Department of Bioinformatics and Computational Biology

Home > Public Software > Archive > Bayes Mix > Bayes Mix User's Guide

This project is archived and no longer maintained.

Bayes Mix User Guide

hidden rowfor table layout
Overview
DescriptionA tool for performing model-based inference for differential gene expression, using a non-parametric Bayesian mixture probability model for the distribution of gene intensities under different conditions
Development Information
LanguageR and C
Current version0.8.8
PlatformsWindows and Linux
StatusInactive
Last updated15th August 2003

Bayes Mix User’s Guide

BayesMix is a model-based inference for differential gene expression, using a non-parametric Bayesian probability model for the distribution of gene intensities under different conditions, similar to the Empirical Bayes approach.

The BayesMix distribution consists of a Java front-end for an R back-end which calls various C functions implementing the statistical method. A pseudoterminal application qua agent accepts commands from the Java front-end and returns output from R.

User Interface

The Java application processes matrix file data. The information input by the user is checked for validity. Invalid input will cause the offending field to be displayed in reverse video. ToolTips with numeric range information are provided. The display button will be enabled when all fields are considered valid.

Differential Matrix

Specifies the filename of the differential matrix. The matrix file must have its fields separated by tabs. The first line of the file [header] contains the names of the variables (columns), followed by some number of lines containing the data to be analyzed. Each data line begins with a row name, followed by data values. The header line should contain one less column than a data line. It is recommended that all header fields be quoted to avoid ambiguity.

Null Difference Matrix

Specifies the filename of the null difference matrix. The format of the matrix is the same as that of the differential matrix.

False Discovery Rate

Specifies a list of real numbers representing false discovery rates. There can be a minumum of four and a maximum of ten values. Each value must be within the range [0.0, 1.0).

Analysis Mode

Specifies the desired processing to perform. Choices are:

SD Correction Factor

Specifies the calibration factor(s) for standard deviation correction. There can be either one or two real number values. When two values are provided, the first is used with the differential matrix and the second with the null difference matrix. If only one value is provided, it is used for both matrices. NPBA analysis mode only.

Simulations

Specifies a positive integer value representing the number of MCMC simulations to perform. NPBA or Sequential analysis mode only.

MCMC Burn-In

Specifies a non-negative integer value representing the number of MCMC simulations to discard as part of the burn-in. This value can be at most one half of the number of simulations. NPBA or Sequential analysis mode only.

MCMC Interval

Specifies a non-negative integer value representing the MCMC sample rate. As such, higher values will cause less samples to be gathered. At least ten samples are required to get meaningful output. NPBA or Sequential analysis mode only.

Back to top

File Pulldown Menu

The File pulldown menu contains all the generic file handling options.

Help Pulldown Menu

The Help pulldown menu contains all the options providing basic assistance in using the application.

Back to top

Input/Output

Environment variables

Files

Command line activation

java

Back to top