amUniqueProfile {allelematch} | R Documentation |

## Determine optimal parameter values for the identification of unique genotypes

### Description

Function to automatically run `amUnique`

at a sequence of parameter values to
determine an optimal setting, and optionally plot the result

### Usage

```
amUniqueProfile(
amDatasetFocal,
multilocusMap = NULL,
alleleMismatch = NULL,
matchThreshold = NULL,
cutHeight = NULL,
guessOptimum = TRUE,
doPlot = TRUE,
consensusMethod = 1,
verbose = TRUE
)
```

### Arguments

`amDatasetFocal` |
An |

`multilocusMap` |
Optionally a vector of integers or strings giving the mappings onto loci for all
genotype columns in amDatasetFocal. |

`alleleMismatch` |
A vector giving a sequence, where elements give the maximum number of mismatching
alleles which will be tolerated when identifying individuals; also known as the
m-hat parameter. |

`matchThreshold` |
A vector giving a sequence, where elements give the minimum dissimilarity score
which constitutes a match when identifying individuals; also known as the s-hat
parameter. |

`cutHeight` |
A vector giving a sequence, where elements give the |

`doPlot` |
If |

`guessOptimum` |
If |

`consensusMethod` |
The method (an integer) used to determine the consensus multilocus genotype from a
cluster of multilocus genotypes. |

`verbose` |
If |

### Details

Selecting the appropriate value for `alleleMismatch`

, `cutHeight`

, or
`matchThreshold`

is an important task. Use this function to assist in this
process. Typically the optimal value of any of these parameters is found where the
number of multiple matches is minimized (the majority of samples are similar to only
one unique genotype). Usually there is a minimum when these parameters are set to be
very sensitive to differences among samples (i.e., `alleleMismatch`

or
`cutHeight`

are 0, `matchThreshold`

is 1). Simulations suggest that the next
most sensitive minimum in multiple matches is the optimal value. This minimum will
often be associated with a drop in multiple matches as sensitivity drops. For more
discussion of this important step, see the Data S1 Supplementary documentation and
tutorials (PDF) located at <doi:10.1111/j.1755-0998.2012.03137.x>.

Using `guessOptimum = TRUE`

will attempt to estimate the location of this minimum
and add it to the profile plot. Manual assessment of this estimate using the plot is
strongly recommended.

If none of `alleleMismatch`

, `cutHeight`

, or `matchThreshold`

is given,
the function runs a sequence of values for `alleleMismatch`

as follows:
`seq(from = 0, to = floor(ncol(amDatasetFocal$multilocus) * 0.4), by = 1)`

`multilocusMap`

is often not required, as `amDataset`

objects will typically
consist of paired columns of genotypes, where each pair is a separate locus. In cases
where this is not the case (e.g., gender is given in only one column), a map vector
must be specified.

Example: `amDataset`

consists of gender followed by 4 diploid loci in paired
columns

`multilocusMap = c(1, 2, 2, 3, 3, 4, 4, 5, 5)`

or equally

```
multilocusMap=c("GENDER", "LOC1", "LOC1", "LOC2", "LOC2", "LOC3", "LOC4",
"LOC4")
```

For more information on selecting `consensusMethod`

see `amCluster`

.
The default `consensusMethod = 1`

is typically adequate.

### Value

A `data.frame`

containing summary data from multiple runs of `amUnique`

### Author(s)

Paul Galpern (pgalpern@gmail.com)

### References

For a complete vignette, please access via the Data S1 Supplementary documentation and tutorials (PDF) located at <doi:10.1111/j.1755-0998.2012.03137.x>.

### See Also

### Examples

```
## Not run:
data("amExample2")
## Produce amDataset object
myDataset <-
amDataset(
amExample2,
missingCode = "-99",
indexColumn = 1,
metaDataColumn = 2
)
## Usage (uncomment)
myUniqueProfile <-
amUniqueProfile(myDataset)
## Data set with gender information
data("amExample5")
## Produce amDataset object
myDataset2 <-
amDataset(
amExample5,
missingCode = "-99",
indexColumn = 1,
metaDataColumn = 2
)
## Usage
myUniqueProfile <-
amUniqueProfile(
myDataset2,
multilocusMap = c(1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10,
11, 11))
## End(Not run)
```

*allelematch*version 2.5.4 Index]