[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index]

7.2 Sample lm_auto parameter file

Here is an annotated parameter file `par15a' for lm_auto:

 
input pedigree file 'ped15'
input seed file 'seed15'

These statements are already familiar. The pedigree file given, `ped15', is the JV pedigree where the final individual is affected for a rare recessive trait (allele frequency 0.05). See Thompson (2000) Section 10.4.

The seed file contains one or more `set sampler seeds' statements to set the random seeds.

 
select all markers  traits 1

There are five markers and one trait locus. This is similar to the `simulate' statement (see genedrop computing requests).

 
map gender F  markers            recomb fracts .2 .2 .2 .2 
map gender M  markers            recomb fracts .1 .3 .1 .3 
map gender F  trait 1  marker 2  recomb fract  .1127
map gender M  trait 1  marker 2  recomb fract  .055

The `map' statements specify the marker map and trait position in terms of recombination fractions. In this example there are five markers with gender--specific maps. The trait locus is between markers 2 and 3, with recombination fractions given in terms of the marker to its left (2). See genedrop mapping model parameters.

 
set markers 1 2 3 4 5  freqs  .2 .2 .4 .2
set trait   1          freqs  .95 .05

The markers all have four alleles (labeled 1--4) with frequencies 0.2, 0.2, 0.4, and 0.2. The trait has a rare allele with frequency 0.05. See genedrop population model parameters.

 
set marker data 5
  333   1 3  1 3  1 3  1 3  1 3 
  331   3 4  3 4  3 4  3 4  3 4 
  334   2 3  2 3  2 3  2 3  2 3 
  431   3 4  3 4  3 4  3 4  3 4 
  531   3 3  3 3  3 3  3 3  3 3 

This sets the marker data for typed individuals (can also be given in a separate file specified in `input marker data file' statement). The number of markers, 5, follows `set marker data'.

The trait data are actually given in the pedigree file. Here trait number is 1, which defaults to the first integer after the "regular" items (names and gender).

The default trait type is genotypic. Trait data are given as coded genotypes: 1 = homozygous for the first allele (with respect to the order in the allele frequency statement); 3 = heterozygous; 4 = homozygous for the second allele; 0 = unobserved. Since the trait genotypes are not observed, they are postulated from the phenotypes by assuming a recessive model. In this pedigree file `ped15', the genotype of affected individual 531 is 4 (the second allele being the rare disease allele) while the others all are 0.

 
set proband gametes  531 1  531 0  331 0  333 1

The `set proband gametes ...' is the key statement of lm_auto. It specifies for which haplotypes to score ibd probabilities. After the statement keywords, a list of individual IDs and an indicator of paternal (1) or maternal (0) haplotype is given. For example, `531 1' means the paternal (1) haplotype of 531.

The example statement requests scoring both haplotypes of 531, the maternal(0) haplotypes of 331 and the paternal (1) haplotype of 333. See ibddrop statements, for the key statement `set proband gametes'.

 
set window patterns  0 4
set locus window   3

This example includes a pair of statements to jointly score ibd patterns for a moving window over adjacent loci. The first statement selects the ibd states to be scored, the second specifies the size of the moving window. This will make more sense when we look at the output from the example.

 
set L-sampler probability 0.2
set MC iterations 2000

This specifies 2000 sampling iterations, using the L-sampler 20 percent of the time (using the M-sampler otherwise). These iterations are preceded by burn-in iterations; lm_auto uses a default of 10 percent of the number of main iterations for burn-in. In practice, one would run the MCMC sampler much longer than 2000 iterations. This is just for demonstration.


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index]

This document was generated using texi2html