In today's practical we will be analysing a data set that was generated as part of a genome-wide screen for Multiple Sclerosis For the first analysis we will be using a data set that consists of 37 nuclear families, each with a single affected sib pair. The parents are typed and fully informative at the locus of interest (HLA-DQ). The pedigree data is in the linkage-format pedigree file MS-ped.txt and the associated locus data file is MS-loc.txt. The locus data file will include some parameters for an assumed disease model that will be ignored when we perform non-parametric (model-free) analysis in Genehunter (but would be used if we were to perform parametric (model-based) analysis in Genehunter)
1. Save the data files above in the appropriate folder where you saved the Genehunter program.
2. Open up an MSDOS window (Click on Start, Run, then type cmd ). Once the window has opened, type dir to see all the files and directories
(folders) that are in your home space, and move into the directory where you saved the data files e.g. by typing
cd xxxxx
(where xxxxx is replaced by the name of the appropriate folder).
Type dir again to check the required files are available in the directory.
3. Start up Genehunter by typing gh
4. Read in the locus data by typing
load markers MS-loc.txt
3. Set analysis to non-parametric (model-free)
by typing analysis npl
4. Read in the pedigree data and analyse each family individually
by typing scan pedigrees MS-ped.txt
5. Total results for all families by typing total stat
You should see the following four numbers on the screen
Column 1: position in cM of locus being analysed
Column 2: NPL score (here equal to 1.627)
Column 3: p value estimated by Genehunter itself (here equal to 0.04)
Column 4: `information' which is an estimate of the available
IBD information as a proportion of the maximum possible (here equal to 1.0 as the parents are all fully informative)
6. To perform the likelihood ratio tests with `possible triangle' restrictions, type
estimate followed by (when the program asks you) n . Suggested output name is mls-out.txt
7. Output the IBD sharing estimates for each ASP by typing dump ibd .
Suggested output file name is ibd.txt .
8. Exit Genehunter by typing q . Examine the output files
which should now be in your directory. The file mls-out.txt should give estimates of
the z0, z1, z2 IBD sharing parameters and the MLS test statistic
(called `loglike'), which here equals 0.738, coresponding to a p value of approximately 0.05.
The data set here consists of 37 sib pairs, of which 4 share 0 alleles IBD, 22 share 1 allele IBD and 11 share 2 alleles IBD. Does this seem to be consistent with the output file ibd.txt? (Hint: you may need to open the file up in excel, and take a look at the prior and posterior IBD sharing probabilities for the different type of relative pair in your data set. The posterior IBD sharing probabilities are given in the last 3 columns, and the prior IBD sharing probabilities are given in the 3 columns before that).
For the second analysis we shall be analysing the
full HLA data from the MS genome screen. This consists
of nuclear families of different sizes (mainly ASPs
and trios) typed at 20 markers in the HLA region.
Parents may or may not be fully informative: they may be typed but uninformative,
they may be untyped, or we may be performing multipoint inference between markers
(in which case parents cannot possibly be fully informative since
they are not typed in between markers!). Pedigree data is in the file
MSfull-ped.txt
and the locus data file is
MSfull-loc.txt
1. Save the data files above in the appropriate folder where you saved the Genehunter program.
2. Open up an MSDOS window (Click on Start, Run, then type cmd ). Once the window has opened, type dir to see all the files and directories
(folders) that are in your home space, and move into the directory where you saved the data files e.g. by typing
cd xxxxx
(where xxxxx is replaced by the name of the appropriate folder).
Type dir again to check the required files are available in the directory.
3. Start up Genehunter by typing gh
4. Read in the locus data by typing
load markers MSfull-loc.txt
3. Set analysis to non-parametric (model-free)
by typing analysis npl
4. Read in the pedigree data and analyse each family individually
by typing scan pedigrees MSfull-ped.txt
5. Keep a log of results by typing
postscript on
photo multipoint-results.txt
6. Total results for all families by typing total stat . Accept the suggested output file names.
6. For the likelihood-based MLS analysis, select for analysis
all ASPs (unweighted) by typing pairs used followed
by (when the list of options appears) 4 .
7. Aanalyse by typing estimate followed by (when the program asks you) n . Suggested output file names are mlsfull-out.txt, sharefull.ps, mlsfull.ps.
8. Exit Genehunter by typing q . Examine the output
files which should now be in your directory. (Postscript files of the multipoint plots
can be viewed by double clicking).
You should find
that the maximum NPL
score is 3.44 at 32 cM. You should find that
the maximum MLS is 2.83 at 28 cM.
Genehunter documentation is available here: ghdocumentation.ps