Superlink-GM


Introduction
Germline Mosaicism (GM) is a genetic condition in which some germ cells of an individual carry a mutation. Individuals with germline mosaicism may give birth to affected children and unaffected children that carry the same measured markers in the vicinity of the trait locus. This condition violates some of the assumptions underlying standard genetic linkage analysis. Consequently, there is relatively low power to detect genes created by GM using standard genetic linkage analysis.

Superlink-GM is a software that performs genetic linkage analysis in the presence of GM. The software extends the standard model used for genetic linkage analysis in order to incorporate GM.


Description
Superlink-GM is composed of three parts.
  • csvToGMPed.py is a Python 2.7 script that accepts an input .csv file representing pedigree data, and produces a .xmlbif file representing the same data in an XMLBIF format.
  • BayesNetGeneration is a linux32 program that performs various Bayesian Networks optimization, taken mostly from Allen and Darwiche (2008). This software is used internally by csvToGMPed.py
  • BayesNetPlus is a linux32 program that computes the likelihood of the evidence in a bayesian network represented by a .xmlbif file.
Note that the software runs on linux32 platforms only.

Usage Instructions
In order to use the software one must follow the following steps:

1. Create a .csv file with the pedigree data. An example .csv file is available in the package. A description of the file format is provided in the readme file.

2. Convert the .csv file to a .xmlbif file using the Python script csvToGMPed.py. .

3. Compute the likelihood of evidence in the .xmlfile file with the program BayesNetPlus.bin.

Please refer to the readme file for detailed instructions


Code and Executables
  • csvToGMPed.tar.gz - this package contains the Python script csvToGMPed, as well as the program BayesNetGeneration.
  • BayesNetPlus.tar.gz - this package contains files needed for the program BayesNetPlus.
  • The Apache Xerces Project - This software is internally used by BayesNetPlus. It should be installed in the system path or in the same directory as BayesNetPlus.

Authors
BayesNetPlus was developed by Mark Silberstein. It uses the program Orderfinder, developed by Nickolay Dovgolevsky and Maayan Fishelson.

BayesNetGeneration was developed by Andrei Anisenia and is based on algorithms developed by Maayan Fishelson and Allen and Darwiche (2008).

The script csvToGMPed was written by Omer Weissbrod under the supervision of Dan Geiger.


Contact Information
Any comments or bug reports should be directed to Omer Weissbrod.

Literature
Allen, D. and Darwiche, A. (2008). "RC Link: Genetic linkage analysis using Bayesian networks." International journal of approximate reasoning, 48(2), 499.525.
Weissbrod, Omer and Geiger, Dan (2011): "Genetic Linkage Analysis in the Presence of Germline Mosaicism," Statistical Applications in Genetics and Molecular Biology: Vol. 10: Iss. 1, Article 46.