calculateHeight: Calculate Peak Height.

In OskarHansson/strvalidator: Process Control and Validation of Forensic STR Kits

Calculate Peak Height.

Description

Calculate peak height metrics for samples.

Usage

calculateHeight(
  data,
  ref = NULL,
  na.replace = NULL,
  add = TRUE,
  exclude = NULL,
  sex.rm = FALSE,
  qs.rm = FALSE,
  kit = NULL,
  ignore.case = TRUE,
  exact = FALSE,
  word = FALSE,
  debug = FALSE
)

Arguments

data	data.frame with at least columns 'Sample.Name' and 'Height'.
ref	data.frame with at least columns 'Sample.Name' and 'Allele'.
na.replace	replaces NA values in the final result.
add	logical default is TRUE which will add or overwrite columns 'TPH', 'Peaks', 'H', and 'Proportion' in the provided 'data'.
exclude	character vector (case sensitive) e.g. "OL" excludes rows with "OL" in the 'Allele' column (not necessary when a reference dataset is provided).
sex.rm	logical, default FALSE to include sex markers in the analysis.
qs.rm	logical, default TRUE to exclude quality sensors from the analysis.
kit	character, required if sex.rm=TRUE or qs.rm=TRUE to define the kit.
ignore.case	logical TRUE ignores case in sample name matching.
exact	logical TRUE for exact sample name matching.
word	logical TRUE to add word boundaries to sample name matching.
debug	logical indicating printing debug information.

Details

Calculates the total peak height (TPH), and number of observed peaks (Peaks), for each sample by default. If a reference dataset is provided average peak height (H), and profile proportion (Proportion) are calculated.

H is calculated according to the formula (references [1][2]): H = sum(peak heights)/(n[het] + 2n[hom] Where: n[het] = number of observed heterozygous alleles n[hom] = number of observed homozygous alleles

Important: The above formula has a drawback that when many alleles have dropped out, i.e. when only few alleles are detected, H can be overestimated. For example, if there are only 1 (homozygote) peak observed in the profile, with a height of 100 RFU, then H=100 RFU. This means that the value of H will always be between half the analytical threshold (AT/2) and the peak height of the observed allele (if only one). For this reason Tvedebrink et al. actually modified the estimate to take the number of expected alleles into account when estimating the expected peak height (reference [3]). Basically, they adjust the estimated peak height for the fact that they know how many alleles that fall below the AT, such that the expected peak height could be estimated lower than AT. In addition, they account for degradation using a log-linear relationship on peak heights and fragment length.

Tip: If it is known that all expected peaks are observed and no unexpected peaks are present, the dataset can be used as a reference for itself.

Note: If a reference dataset is provided the known alleles will be extracted from the dataset.

Value

data.frame with with at least columns 'Sample.Name', 'TPH', and 'Peaks'.

References

[1] Torben Tvedebrink, Poul Svante Eriksen, Helle Smidt Mogensen, Niels Morling, Evaluating the weight of evidence by using quantitative short tandem repeat data in DNA mixtures Journal of the Royal Statistical Society: Series C (Applied Statistics), Volume 59, Issue 5, 2010, Pages 855-874, 10.1111/j.1467-9876.2010.00722.x. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1111/j.1467-9876.2010.00722.x")}

[2] Torben Tvedebrink, Helle Smidt Mogensen, Maria Charlotte Stene, Niels Morling, Performance of two 17 locus forensic identification STR kits-Applied Biosystems's AmpFlSTR NGMSElect and Promega's PowerPlex ESI17 kits Forensic Science International: Genetics, Volume 6, Issue 5, 2012, Pages 523-531, 10.1016/j.fsigen.2011.12.006. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.fsigen.2011.12.006")}

[3] Torben Tvedebrink, Maria Asplund, Poul Svante Eriksen, Helle Smidt Mogensen, Niels Morling, Estimating drop-out probabilities of STR alleles accounting for stutters, detection threshold truncation and degradation Forensic Science International: Genetics Supplement Series, Volume 4, Issue 1, 2013, Pages e51-e52, 10.1016/j.fsigss.2013.10.026. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.fsigss.2013.10.026")}

OskarHansson/strvalidator documentation built on July 22, 2023, 12:04 p.m.