Package {rabhit}

.onAttach start message

Description

.onAttach start message

Usage

.onAttach(libname, pkgname)

Arguments

Value

invisible()

Human Gene order on the chromosome

Description

A list of the chains genes order by their location on the chromosomes

Usage

GENE.loc

Format

A nested list with three enteries, each a vector of the IG chains (IGH, IGL, and IGK) genes ordered by location.

Human germlines

Description

A list of the germline genes from the human immunoglobulin loci

Usage

GERM

Format

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

Human IGHD germlines

Description

A character vector of all 37 human IGHD germline gene segment alleles in IMGT Gene-db release 2018-12-4.

Usage

HDGERM

Format

Values correspond to IMGT nuceltoide sequences.

References

Xochelli et al. (2014) Immunoglobulin heavy variable (IGHV) genes and alleles: new entities, new names and implications for research and prognostication in chronic lymphocytic leukaemia. Immunogenetics. 67(1):61-6.

Human IGHJ germlines

Description

A character vector of all 13 human IGHJ germline gene segment alleles in IMGT Gene-db release 2018-12-4.

Usage

HJGERM

Format

Values correspond to IMGT nuceltoide sequences.

References

Xochelli et al. (2014) Immunoglobulin heavy variable (IGHV) genes and alleles: new entities, new names and implications for research and prognostication in chronic lymphocytic leukaemia. Immunogenetics. 67(1):61-6.

Human IGHV germlines

Description

A character vector of all 342 human IGHV germline gene segment alleles in IMGT Gene-db release 2018-12-4.

Usage

HVGERM

Format

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

References

Xochelli et al. (2014) Immunoglobulin heavy variable (IGHV) genes and alleles: new entities, new names and implications for research and prognostication in chronic lymphocytic leukaemia. Immunogenetics. 67(1):61-6.

Human IGKJ germlines

Description

A character vector of all 342 human IGKJ germline gene segment alleles in IMGT Gene-db release 2019-11-18.

Usage

KJGERM

Format

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

Human IGKV germlines

Description

A character vector of all 342 human IGKV germline gene segment alleles in IMGT Gene-db release 2019-11-18.

A character vector of all 342 human IGLV germline gene segment alleles in IMGT Gene-db release 2019-11-18.

Usage

KVGERM

LVGERM

Format

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

Human IGLJ germlines

Description

A character vector of all 342 human IGLJ germline gene segment alleles in IMGT Gene-db release 2019-11-18.

Usage

LJGERM

Format

Values correspond to IMGT-gaped nuceltoide sequences (with nucleotides capitalized and gaps represented by '.').

Prepare AIRR-seq data and germline for haplotype inference with allele similarity clusters (ASC)

Description

convertToASC converts IMGT-named V alleles to PIgLET Allele Similarity Cluster (ASC) names and returns the pieces needed to run createFullHaplotype on the clustered data. ASCs collapse near-identical and duplicated IGHV alleles (for example IGHV1-69 / IGHV1-69D) that are otherwise difficult to assign unambiguously, and provide a naming scheme compatible with OGRDB reference sets. It is a thin wrapper around the piglet package.

Usage

convertToASC(
  clip_db,
  germline,
  chain = c("IGH", "IGK", "IGL", "TRB"),
  allele_cluster_table = NULL,
  v_call = "v_call",
  ...
)

Arguments

Details

The returned objects feed directly into the standard workflow:

asc <- convertToASC(clip_db, HVGERM, chain = "IGH")
hap <- createFullHaplotype(asc$clip_db, toHap_col = "v_call",
                           hapBy_col = "j_call", hapBy = "IGHJ6",
                           toHap_GERM = asc$germline, chain = "IGH")
plotHaplotype(hap, genes_order = asc$genes_order)

Allele similarity clustering is applied to the V genes only; D and J calls are left unchanged.

Value

a list with:

• clip_db - the input data with v_call converted to ASC names.

• germline - the ASC-named germline vector to pass as toHap_GERM.

• genes_order - an ASC gene order (derived from the chromosomal order in GENE.loc) to pass to the plotting functions.

• allele_cluster_table - the ASC table that was used.

See Also

createFullHaplotype; the piglet package for the underlying clustering (inferAlleleClusters, assignAlleleClusters, germlineASC).

Examples

# requires the Suggested 'piglet' package
if (requireNamespace("piglet", quietly = TRUE)) {
  data(samples_db, HVGERM)
  clip_db <- samples_db[samples_db$subject == "I5", ]
  asc <- convertToASC(clip_db, HVGERM, chain = "IGH")
  hap <- createFullHaplotype(asc$clip_db, toHap_col = "v_call",
                             hapBy_col = "j_call", hapBy = "IGHJ6",
                             toHap_GERM = asc$germline, chain = "IGH")
  plotHaplotype(hap, genes_order = asc$genes_order)
}

Anchor gene haplotype inference

Description

The createFullHaplotype functions infers haplotype based on an anchor gene.

