This script shows how to use makeped. I am using the same pedigree structure as on pages 14-15 of Terwilliger-Ott ("TO"), but there are some differences that I wish to highlight. Here is the file pedin.pre that I entered:
ex1 father 0 0 1 2 1 3 1 2 ex1 mother 0 0 2 1 2 3 1 1 ex1 dau1 father mother 2 1 2 3 1 2 ex1 dau2 father mother 2 2 2 3 1 2 ex1 son1 father mother 1 2 1 2 1 2 ex1 dau3 father mother 2 1 3 3 1 1 ex1 son2 father mother 1 1 3 3 1 1
There is one row per person. The first column identifies the pedigree. The second column is an identifier for the person. The third column is the identifier for the person's father. The fourth column is the identifier for the person's mother. The fifth column is gender: 1 is male, 2 is female. The remaining columns describe the genotypes.
Locus 1 gets only 1 column for affected/unaffected because there is only 1 liability class. The first number (column 6) is the affection status: 1 is unaffected, 2 is affected. Multiple liability classes are used to give different people different penetrances (e.g., because of age or gender).
Columns 7 and 8 are alleles for locus 2.
Columns 9 and 10 are the alleles for locus 3, using the same values as for
locus 2 in TO.
Use 0 for an individual whose genotype is unknown at a locus. The LINKAGE convention is that either both alleles are known or both are unknown. In those cases where one allele can be inferred, UNKNOWN is usually capable of the inference.
Enter makeped
and get the prompt:
MAKEPED Version 2.1
Pedigree file ->
Fill in pedin.pre to the right of the arrow and hit return.
This produces:
Pedigree file -> pedin.pre Output file ->
The conventional output file name is pedin.dat, so I enter that and get:
Does your pedigree file contain any loops? (y/n) ->
Loops are an advanced topic (TO, Chapter 7).
Enter n and get the following:
Do you want probands selected automaticaly? (y/n) ->
[sic]
Enter y and the program exits.
Proband issues are confusing.
First, in this context, proband means:
"who is the last person updated as the pedigree is traversed?" (see
the file traverse.ps in FASTLINK documentation ("FD")).
This is completely different from the usual meaning of "the first diseased
person in the pedigree to be identified".
Until you understand more about proband selection, follow these two rules:
Rule 1. Always let makeped select the proband for you
Rule 2. Put the patriarch and matriarch of the family (or one of them
if they are not unique) as the first two people in pedin.pre.
Here is what the output file pedin.dat looks like:
1 1 0 0 3 0 0 1 1 2 1 3 1 2 Ped: ex1 Per: father 1 2 0 0 3 0 0 2 0 1 2 3 1 1 Ped: ex1 Per: mother 1 3 1 2 0 4 4 2 0 1 2 3 1 2 Ped: ex1 Per: dau1 1 4 1 2 0 5 5 2 0 2 2 3 1 2 Ped: ex1 Per: dau2 1 5 1 2 0 6 6 1 0 2 1 2 1 2 Ped: ex1 Per: son1 1 6 1 2 0 7 7 2 0 1 3 3 1 1 Ped: ex1 Per: dau3 1 7 1 2 0 0 0 1 0 1 3 3 1 1 Ped: ex1 Per: son2
Notice that makeped assigns numbers to everyone. People like to use names, but computer programs prefer to use numbers. That's why.
Column 1: pedigree number
Column 2: person number
Column 3: number of father
Column 4: number of mother
Column 5: first child
Column 6: next (half)sibling with same father
Column 7: next (half)sibling with same mother
Column 8: gender
Column 9: Proband has a 1; others have 0.
Column 9 is very important for loops.
See: FD: loops.ps
After that follow the genotypes and some annotation.