Type:	Package
Title:	Local Haplotype Clustering and Visualization
Version:	1.4.0
Maintainer:	Jacob Marsh <jake.marsh@live.com.au>
Description:	A local haplotyping visualization toolbox to capture major patterns of co-inheritance between clusters of linked variants, whilst connecting findings to phenotypic and demographic traits across individuals. 'crosshap' enables users to explore and understand genomic variation across a trait-associated region. For an example of successful local haplotype analysis, see Marsh et al. (2022) <doi:10.1007/s00122-022-04045-8>.
Encoding:	UTF-8
LazyData:	true
LazyDataCompression:	xz
RoxygenNote:	7.2.3
Imports:	cli, clustree, data.table, dbscan, dplyr, ggdist, ggplot2, ggpp, gridExtra, gtable, magrittr, patchwork, rlang, scales, tibble, tidyr
Suggests:	covr, knitr, rmarkdown, testthat (≥ 3.0.0), umap, vdiffr
Depends:	R (≥ 4.00)
URL:	https://jacobimarsh.github.io/crosshap/
BugReports:	https://github.com/jacobimarsh/crosshap/issues
License:	MIT + file LICENSE
Config/testthat/edition:	3
NeedsCompilation:	no
Packaged:	2024-03-31 15:17:05 UTC; jmarsh96
Author:	Jacob Marsh [aut, cre], Brady Johnston [aut], Jakob Petereit [aut]
Repository:	CRAN
Date/Publication:	2024-03-31 15:40:02 UTC

crosshap: Local Haplotype Clustering and Visualization

Description

A local haplotyping visualization toolbox to capture major patterns of co-inheritance between clusters of linked variants, whilst connecting findings to phenotypic and demographic traits across individuals. 'crosshap' enables users to explore and understand genomic variation across a trait-associated region. For an example of successful local haplotype analysis, see Marsh et al. (2022) doi:10.1007/s00122-022-04045-8.

Author(s)

Maintainer: Jacob Marsh jake.marsh@live.com.au (ORCID)

Authors:

• Brady Johnston brady.johnston@me.com (ORCID)

• Jakob Petereit petinho86@gmail.com (ORCID)

See Also

Useful links:

• https://jacobimarsh.github.io/crosshap/

• Report bugs at https://github.com/jacobimarsh/crosshap/issues

Pipe operator

Description

See magrittr::%>% for details.

Usage

lhs %>% rhs

Arguments

Value

The result of calling 'rhs(lhs)'.

Example Haplotype object

Description

A haplotyping object created by run_haplotyping() for example cqProt-003 soy data

Usage

HapObject

Format

A haplotype (S3) object containing results needed for haplotype visualization across five epsilon values (0.2,0.4,0.6,0.8,1)

epsilon

Epsilon value chosen for haplotyping with DBSCAN

MGmin

MGmin value (minPts) chosen for haplotyping with DBSCAN

Hapfile

Summary of Marker Groups defining haplotype combinations

Indfile

Haplotype assignments for individuals

Varfile

Marker Group assignments for SNPs, with additional calculated information

Example LD matrix

Description

A pairwise R^2 linkage matrix generated by PLINK for example cqProt-003 soy data

Usage

LD

Format

A square matrix read in by read_LD()

Mode utility function

Description

Mode utility function

Usage

arith_mode(x)

Arguments

Value

Mode numerical values

Bot hap-pheno raincloud plot

Description

build_bot_halfeyeplot() builds a vertical plot displaying the phenotypic scores for each individual, grouped by haplotype, coloured by metadata variable. Metadata groups can be isolated using the isolate_groups argument. Makes use of the $Indfile information from haplotype object. It is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot.

Usage

build_bot_halfeyeplot(
  HapObject,
  epsilon,
  hide_labels = TRUE,
  isolate_group = NA
)

Arguments

Value

A ggplot2 object.

Examples

build_bot_halfeyeplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Left SNP-allele plot

Description

build_left_alleleplot() builds a horizontal plot displaying mean allelic frequencies (reference/alternate/missing/heterozygous) of all SNP loci, grouped by marker group. Makes use of $Varfile information from a HapObject created by run_haplotyping(). This is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot.

Usage

build_left_alleleplot(HapObject, epsilon, hide_labels = TRUE)

Arguments

Value

A ggplot2 object.

