2005-01-26
Features of DMLE+

The DMLE+ program allows multipoint LD mapping using an arbitrary number of SNPs or microsatellite markers. DMLE+ implements Markov chain Monte Carlo (MCMC) methods to allow Bayesian estimation of the posterior probability density of the position of a disease mutation relative to a set of markers.

Current Release (DMLE+ 2.2)

• multipoint LD mapping using an arbitrary number of markers
• integrates over uncertain genealogy of disease chromosomes
• estimates Bayesian posterior probabilities of mutation location, disease allele age, and other parameters
• uses either haplotype or genotype data (under simple models of disease inheritance)
• allows for multiple marker alleles
• integrates over uncertain allele frequencies in the normal (control) population
• incorporates information from an annotated human genome sequence
• Windows GUI and portable (UNIX) command line versions available
• accomodates either linkage equilibrium or linkage disequilibrium in sample of controls*
• allows slice sampling, uniform selection, or sliding window mechanisms for modifying mutation location in MCMC*
• mutation age and branch length parameters adjusted separately in MCMC to improve mixing*
• faster algorithms implemented for sampling topologies in MCMC*
• linux version compiled with ICPC, rather than gcc, resulting in substantially faster execution*

Next Release (2.3)(late 2005)

• uses either haplotype or genotype data (under complex models of disease inheritance)
• allows for locus and allelic heterogeneity
• allows for marker mutation under simple models (microsatellites)
• takes account of the uncertain relationship between physical and linkage maps