Package 'fingers'

Title: Identifying Clusters of Related Individuals
Description: Identifying clusters of related individuals.
Authors: Karl W Broman [aut, cre] , Laura Plantinga [ctb]
Maintainer: Karl W Broman <[email protected]>
License: GPL-3
Version: 0.60-6
Built: 2026-05-09 20:41:57 UTC
Source: https://github.com/kbroman/fingers

Help Index

Data on Aedes aegypti

Description

This is RAPD data for 40 loci typed on a set of 10 full-sibling families, with 15 individuals in each family.

Usage

data(aedes)

Format

The data is a matrix of 150 rows (the individuals) by 40 columns (the RAPD loci). Each entry is a RAPD phenotype, indicating the presence (1) or absence (0) of a band.

Author(s)

Karl W Broman [email protected]

Source

FINGERS software, WC Black IV, Colorado State University

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

shiff1, simrapd

Examples

data(aedes)

Calculate simple distance matrix

Description

Calculate the simple distance matrix, by the proportion of mismatches, for a RAPD data set.

Usage

calc.dist(dat)

Arguments

dat

A matrix of size (n.ind x n.mar) containing RAPD phenotypes, with 1 indicating the presence of a band and 0 indicating absence.

Details

For each pair of individuals, we calculate the proportion of RAPD markers (among those where both individuals have complete data) at which one individual shows a band and the other doesn't.

Value

A symmetric matrix of dimension (n.ind x n.ind), containing the distances between individuals.

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

llrdist, fingers

Examples

data(aedes)
d <- calc.dist(aedes)

Calculate measure of quality of inferred clusters

Description

Calculate a score indicating how well two sets of clusters conform.

Usage

cluster.stat(fam1,fam2,method=c("all","rand","adj","fm","kb"))

Arguments

fam1

A list of clusters; each component in the list is one family, containing the indices of the individuals in that family.
fam2

A list, just like fam1.
method

A character string indicating whether to calculate the Rand index, the adjusted Rand index, the Fowlkes and Mallows B index, or Karl Broman's index. If method=all, a vector with all four indices is returned.

Details

In the Rand index (Rand 1971), one considers all pairs of individuals, and assigns a 1 to a pair if the individuals are either in the same cluster in both fam1 and fam2 or are not in the same cluster in both fam1 and fam2, and assigns a 0 to the pair otherwise, and then takes the sum of these, divided by the number of pairs of individuals.

Karl Broman's index (which we don't recommend, but we implement here in order to allow comparisons to be made) is just like the Rand index, but fam2 is assumed to be the true partition, and the set of all pairs in the same group (by fam2) and the set of all pairs in different groups (by fam2), are given equal weight.

Let nijn_{ij} be the number of individuals in group i by partition 1 and group j by partition 2. Let ni.=∑jnijn_{i.} = \sum_{j} n_{ij} and define n.jn_{.j} similarly.

In the adjusted-Rand index (Hubert and Arabie 1985), ...

In the Fowlkes and Mallows B index (Fowlkes and Mallows 1983), ...

Value

The value of a score for comparing two sets of clusters.

Author(s)

Karl W Broman [email protected]

References

WM Rand (1971) Objective criteria for the evaluation of clustering methods. Journal of the American Statistical Association 66:846-850.

L Hubert and P Arabie (1985) Comparing partitions. Journal of Classification. 2:193-218.

EB Fowlkes and CL Mallows (1983) A method for comparing two hierarchical clusterings. Journal of the American Statistical Association 78:553-584.

BS Everitt, S Landau and M Leese (2001) Cluster analysis, 4th edition. Arnold, London, pp. 181-3.

See Also

fingers, true.fams

Examples

data(aedes)
f <- freq(aedes)
co <- cutoff(f)
d <- calc.dist(aedes)
fam <- fingers(d,co,make.plot=TRUE)
tf <- true.fams(aedes)
cluster.stat(fam,tf)
cluster.stat(fam,tf,method="fm")

Compare two sets of clusters

Description

Give diagnostic information indicating how well two sets of clusters conform.