Examples

build_left_alleleplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Left SNP-position plot

Description

build_left_alleleplot() builds a horizontal plot displaying the chromosomal position of each SNP locus, grouped by marker group. Makes use of the $Varfile file from haplotype object. It is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot.

Usage

build_left_posplot(HapObject, epsilon, hide_labels = TRUE)

Arguments

Value

A ggplot2 object.

Examples

build_left_posplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Middle MG/hap dot plot

Description

build_mid_dotplot() builds a central dot plot displaying the relationship between haplotype combinations and the characteristic marker group alleles that define them. Makes use of the $Hapfile information from a haplotype object. This is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot (can be useful when patched to a peripheral plot).

Usage

build_mid_dotplot(HapObject, epsilon, hide_labels = FALSE)

Arguments

Value

A ggplot2 object.

Examples

build_mid_dotplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Right intra-cluster linkage plot

Description

build_right_jitterplot() builds a horizontal plot displaying the mean pairwise R^2 linkage between each SNP and all other SNPs in its marker group, grouped by marker group, coloured by alternate allele frequency. Makes use of the $Varfile information from haplotyping object. It is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot.

Usage

build_right_clusterplot(HapObject, epsilon, hide_labels = FALSE)

Arguments

Value

A ggplot2 object.

Examples

build_right_clusterplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Right SNP-pheno phenoplot

Description

build_right_phenoplot() builds a horizontal plot displaying the mean difference in phenotype score between individuals with the alternate vs reference alleles for each SNP locus, grouped by marker group, coloured by the alternate allele frequency of each SNP. Makes use of the $Varfile phenotypic information from haplotyping object. It is an internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot.

Usage

build_right_phenoplot(HapObject, epsilon, hide_labels = TRUE)

Arguments

Value

A ggplot2 object.

Examples

    build_right_phenoplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Hap/MG summary tables

Description

build_summary_tables() builds summary tables for each haplotype and Marker Group with some of the information shown in the peripheral crosshap plots. It is an internal function called by crosshap_viz(), though can be called separately to build stand-along grob tables.

Usage

build_summary_tables(HapObject, epsilon)

Arguments

Value

A list containing two TableGrob objects.

Top metadata-hap bar plot

Description

build_top_metaplot() builds a vertical stacked bar plot displaying the frequency of each haplotype combination, broken down by each categorical metadata variable provided. Makes use of the $Indfile information from a haplotype object. This is an in internal function called by crosshap_viz(), though can be called separately to build a stand-alone plot

Usage

build_top_metaplot(HapObject, epsilon, hide_labels = FALSE)

Arguments

Value

A ggplot2 object.

Examples

build_top_metaplot(HapObject, epsilon = 0.6, hide_labels = FALSE)

Clustering tree

Description

clustree_viz() builds a clustering tree displaying changes in haplotype assignment between individuals or changes in Marker Group assignment for SNPs, across different epsilon values. This function is a 'clustree' wrapper.

Usage

clustree_viz(HapObject, type = "MG")

Arguments

Value

A ggplot2 object.

Visualize haplotypes

Description

crosshap_viz() builds five individual plots using various elements of a HapObject created by run_haplotyping(). The central dotplot displays relationship between clusters of linked SNPs (Marker Groups), and distinct haplotypes present within the population. Vertical plots (top/bottom) visualize individuals and populations, grouped by haplotype. Horizontal plots (left/right) visualize SNP information, grouped by Marker Group cluster.

Usage

crosshap_viz(
  HapObject,
  epsilon,
  plot_left = "allele",
  plot_right = "pheno",
  hide_labels = FALSE,
  isolate_group = NA
)

Arguments

Value

A patchwork object.

Mean utility function

Description

Mean utility function

Usage

mean_na.rm(x)

Arguments

Value

Mean numerical values

Example Domestication metadata

Description

Metadata file with level of domestication for each individual in example cqProt-003 soy data

Usage

metadata

Format

A two-column tibble read in by read_metadata()

Source

https://doi.org/10.1007/s00122-022-04045-8

Example phenotype data

Description

Seed protein scores for each individual in example cqProt-003 soy data

Usage

pheno

Format

A two-column tibble read in by read_pheno()

Source

https://doi.org/10.1007/s00122-022-04045-8

UMAP haplotype visualization helper

Description

prepare_hap_umap() builds a large composite ggplot2 object ready for faceting and animation (see vignette) for visualizing SNP alleles (coloured by Marker Group) possessed by individuals with each haplotype. UMAP coordinates for each SNP can be generated using umap::umap(), with the LD matrix generated for run_haplotyping() as input. When fully rendered and faceted, the resultant GIF intuitively visualizes the shared loci within each Marker Group that are constant within each haplotype combination.

