QTLRel Package: functionalities (1)

QTLRel: Tools for mapping of quantitative traits of genetically related individuals and calculating identity coefficients from a pedigree.

This software provides tools for quantitative trait mapping in populations such as advanced intercross lines where relatedness among individuals should not be ignored. It can estimate background genetic variance components, imputing missing genotypes, simulate genotypes, perform a genome scan for putative quantitative trait loci (QTL), and plot mapping results. It also has functions to calculate identity coefficients from pedigree, especially suitable for pedigrees that consist of a large number of generations, or estimate identity coefficients from genotypic data in certain circumstances.

Functions:

1.aicVC: AIC Model Selection, Select genetic variance components via Akaike's information criteria (AIC).

In genome-wide association studies (GWAS), random effects are usually added to a model to account for polygenic variation. Abney et al (2000) showed that five variance components including the most interesting additive and dominance variance components are potentially induced by polygenes. The above function is intended for selecting variance components that contribute "most" to a quantitative trait.

 

2.blup: Best Linear Unbiased Prediction, Estimate the best linear unbiased prediction (BLUP) for various effects in the model.

3.cic: Calculate Jacquard condensed identity coefficient, Calculate Jacquard condensed identity coefficients from a pedigree.

The coefficients will be calculated for individuals with IDs specified by ids. All individuals will be considered if ids is missing. This is not recommended if the total number of individuals in the pedigree is large. Instead, it is recommended that ids is specified for interested individuals only. df is a tuning parameter. It should not be 0 (or smaller than 1) if the pedigree is large in depth (many generations) but the number of individuals is not small; otherwise, it can take forever to finish. It should not be Inf (or a large number) if the number of individuals in certain intermediate generation is very large.

4.estVC: Estimate Variance Components Parameters, Estimate model parameters for covariates, genetic variance components and residual effect.

The optimization function optim is adopted in the above function to estimate the parameters and maximum likelihood. Several optimization methods are available for the optimization algorithms in optim, but we recommend "Nelder-Mead" for the sake of stability. Alternatively, one may choose other options, e.g., "BFGS" to initialize and speed up the estimation procedure and then the procedure will automatically turn to "Nelder-Mead" for final results.

Normality is assumed for the random effects. Input data should be free of missing values. 

5. genMatrix: Derive genetic matrices, Derive genetic matrices from Jacquard condensed identity coefficients or genotypic data.

6.genImpute: Impute Genotypic Data, Impute missing genotypic data in advance intercross lines (AIL)

The missing genotypic value is randomly assigned with a probability conditional on the genotypes of the flanking SNPs (markers).

An object, prd, from genoProb alone can be used for the purpose of imputation. Then, the output (especially the putative loci) will be determined by prd. Optionally, it can be used together with gdat so that missing values in gdat will be imputed if possible, depending on whether loci in the columns of gdat can be identified in the third dimension of prd; this won't change the original genotypic data,

7.genoProb: Probability of a Genotype, Calculate the probability of a genotype at a locus conditional on the genotypes of its flanking markers in advance intercross lines (AIL)

The "cumulative" genetic distance between any two adjacent loci for which probabilities are calculated is not larger than step. If step=Inf, probabilities will only be calculated at loci in both the columns of gdat and the rows of gmap. if step is small, a large set of putative loci will be considered, including all loci defined by the columns of gdat and the rows of gmap.

8. genoSim: Generate Genotypic Dta, Simulate genotypic data from a pedigree in advanced intercross lines (AIL).

The pedigree should be in the same format as an output of pedRecod. Note that two and only two founders are allowed; Otherwise, errors will occur without notice.

A matrix, with entry value s-1 where s is the summation of the numbers representing two alleles at a locus. For instance, 1, 2, and 3 representing genotypes "AA", "AB", and "BB" respectively if hap is not specified. Each row represent an observation, and each column corresponds to SNP in gmap.