Usage

comp.fams(fam1,fam2)

Arguments

fam1

A list of clusters; each component in the list is one family, containing the indices of the individuals in that family.
fam2

A list, just like fam1.

Value

A list with two components. The first component is a contingency table whose (i,j)th element is the number of individuals in cluster i in fam1 and cluster j in fam2. The second component is a list indicating, for each cluster from fam1, the cluster assignment in fam2.

Author(s)

Karl W Broman [email protected]

See Also

cluster.stat, fingers, true.fams

Examples

data(aedes)
f <- freq(aedes)
co <- cutoff(f)
d <- calc.dist(aedes)
fam <- fingers(d,co,make.plot=TRUE)
tf <- true.fams(aedes)
comp.fams(fam,tf)

Calculate cutoff for clustering with RAPD markers

Description

Calculate the cutoff for hierarchical cluster analysis to infer groups of related individuals with RAPD data.

Usage

cutoff(f,method=c("qu","meansib","qs","lr"),value=0.2)

Arguments

f

A vector of band allele frequencies for a set of RAPD markers.
method

The method to use to form the cutoff: a quantile of the distribution of distances among unrelated (qu), the mean distance between siblings (meansib), a quantile of the distribution of distances among siblings (qs), or the likelihood ratio for unrelateds vs. siblings (lr).
value

For method="qu" or method="qs", this should specify the quantile; for method="lr", this should specify the likelihood ratio.

Value

The cutoff (a single value).

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

cutoff.llr, freq, pull.markers, fingers

Examples

data(aedes)
f <- freq(aedes)
co1 <- cutoff(f,method="meansib")
co2 <- cutoff(f,method="qu",value=0.2)
co3 <- cutoff(f,method="qs",value=0.9)
co4 <- cutoff(f,method="lr",value=4.0)

Calculate cutoff for clustering with RAPD markers

Description

Calculate a cutoff (for the LLR distance measure) for hierarchical cluster analysis to infer groups of related individuals with RAPD data.

Usage

cutoff.llr(f,method=c("qu","meansib","qs","lr"),value=0.2)

Arguments

f

A vector of band allele frequencies for a set of RAPD markers.
method

The method to use to form the cutoff: a quantile of the distribution of distances among unrelated (qu), the mean distance between siblings (meansib), a quantile of the distribution of distances among siblings (qs), or the likelihood ratio for unrelateds vs. siblings (lr).
value

For method="qu" or method="qs", this should specify the quantile; for method="lr", this should specify the likelihood ratio.

Value

The cutoff (a single value).

Author(s)

Karl W Broman [email protected]

See Also

cutoff, llrdist, freq, pull.markers, fingers

Examples

data(aedes)
f <- freq(aedes)
co1 <- cutoff.llr(f,method="meansib")
co2 <- cutoff.llr(f,method="qu",value=0.2)
co3 <- cutoff.llr(f,method="qs",value=0.9)
co4 <- cutoff.llr(f,method="lr",value=4.0)

Plot distance matrix

Description

Plot the distance matrix for a RAPD data set, with (optionally) lines drawn separating clusters of individuals.

Usage

dist.image(dist,fams=NULL,col=topo.colors(1+ncol(dist)),...)

Arguments

dist

A matrix of size (n.ind x n.ind), containing the distances between pairs of individuals.
fams

A list of clusters; each component in the list is one inferred family, containing the indices of individuals placed in that family.
col

Colors to use in the plot; see image.
...

Other arguments to pass to image.

Value

The function calls image in order to create an image of the distance matrix.

Author(s)

Karl W Broman [email protected]

See Also

calc.dist, true.fams

Examples

data(aedes)
f <- freq(aedes)
co <- cutoff(f)
d <- calc.dist(aedes)
fam <- fingers(d,co,make.plot=TRUE)
dist.image(d,fam)

Infer clusters of related individuals

Description

Perform hierarchical clustering to infer groups of related individuals with RAPD data.