Usage

prepare_hap_umap(
  umap_in,
  hetmiss_as = "allele",
  HapObject,
  epsilon,
  vcf,
  nsamples = 25
)

Arguments

Value

A large ggplot2 object.

Identify haplotypes from clustered SNPs

Description

pseudo_haps() calls the most common allelic states for each SNP marker group across individuals, before building dummy SNPs for each marker group that mimic the binary vcf format. This is the step which determines the haplotype combinations, and therefore enables several summaries to be returned - as contained in the $Hapfile and preliminary $Indfile and finalised $MGfile, following marker group smoothing. This is an internal function not intended for external use.

Usage

pseudo_haps(preMGfile, bin_vcf, minHap, LD, keep_outliers)

Arguments

Value

Returns intermediate of haplotype object

Read LD correlation matrix to tibble

Description

If your correlation matrix does not have rownames and column names, a VCF will need to be provided so it can be added with read_LD().

Usage

read_LD(LDin, vcf = NULL)

Arguments

Value

A tibble.

Read metadata to tibble

Description

Requires two column text file without a header (Ind | Metadata)

Usage

read_metadata(Metain)

Arguments

Value

A tibble.

Read phenotype data to tibble

Description

Requires two column text file without a header (Ind | Pheno)

Usage

read_pheno(Phenoin)

Arguments

Value

A tibble.

Read VCF to tibble

Description

Dashes,'-', in individual names are recoded to '.' for downstream compatability.

Usage

read_vcf(VCFin)

Arguments

Value

A tibble.

Cluster SNPs and identify haplotypes

Description

run_haplotyping() performs density-based clustering of SNPs in region of interest to identify Marker Groups. Individuals are classified by haplotype combination based on shared combinations of Marker Group alleles. Returns a haplotyping object (HapObject), which can be used as input to build clustering tree for epsilon optimization using clustree_viz(), and can be visualized with reference to phenotype and metadata using crosshap_viz().

Usage

run_haplotyping(
  vcf,
  LD,
  pheno,
  metadata = NULL,
  epsilon = c(0.2, 0.4, 0.6, 0.8, 1),
  MGmin = 30,
  minHap = 9,
  hetmiss_as = "allele",
  het_phenos = FALSE,
  keep_outliers = FALSE
)

Arguments

Value

A comprehensive haplotyping S3 object (HapObject) for each provided epsilon value, needed for clustree_viz() and crosshap_viz().

Cluster SNPs with HDBSCAN and identify haplotypes

Description

run_hdbscan_haplotyping() performs HDBSCAN clustering of SNPs in region of interest to identify marker groups. Individuals are classified by haplotype combination based on shared combinations of marker group alleles. Returns a comprehensive haplotyping object (HapObject), which can be visualized with reference to phenotype and metadata using crosshap_viz() (set epsilon to 1 as a dummy value).

Usage

run_hdbscan_haplotyping(
  vcf,
  LD,
  pheno,
  MGmin,
  minHap = 5,
  hetmiss_as = "allele",
  metadata = NULL,
  keep_outliers = FALSE
)

Arguments

Value

A comprehensive haplotyping S3 object (HapObject) for each provided epsilon value, needed for clustree_viz() and crosshap_viz().

Calculate SNP phenotypic associations

Description

tagphenos() reports the frequency of allele types for each SNP and calculates phenotype associations for the different alleles, before returning this information in a $Varfile in a HapObject. This is an internal function that is not intended for external use.

Usage

tagphenos(MGfile, bin_vcf, pheno, het_phenos = FALSE)

Arguments

Value

Returns intermediate of haplotype object.

Example VCF

Description

A VCF containing SNPs for example cqProt-003 soy data

Usage

vcf

Format

A VCF read in by read_vcf()

Source

https://doi.org/10.1007/s00122-022-04045-8