Instructions for running models described In:

Devlin, B., Jones, B.L., Bacanu, S., Roeder, K., "Mixture Models for Linkage Analysis of Affected Sibling Pairs and Covariates," Genetic Epidemiology, 2002, 22(1):52-65.

Devlin, B., Jones, B.L., Bacanu, S., Roeder, K., "Reply to Olson," Genetic Epidemiology, 2002, 23(4):449-455.



CovIBD refines linkage analysis of affected sibpairs by considering attributes or environmental exposures thought to affect disease liability. This refinement utilizes a mixture model in which a disease mutation segregates in only a fraction of the sibships, with the rest of the sibships unlinked. Covariate information is used to predict membership within the two groups corresponding to the linked and unlinked sibships. The pre-clustering model uses covariate information to first form two probabilistic clusters and then tests for excess IBD-sharing in the clusters. The Cov-IBD model determines probabilistic group membership by joint consideration of covariate and IBD values.

The main program (covibd) is implemented in the R programming language (freeware version of S-Plus) for Linux (version 1.7.0). Thus, the first step in running the program will be to download a copy of R if you do not already have it installed. To download R, go to

and follow the necessary links to download R for Linux.

Once R is installed, simply type R at the command prompt to start. To quit R, type q() at the R command prompt. To cancel an R command, type control-c.

Download program

General Instructions


Example 1

Example using CovIBD model.

Driver file

Data file

Covariate file

Output file

Covariate output file


Example 2

Example using Preclustering Model (clustering input by user).

Driver file

Data file

Covariate file (inclusion probabilities)

Output file

Click here for recommended steps to performing a cluster analysis in R.