Usage

fingers(dist,cutoff=NULL,method=c("average","complete",
         "mcquitty","single","ward"),truefam=NULL,
         make.plot=FALSE,just.plot=FALSE)

Arguments

dist

A matrix of size (n.ind x n.ind) containing the distances between individuals.
cutoff

A value to use to cut off the dendogram formed by hierarchical clustering in order to define a set of clusters. (Optional, but if NULL, the argument truefam must be included.)
method

A hierarchical clustering method. See hclust. Note: We haven't allowed centroid or median, because these weren't working for us.
truefam

The true family structure; used only if cutoff is NULL, in which case all possible cutoffs are tried, and that giving the maximum adjusted Rand index is used.
make.plot

If TRUE, make a plot of the dendogram formed by hierarchical clustering.
just.plot

If TRUE, just make the plot; don't return the inferred families. (In this case, the cutoff argument is not needed.)

Details

We use the function hclust to do the cluster analysis.

Value

A list of clusters; each component in the list is one inferred family, containing the indices of individuals placed in that family. The cutoff used is included as an attribute. Use attr(result,"cutoff") to obtain this value.

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

cutoff, cutoff.llr, calc.dist, llrdist, cluster.stat, true.fams, freq, pull.markers

Examples

data(aedes)
f <- freq(aedes)
co <- cutoff(f)
d <- calc.dist(aedes)
fam <- fingers(d,co,make.plot=TRUE)
tf <- true.fams(aedes)
cluster.stat(fam,tf)

Estimate RAPD allele frequencies

Description

Estimate the frequency of the band allele for a set of RAPD markers.

Usage

freq(dat)

Arguments

dat

A matrix of size (n.ind x n.mar) containing RAPD phenotypes, with 1 indicating the presence of a band and 0 indicating absence.

Details

The RAPDs are assumed to be in Hardy-Weinberg equilibrium, and so the frequency of the band allele is estimated as p^=1−1−xˉ\hat{p} = 1-\sqrt{1-\bar{x}} where xˉ\bar{x} is the proportion of individuals showing a band.

Value

A vector of length n.mar, containing the estimated frequencies of the band allele for each RAPD marker.

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

pull.markers

Examples

data(aedes)
f <- freq(aedes)

Calculate distance matrix based on log likelihood ratio

Description

Calculate a distance matrix, based on the log likelihood ratio comparing the hypotheses of full sibling versus unrelated, for a RAPD data set.

Usage

llrdist(dat,p=freq(dat))

Arguments

dat

A matrix of size (n.ind x n.mar) containing RAPD phenotypes, with 1 indicating the presence of a band and 0 indicating absence.
p

A vector of band allele frequencies.

Details

For each pair of individuals, at each locus, we calculate the log likelihood ratio (LLR) comparing the hypotheses unrelated with siblings, with the data being B (both have band), N (neither have band) or D (one has band, the other doesn't). These LLRs are averaged across individuals. Note: at each locus, we re-center the LLRs so that the minimum of the LLRs among B/N/D is 0; this makes the resulting distances ≥\ge 0.

Calculations are performed in a C program.

Value

A symmetric matrix of dimension (n.ind x n.ind), containing the distances between individuals.

Author(s)

Karl W Broman [email protected]

See Also

calc.dist, fingers

Examples

data(aedes)
f <- freq(aedes)
dis <- llrdist(aedes,f)

Extract markers with allele frequencies in specified range

Description

Extract markers from a RAPD data set that have allele frequencies within a specified range.

Usage

pull.markers(dat,lo=0.1,hi=0.6,f=freq(dat))

Arguments

dat

A matrix of size (n.ind x n.mar) containing RAPD phenotypes, with 1 indicating the presence of a band and 0 indicating absence.
lo

Lower bound for band allele frequency.
hi

Upper bound for band allele frequency.
f

Vector of band allele frequencies (included in order to avoid recalculating it, if possible).

Value

A matrix, like the argument dat, but containing only those markers with band allele frequency between lo and hi.

Author(s)

Karl W Broman [email protected]

See Also

freq

Examples

data(shiff1)
f <- freq(shiff1)
subset <- pull.markers(shiff1, 0.1, 0.6, f)

Schistosome data

Description

This is RAPD data for 35 loci typed on a set of 135 individuals.

Usage

data(shiff1)

Format

The data is a matrix of 135 rows (the individuals) by 35 columns (the RAPD loci). Each entry is a RAPD phenotype, indicating the presence (1) or absence (0) of a band.

