Mixtures

In disclapmix: Discrete Laplace Mixture Inference using the EM Algorithm

knitr::opts_chunk$set(fig.width = 5, fig.height = 5)

Introduction

This vignette shows how to use the R package disclapmix that implements the method described in [@AndersenDisclap2013] and [@AndersenDisclapMixture2015]. For a more gentle introduction to the method, refer to the introduction vignette and [@AndersenDisclapIntroduction2013].

Mixture analysis

We again use the Danish reference database [@Hallenberg2005YchromosomeSH] with $n = 185$ observations (male Y-STR haplotypes) at $r=10$ loci is available in the danes dataset. Let us load the package as well as the data:

library(disclapmix)
data(danes)

The database is in compact format, i.e. one unique haplotype per row. To fit the model, we need one observation per row. This is done for example like this:

db <- as.matrix(danes[rep(seq_len(nrow(danes)), danes$n), seq_len(ncol(danes)-1)])
str(db)

Also, note that the database is now an integer matrix.

Assume now that we have a mixture and that the reference database are without these two contributors:

donor1 <- db[1, ]
donor2 <- db[20, ]
refdb <- db[-c(1, 20), ]

We now construct the mixture:

mix <- generate_mixture(list(donor1, donor2))

We can then see some properties of possible pairs:

pairs <- contributor_pairs(mix)
pairs

To do much more, we need a model assigning hpalotype probabilies. In the introduction vignette, we found that 4 clusters seemed fine, so let us fit this model:

fit <- disclapmix(x = refdb, clusters = 4L)

We can now use this model to e.g. rank the contributor pairs:

ranked_pairs <- rank_contributor_pairs(pairs, fit)
ranked_pairs

We can get the ranks for the donors:

get_rank(ranked_pairs, donor1)
get_rank(ranked_pairs, donor2)

References

disclapmix documentation built on June 29, 2022, 5:06 p.m.