Input data

The input data panel allows to select the input files used in the power study and to specify some parameters.

In a single pedigree file study only the Ped file (A) should be selected.

In a comparative study both Ped file (A) and Ped file (B) need to be specified. In that case Ped file (A) should describe the bigger sample, both in term of number of terms of family units and in terms of family members. When exploring the effect on power derived from the partitioning of a single large genealogy (or of very large different families) in sub-pedigrees, Ped file (A) should contain the undivided sample and Ped file (B) the extracted sub-pedigrees (see for example Falchi et al, 2004). The extracted sub-pedigrees can also have a subset of the individuals described in Ped file (A).

The Map file describes the relative location in the chromosome of the markers and the QTL.

#of replicates states the number of replicas of the family (s) described in the pedigree file that should be simulated within the study.

If the pedigree file describes the entire set of families that are going to be used in the study, #of replicates contains 1. If the pedigree file describes one or more sampling units (for instance contains a single nuclear family with 3 sibs), #of replicates specify how many replicas of the sampling unit should be considered in the power study.

#of simulations are the number of simulations cycles that should be evaluated in the power study. Each cycle correspond to a single analysis of simulated data for the input pedigrees and all the markers described in the map file.

The Random seed is the number used by the random number generator to simulate the first dataset and is automatically incremented within the simulations. When the same study is evaluated more than one time using the same initial random seed, it will produce exactly the same results.

Suggestive LOD, Significant LOD and Highly significant LOD are the user specified LOD threshold that will be used to generate the power histograms.

Genetic model

The genetic model panel allows describing the genetic model for the simulated trait. It currently comprises a diallelic QTL with additive effects (having alleles a and A), a residual additive polygenic effect and a random individual specific effect.

aa homozygous mean is the genotypic value of the “aa” genotype. The Displacement specifies the number of standard deviations between the aa genotypic value and the AA genotypic value. Since additive effects are modelled, the Aa genotypic value is midway between the aa and the AA genotypic values.

Standard deviation specifies the standard deviation within each genotype, which is assumed to be the same for each genotype and can be determined by environmental and polygenic effects.

The Residual h2 is the portion of the variance within each genotype (the squared Standard deviation) due to additive polygenic effects.

In the right side of the panel will be shown, according to the specified parameters, the total narrow heritability for the trait (h2), the QTL heritability (h2 QTL), the residual polygenic heritability (h2R), and the variances due to each factor.

When the values are manually changed, by writing a number in the respective fields, the model should be updated pressing the Evaluate button.

The distribution of LOD scores under a null model can be easily tested by imposing a QTL heritability of zero (using a displacement=0), or assuming a major QTL is present in the genome but is unlinked with the tested markers by specifying a distance from the QTL and the markers in the map file big enough to allow independent assortment.