Werner's Syndrome Data: 2006 version

This year (2006) the Werner's Syndrome example and data files have all been converted to use Merlin rather than Genehunter. The Genehunter files are still on the system, and can be found through this link, but Merlin is a better supported and more useful software to get to know.
One disadvantage of Merlin is that it needs more input files!

A little background about Werner's Syndrome as well as some references can be found here.

ApoB on Chromosome 2

ApoB is on chromosome 2 and is unlinked to the Werner's syndrome locus which is known to be on chromosome 8p. Some alleles for ApoB are quite different in frequency in the Japanese compared to frequencies in caucasians. What is interesting is that if caucasian frequencies are used, spurious evidence about linkage can result from the analysis (although a lod score of 3 is not achieved, an "interesting" lod score of over 2 is obtained, while if Japanese frequencies are used, the maximum lod score is under 1).

Homozygosity mapping

We could do this "by hand", using the details of homozygosity mapping from the lecture notes. (In a few past years, this has been a homework exercise.) The marker data together with ApoB allele frequencies from different control populations are given in homoApob.dat.

These data are reproduced in homoApob.dat, just because it is easier to see than in the big data set -- you do NOT need this file for running either Genehunter or Merlin.

Linkage analysis uising Merlin

We can use Merlin to calculate lod score based on the pedigree data with the single marker Apob and see what's the influence of using different allele frequency estimates.

Note that, since WS is a lethal disease, we do know that the ancestors are unaffected: this is not taken into account in the simple homozygosity mapping formulae. For a rare disease it has little impact, but for completeness we should use that information.

1. Data with 19 Japanese families: apob3_merlin.ped.

   The fields are:

   1. Family Id
   2. Personal ID
   3. Father ID
   4. Mother ID
   5. Sex: 1 = Male, 2 = Female
   6. Disease Status: 1 = Normal, 2 = Diseased, 0 = Unknown
   7. Next two fields are Apob genotypes, X = Unknown, 1-9 allele codes
   8. Next two fields are a dummy marker locus, X = Unknown,-- everyone is "X X".
      I included this dummy marker becasue it's the simplest way to get Merlin to run with just one real marker.


2. Merlin also requires several other files:
   • The .dat file: apob3_merlin.dat
     This file gives information on the fields of the pedigree file, after the ID's and gender. It says we have a discrete trait (code "A") which we call ws-disease, followed by two markers (code "M") with the names we have given them.
   • the .model file : ws_merlin.model
     This file specifies the model for the disease trait ws-disease (we can use this file for both the APOOB3 data example, and for the Chromosome 8 multipoint mapping below. It gives a frequency for the disease allele (0.005) and three penetrances -- the probabilities of being affected is one has two normal alleles (0.0), is a carrier (0.0), or has two diseasea alleles (1.0). That is, we are here specifying a simple recessive disease. (Note the penetrances must be separated by commas, with no spaces -- I learned that the hard way.)
   • The .map file: apob3_merlin.map
     This file specified the marker map, by giving the positions, measured in centiMorgans (cM). Here we have just one real marker, and a dummy, so I arbitrarily put the APOB3 marker at position 100.0, and the dummy at some big distance away (at position 300.00). The APOB3 marker is in fact on Chromosome 2, so I gave it that, although it does not matter for this example.

3. Merlin marker allele frequency files
   Merlin uses a separate file for marker alleles frequencies. Here we have 3 such files. Each marker has two lines:
   M (for marker) followed by the marker name
   F (for frequency) folloed by a list of allele freqs for alleles 1,2,...,
   (we do not need to tell it how many alleles there are)
   • Standard Caucasian: merlin frequency file
   • UW Caucasian controls: merlin frequency file
   • UW Japanese controls: merlin frequency file

Chromosome 8 Markers

Multipoint linkage analysis with Merlin

Some of the affected individuals in those Japanese families had data on 35 markers close to WRN on Chromosome 35. Some of the markers are very close together and not informative for linkage analysis, so we are only going to use a subset (13) of the markers. Again we have allele frequencies from both Japanese and Caucasian controls.

• Pedigree data for linkage analysis with Merlin: ws_merlin.ped.
  The format is as before, but we now have 13 markers instead of 2.
• Other Merlin files for this analysis: the data specification file ws_merlin.dat
  , the model file ws_merlin.model, (same as above), and the marker map file ws_merlin.map
  Note the data spcification format and map file format are as described above: the only difference is we now have 13 markers.
• Merlin frequency files with two sets of allele frequencies: Japanese: ws_jp.freq. Caucasians: ws_ca.freq.
  The format is as before: we now have 13 markers.