Usage

createFullHaplotype(
  clip_db,
  toHap_col = c("v_call", "d_call"),
  hapBy_col = "j_call",
  hapBy = "IGHJ6",
  toHap_GERM = NULL,
  relative_freq_priors = TRUE,
  kThreshDel = 3,
  rmPseudo = TRUE,
  deleted_genes = c(),
  nonReliable_Vgenes = c(),
  min_minor_fraction = 0.3,
  single_gene = TRUE,
  chain = c("IGH", "IGK", "IGL", "TRB")
)

Arguments

Details

Function accepts a data.frame in AIRR format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'subject': The subject name

• 'v_call': V allele call(s) (in an IMGT format)

• 'd_call': D allele call(s) (in an IMGT format, only for heavy chains)

• 'j_call': J allele call(s) (in an IMGT format)

Value

A data.frame, in which each row is the haplotype inference summary of a gene from the column selected in toHap_col.

The output containes the following columns:

• subject: the subject name.

• gene: the gene name.

• Anchor gene allele 1: the haplotype inference for chromosome one. The column name is the anchor gene with the first allele.

• Anchor gene allele 2: the haplotype inference for chromosome two. The column name is the anchor gene with the second allele.

• alleles: allele calls for the gene.

• proirs_row: priors based on relative allele usage of the anchor gene.

• proirs_col: priors based on relative allele usage of the inferred gene.

• counts1: the appereance count on each chromosome of the first allele from alleles, the counts are seperated by a comma.

• k1: the Bayesian factor value for the first allele (from alleles) inference.

• counts2: the appereance count on each chromosome of the second allele from alleles, the counts are seperated by a comma.

• k2: the Bayesian factor value for the second allele (from alleles) inference.

• counts3: the appereance count on each chromosome of the third allele from alleles, the counts are seperated by a comma.

• k3: the Bayesian factor value for the third allele (from alleles) inference.

• counts4: the appereance count on each chromosome of the fourth allele from alleles, the counts are seperated by a comma.

• k4: the Bayesian factor value for the fourth allele (from alleles) inference.

Examples

# Load example data and germlines
data(samples_db, HVGERM, HDGERM)

# Selecting a single individual
clip_db = samples_db[samples_db$subject=='I5', ]

# Infering haplotype
haplo_db = createFullHaplotype(clip_db,toHap_col=c('v_call','d_call'),
hapBy_col='j_call',hapBy='IGHJ6',toHap_GERM=c(HVGERM,HDGERM))

Create haplotype table

Description

createHaplotypeTable Haplotype of a specific gene

Usage

createHaplotypeTable(
  df,
  HapByPriors = c(0.5, 0.5),
  toHapByCol = TRUE,
  toHapPriors = c(0.5, 0.5)
)

Arguments

Value

data frame with chromosomal associasions of alleles of a specific gene

Graphical output of single chromosome deletions

Description

The deletionHeatmap function generates a graphical output of the single chromosome deletions in multiple samples.

Usage

deletionHeatmap(
  hap_table,
  chain = c("IGH", "IGK", "IGL", "TRB", "TRA"),
  kThreshDel = 3,
  genes_order = NULL,
  html_output = FALSE
)

Arguments

Details

A data.frame created by createFullHaplotype.

Value

A single chromosome deletion visualization.

Examples

# Plotting single choromosme deletion from haplotype inference
deletionHeatmap(samplesHaplotype)

Double chromosome deletion by relative gene usage

Description

The deletionsByBinom function inferes double chromosome deletion events by relative gene usage.

Usage

deletionsByBinom(
  clip_db,
  chain = c("IGH", "IGK", "IGL"),
  nonReliable_Vgenes = c(),
  genes_order = NULL
)

Arguments

Details

The function accepts a data.frame in AIRR format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'subject': The subject name

• 'v_call': V allele call(s) (in an IMGT format)

• 'd_call': D allele call(s) (in an IMGT format, only for heavy chains)

• 'j_call': J allele call(s) (in an IMGT format)

Value

A data.frame, in which each row is the double chomosome deletion inference of a gene.

The output containes the following columns:

• subject: the subject name.

• gene: the gene call

• frac: the relative gene usage of the gene

• cutoff: the the cutoff of for the binomial test

• pval: the p-value of the binomial test

• deletion: if a double chromosome deletion event of a gene occured.

