## Fitting imprinting effects

!------------------------------------------------------------------------------ ! Program GIN4IMPRINT ! ! Auxiliary program for WOMBAT to set up inverse of gametic relationship matrix ! ! Files written out allow imprinting effects to be fitted with a GIN covariance ! structure: ! 1. The .gin file -> default name: imprint.gin ! 2. The corresponding .codes file -> default name: imprint.codes ! Gametes are coded as: animal ID * 10 + 1 (paternal) and ! animal ID * 10 + 2 (maternal) ! => need to add corresponding column to data file! ! Input required: ! Pedigree file (after pruning) with animal, sire and dam ID recoded in ! running order and animals' inbreeding coefficients; ! this can easily be generated by WOMBAT by running a set-up step for a ! corresponding model without the imprinting effect(s) fitted ! ! Code for gametic relationship matrix written by Bruce Tier !-------------------------------------------------------------km----02/2011-----

WOMBAT readily allows for models fitting imprinting effects. These are best fitted as gametic effects, setting up the inverse of the gametic relationship matrix externally and supplying it to the analysis as `.gin`

file.

To make this task simpler, a FORTRAN program – written by **Bruce Tier** – is available:

- Source code: Gin4Imprint.f90
- Source code & compiled executables (Linux and Windows 64-bit):Gin4Imprint.zip

Details on its use are given as comments at the beginning of the source code and in the box on the right; an example analysis can be found in Example4/D. In brief:

- Run the `Set-up' step in WOMBAT for a simple animal model, i.e. NOT fitting imprinting effects. This will generate the file:
`PrunedPedFile1.dat`

This file contains the pedigree information for animals in the analysis (after pruning), both as original codes and recoded from 1 to number of animals, together with their inbreeding coefficients. - Run
`Gin4Imprint`

. This will use the pedigree list and inbreeding coefficients from`PrunedPedFile1.dat`

to:- set up the inverse of the gametic relationship matrix and write it out in the format required by WOMBAT to be used as user-supplied covariance structure. The file containing this information has the default name
`imprint.gin`

- write out a list of codes linking gamete effects to running numbers as file
`imprint.codes`

For each animal, the paternal gamete is coded as ID*10+1 where ID is the original animal code as supplied in data and pedigree files. Correspondingly, the maternal gamete is ID*10+2.

Note that this requires that for the largest ID, ID*10+2 does not exceed the biggest FORTRAN integer (2147483647;`Gin4Imprint`

will stop with an error message if this occurs).- imprint_example.par
PED pedigree.dat DAT impdata.dat fixeff 2 animal record imprint end ANAL UNI MODEL RAN animal NRM RAN imprint GIN FIX fixeff TR record END VAR animal 1 16 VAR imprint 1 2 VAR error 1 20

- Add a column containing ID*10+i (i=1 or 2) to your data file.
- Update your parameter file by
- calling this new column
`imprint`

, - adding
`RAN imprint GIN`

to the model block, and - providing starting value(s) for the corresponding (co)variance components.

Ready!

**N.B.** The above shows how to fit a gametic effect in addition to animals' additive genetic effects. Other models can be fitted analogously, e.g. we may wish to fit both maternal and paternal gametic effects and allow for a correlation between them, but exclude additive genetic effects.

- [2011, article]
- Tier, B., & Meyer, K. (2011). The effect of imprinted genes on carcass traits in Australian Angus and Hereford cattle.
*Proc. Ass. Advan. Anim. Breed. Genet.*, 19, 63-66. - [2012, article | www]
- Tier, B., & Meyer, K. (2012). On the analysis of parent-of-origin effects with examples from ultrasonic measures of carcass traits in Australian beef cattle.
*Journal of Animal Breeding and Genetics*, 129(5), 359-368. - [2012, article | www]
- Meyer, K., & Tier, B. (2012). Estimates of variances due to parent of origin effects for weights of Australian beef cattle.
*Animal Production Science*, 52, 215-224.