"GenotypeBayes.py":
This script is written in the Python language (http://www.python.org).  It is executable in Windows or UNIX if Python is installed.

This program uses an error rate table to change dye ratios into probabilities of genotypic state.  It then computes the Bayesian posterior probability of a genotypic state at each marker by assessing the probability of all possible combinations of genotypes around the marker of interest, based on the genetic map and prior probability scores.  The posterior probability calculation is not conducted for the end markers on a chromosome.  See the published manuscript for mathematical detail.

For Columbia/Landsberg samples, our prior error rates have been fully calculated for Cy5 labeled Columbia hybridization only.

There is no reason that the array can only be used with Columbia/Landsberg samples.  However, we have not extensively tested a set of prior error-prediction data for other accessions.  

Input Files:

"GenotypeBayesSettings.txt":
This is the settings file.  The filename nor the headers can be changed.  It must be in the directory of the python script.  Settings are:
Type of Data:  Which dye Columbia was labelled with, allowable values are 'Cy3','Cy5', or 'Both' for a dye swap
Input Data Directory:  Where the data input files are located.  All directories should be in the format c:\directory -- without a final \.
Output Data Directory:  Where the output should be placed.
Probability File Directory:  Where the training set probability files are stored.
Cy3 Input Data Filename:  All filenames must include the appropriate extension
Cy5 Input Data Filename--if only one dye is used, do not delete the other header in the settings file!
Output Data Filename
Columbia Cy3 Probability Table Filename
Columbia Cy5 Probability Table Filename--if only one dye is used, do not delete the other header in the settings file.!
Allowed error rates:  The script will call markers (as COL/LER) based on these false-positive cutoffs for prior and posterior probabilities.  	The values are two-tailed; i.e. a cutoff of 0.05 will lead to probabilities greater than 0.975 called as Col; less than 0.025, as Ler.

Cy3/Cy5 Input Data Files:
These files must be comma-delimited.  The first column contains marker names; the second contains the chromosome each marker resides on; and the third contains genetic positions from a telomere in cM.  The markers should be sorted in the order they occur on each chromosome.  Cells in the fourth column and after on the first row contain the sample names; for a dye swap experiment, the sample names must be the same and in the same order.  Data starts with the second row; any number of samples can be accommodated.  Missing data should be left blank (i.e. empty cells in Excel).

Probability Tables:
Two probability tables are required; these files must be comma-delimited.  One expresses the probability of a marker actually being Columbia if it has a particular ratio when Columbia is labeled with Cy5; the other is for Columbia labeled with Cy3.  The first row is ignored, as are columns after the third.  The first column is the lower bound for ratio ranges assigned to a particular probability; the second column is the upper bound; and the third column is the probability of the marker being Columbia associated with that ratio range.  Any number of ranges can be used.