1. This project (Burke_2012) included 2 data sets of a 4-way cross between inbred strain BALB/cJ, C57BL/6J, C3H/HeJ and DBA/2J. The direction of the cross was (BALBxB6) F1 x (C3HxDBA/2) F1. The 1st one had 505 female mice, the 2nd one was merged from the initial population and a replicate population (n=832). They were saved in 2 separated folders ("Burke_2012" and "Burke_2012b"). This is the data set of "Burke_2012". The map of the data was updated with Build37 Cox genetic map (B37) and the new file was saved as "Burke2012_4way_B37_Data.csv". This file was ready to be read into Rqtl.


2. Generated a description file of the project, it included the general information, definitions of the phenotypes, description of the genotypes and information of the missing markers; 


3. For the genotypes of this cross: (cM=centimorgan, bp=basepair)
(1) The original data of "Burke_2012" didn't have detailed information of genetic map. It had both mouse SNP and DMIT markers. Most of the DMIT markers had 2 separated genotype columns to represent maternally-inherited and paternally-inherited alleles. The SNP markers, in contrast, only was used in one side of parental strains, and had one genotype column. I merged the DMIT markers and converted all the genotypes to the codes which can be read by Rqtl. Here is the details of the changes:
In original file: 1=BALB/cJ, 2=C57BL/6J, 3=C3H/HeJ, 4=DBA/2J, 0=genotype could not be scored
In B37 data file: 1=BALB|C3H, 2=B6|C3H, 3=BALB|DBA, 4=B6|DBA, 5=BALB, 6=B6, 7=C3H, 8=DBA, "-"=missing 
The markers with converted genotypes have been rearranged in B37 data by the chromosome ID and basepair positions; 

(2) Marker "D1Nds2" was unmapped in Cox genetic map. The bp position of the marker was assigned based on MGI, then converted it into cM through Map Converter tool (this type of markers was highlighted with light yellow color in description file); 

(3) Could not find any information of SNP marker "rs3023673" in Cox genetic map and MGI database directly, got its B37 basepair position from CGD SNP Database (http://cgd.jax.org/cgdsnpdb/) first, then converted it to cM through Mouse Map Converter tool (this type of markers was highlighted with light yellow color in description file);

(4) Could not find any information of marker "D6Mit198" in Cox genetic map and MGI database directly, received its B37 basepair position from author and converted it to cM through Mouse Map Converter tool (this type of markers was highlighted with lavender color in description file);

(5) Seven DMIT markers missed bp positions in Cox genetic map, assigned the positions from current MGI database. This type of markers were highlighted with gold color and please see description file for details; 

(6) Some markers were at the same position (cM) on chr 1,3,4,6,11,13,14 and 17 in B37 Cox map in this data;
 
(7) Generated genetic map plot, missing genotypes plot, map comparison plot and whole genome RF plot for B37 data. All quality control plots were saved in "QCReport_Burke2012.pdf". 


4. For the phenotypes of the cross:
(1) Changed the symbol of missing values from "" to "-" and saved the changes in B37 data file; 

(2) Added sex information (0 = female; 1=male) in B37 data file; 


5. Except the "csv" and "pdf" files mentioned above, I also saved the description file, original data, data with B37 map, data process and list of missing markers together as a big excel file ("Data_Description_4way_Burke2012.xlsx") (The 1st curating work was finished on 5/2/2012).