Send to

Choose Destination
See comment in PubMed Commons below
Genet Epidemiol. 2006 Jul;30(5):447-58.

Distribution and magnitude of type I error of model-based multipoint lod scores: implications for multipoint mod scores.

Author information

Department of Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, Ohio 44106, USA.


The multipoint lod score and mod score methods have been advocated for their superior power in detecting linkage. However, little has been done to determine the distribution of multipoint lod scores or to examine the properties of mod scores. In this paper we study the distribution of multipoint lod scores both analytically and by simulation. We also study by simulation the distribution of maximum multipoint lod scores when maximized over different penetrance models. The multipoint lod score is approximately normally distributed with mean and variance that depend on marker informativity, marker density, specified genetic model, number of pedigrees, pedigree structure, and pattern of affection status. When the multipoint lod scores are maximized over a set of assumed penetrances models, an excess of false positive indications of linkage appear under dominant analysis models with low penetrances and under recessive analysis models with high penetrances. Therefore, caution should be taken in interpreting results when employing multipoint lod score and mod score approaches, in particular when inferring the level of linkage significance and the mode of inheritance of a trait.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Loading ...
    Support Center