Statistical Programming Core

Contact: Michael Schmidt, PhD mschmidt@med.miami.edu

The Section of Statistical Genetics includes faculty members with research focusing on statistical methods in human genetics studies. Areas of interest include:

• Methods development for disease gene mapping


• Expression and functional analysis methods


• Population genetics methods as they relate to understanding genetic variation in human disease


• Large-scale statistical analysis methods

The Statistical Programming Core interacts with the Genetic Epidemiology Core and the Informatics Core of the MIHG. The Core is responsible for computational aspects related to statistical genetic analysis and methods development. Dr. Mike Schmidt is the Core Leader. He has worked for several years programming genetic algorithms and conducting large simulation studies. As research programs develop in statistical genetics and genetic analysis, the Core includes programmers funded on research proposals to conduct simulation studies and develop statistical software.

The faculty and analysis group are actively involved in research to study existing statistical methods and develop novel methods for gene identification. The following are our in-house developed software and methods, which are maintained by the Statistical Programming Core and made freely accessible to the genetic research community. Click Here for software download page.

SIMLA Simulation Software
SIMLA is a SIMulation program that generates data sets of families for use in linkage and association studies. SIMLA_3.1 is a major upgrade to versions 2.3 and 3.0 that provides the ability to simulate two disease loci and two environmental covariates. Gene-gene and gene-environment interactions may also be simulated which jointly determine the disease risk of all pedigree members.
The SIBLINK program performs linkage analysis on affected sib-pairs.

PDT Analysis Program v 5.1
The Pedigree Disequilibrium Test (PDT) analysis program allows the user to test for linkage and association in general pedigree data. In addition to allele- and genotype-specific analysis of individual markers, PDT version 5.1 adds the ability to perform genotype-specific analysis over marker sets (Solaris, Linux, Windows). The extension versions allow for genotypic association analysis (geno-PDT) and multi-locus effect estimate (multi-locus geno-PDT)

Association in the Presence of Linkage (APL)
APL is a C++ program that provides a novel test for association in the presence of linkage using general pedigree data. The APL provides both single-locus and haplotype analysis. Recently a function for analysis of X-linked markers was added.

Extended Multifactor Dimentionality Reduction (EMDR) , MDR-PDT and MDR-phenomics
EMDR and MDR-PDT are extensions of MDR, which is a model-free non-parametric method for identifying and characterizing gene-gene effect influencing complex diseases. These methods are more powerful for detecting high order gene-gene or gene-environment interactions in both family-based and case-control studies than traditional parametric methods.

MDR-Phenomics was developed to test for high-order interaction with the consideration of genetic heterogeneity.  It inherits the advantages from MDR that no assumptions of genetic models are required and that high dimensionality can be processed by reductions. It also improves the power by integration of phenotypic covariate analysis.