For single-p oint analysis, the theoretical estimates predict the empirical estimates. F or multipoint analysis, the theoretical values overestimate what is observe d. For both single point and multipoint, modest p-values within a single ge nome scan may give highly significant results when replicated in the same r egion in other scans. (C) 1999 Wiley-Liss, Inc.