Author(s)

Karl W Broman [email protected]

Source

Clive Shiff, Molecular Microbiology and Immunology, Bloomberg School of Public Health, The Johns Hopkins University

See Also

shiff2, shiff3, aedes, simrapd

Examples

data(shiff1)

Schistosome data

Description

This is RAPD data for 10 loci typed on a set of 135 individuals. Markers with estimated band allele frequencies outside of the range 0.1-0.6 have been removed.

Usage

data(shiff2)

Format

The data is a matrix of 135 rows (the individuals) by 10 columns (the RAPD loci). Each entry is a RAPD phenotype, indicating the presence (1) or absence (0) of a band.

Author(s)

Karl W Broman [email protected]

Source

Clive Shiff, Molecular Microbiology and Immunology, Bloomberg School of Public Health, The Johns Hopkins University

See Also

shiff1, shiff3, aedes, simrapd

Examples

data(shiff2)

Schistosome data

Description

This is RAPD data for 10 loci typed on a set of 125 individuals. Markers with estimated band allele frequencies outside of the range 0.1-0.6 have been removed. Individuals with one or more missing values have been removed.

Usage

data(shiff3)

Format

The data is a matrix of 125 rows (the individuals) by 10 columns (the RAPD loci). Each entry is a RAPD phenotype, indicating the presence (1) or absence (0) of a band.

Author(s)

Karl W Broman [email protected]

Source

Clive Shiff, Molecular Microbiology and Immunology, Bloomberg School of Public Health, The Johns Hopkins University

See Also

shiff1, shiff2, aedes, simrapd

Examples

data(shiff3)

Simulate RAPD data

Description

Simulates RAPD data for a set of sibling families.

Usage

simrapd(n.sib = rep(15,10), p = c(rep(0.125,8),rep(0.175,5),rep(0.225,5),
           rep(0.275,8),rep(0.325,3),rep(0.375,4),
           rep(0.475,4),rep(0.575,3)))

Arguments

n.sib

A vector giving the number of siblings per family (length is the number of families).
p

A vector of frequencies of the band allele at each marker (length is the number of markers).

Details

The RAPDs are assumed to be in Hardy-Weinberg equilibrium.

Value

A matrix of dimension (n.ind x n.mar), giving the RAPD phenotypes for each individual at each marker, with 1 indicating a band and 0 indicating no band.

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

simulfams

Examples

data <- simrapd(rep(20,5), p=runif(40, 0.1, 0.6))

Simulate RAPD data

Description

Simulates RAPD data for a set of sibling families.

Usage

simulfams(n.sib=sample(5:20,size=sample(5:20,size=1),replace=TRUE),
          p=runif(sample(5:15,size=1),min=0.1,max=0.6))

Arguments

n.sib

A vector giving the number of siblings per family (length is the number of families).
p

A vector of frequencies of the band allele at each marker (length is the number of markers).

Details

The RAPDs are assumed to be in Hardy-Weinberg equilibrium.

Value

A matrix of dimension (n.ind x n.mar), giving the RAPD phenotypes for each individual at each marker, with 1 indicating a band and 0 indicating no band.

Author(s)

Laura Plantinga and Karl Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

simrapd

Examples

data <- simulfams(rep(20,5), p=runif(40, 0.1, 0.6))

Identify the true clusters

Description

Use the row names of a RAPD data set to identify the true sets of families.

Usage

true.fams(dat)

Arguments

dat

A matrix of size (n.ind x n.mar) containing RAPD phenotypes, with 1 indicating the presence of a band and 0 indicating absence. The row names (identifying individuals) are assumed to be of the form "family-individual"

Value

A list of clusters; each component in the list is one inferred family, containing the indices of individuals placed in that family.

Author(s)

Karl W Broman [email protected]

References

BL Apostol, WC Black IV, BR Miller, P Reiter, BJ Beaty (1993) Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti. Theor Appl Genet 86:991-1000.

See Also

aedes, simrapd, fingers, cluster.stat

Examples

data(aedes)
tf <- true.fams(aedes)