Examples

# Load example data and germlines
data(samples_db)

# Selecting a single individual
clip_db = samples_db[samples_db$subject=='I5', ]
# Infering haplotype
del_binom_df = deletionsByBinom(clip_db)
head(del_binom_df)

Single chromosomal D or J gene deletions inferred by the V pooled method

Description

The deletionsByVpooled function inferes single chromosomal deletion for D and J gene .

Usage

deletionsByVpooled(
  clip_db,
  chain = c("IGH", "IGK", "IGL"),
  deletion_col = c("d_call", "j_call"),
  count_thresh = 50,
  deleted_genes = "",
  min_minor_fraction = 0.3,
  kThreshDel = 3,
  nonReliable_Vgenes = c()
)

Arguments

Details

The function accepts a data.frame in AIRR format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'subject': The subject name

• 'v_call': V allele call(s) (in an IMGT format)

• 'd_call': D allele call(s) (in an IMGT format, only for heavy chains)

• 'j_call': J allele call(s) (in an IMGT format)

Value

A data.frame, in which each row is the single chomosome deletion inference of a gene.

The output containes the following columns:

• subject: the subject name.

• gene: the gene call

• deletion: chromosome deletions inferred. Encoded 1 for deletion and 0 for no deletion.

• k: the Bayesian factor value for the deletion inference.

• counts: the appereance count of the gene on each chromosome, the counts are seperated by a comma.

Examples

data(samples_db)

# Infering V pooled deletions
del_db <- deletionsByVpooled(samples_db)
head(del_db)

Double chromosome deletion by relative gene usage

Description

The geneUsage function calculates the relative gene usage.

Usage

geneUsage(
  clip_db,
  chain = c("IGH", "IGK", "IGL", "TRB"),
  genes_order = NULL,
  rmPseudo = TRUE
)

Arguments

Details

The function accepts a data.frame in AIRR format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'subject': The subject name

• 'v_call': V allele call(s) (in an IMGT format)

• 'd_call': D allele call(s) (in an IMGT format, only for heavy chains)

• 'j_call': J allele call(s) (in an IMGT format)

Value

A data.frame, in which each row is the relative gene usage value per individual.

The output containes the following columns:

• subject: the subject name.

• gene: the gene call

• frac: the relative gene usage of the gene

Hierarchical clustering of haplotypes graphical output

Description

The hapDendo function generates a graphical output of an hierarchical clustering based on the Jaccard distance between multiple samples' haplotypes.

Usage

hapDendo(
  hap_table,
  chain = c("IGH", "IGK", "IGL", "TRB", "TRA"),
  genes_order = NULL,
  removeIGH = TRUE,
  mark_low_lk = TRUE,
  lk_cutoff = 1
)

Arguments

Details

A data.frame created by createFullHaplotype.

Value

A multitple samples visualization of the distances between haplotypes.

Examples

# Plotting haplotype hierarchical clustering based on the Jaccard distance
hapDendo(samplesHaplotype)

Graphical output of alleles division by chromosome

Description

The hapHeatmap function generates a graphical output of the alleles per gene in multiple samples.

Usage

hapHeatmap(
  hap_table,
  chain = c("IGH", "IGK", "IGL", "TRB", "TRA"),
  genes_order = NULL,
  removeIGH = TRUE,
  lk_cutoff = 1,
  mark_low_lk = TRUE,
  size_annot = 1.5,
  color_y = NULL,
  order_subject = NULL,
  file = NULL,
  size_text = NULL,
  ylabel_size = 1
)

Arguments

Details

A data.frame created by createFullHaplotype.

Value

A list with the following:

• 'p': heat-map visualization of the haplotype inference for multiple samples.

• 'width': Optimal width value for rendering plot.

• 'height': Optimal width value for rendering plot.

When a file is supplied the graph is also rendered to pdf.

Examples

# Plotting haplotpe heatmap
p <- hapHeatmap(samplesHaplotype)
p$p

Detect non reliable gene assignment

Description

nonReliableVGenes Takes a data.frame in AIRR format and detect non reliable IGHV genes. A non reliable gene is when the ratio of the multiple assignments with a gene is below the threshold.

Usage

nonReliableVGenes(clip_db, thresh = 0.9, appearance = 0.01)

Arguments

Details

The function accepts a data.frame in AIRR format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'subject': subject names

• 'v_call': V allele call(s) (in an IMGT format)

Value

a nested list of non reliable genes for all subject.

Examples

# Example IGHV call data frame
clip_db <- data.frame(subject=rep('S1',6),
v_call=c('IGHV1-69*01','IGHV1-69*01','IGHV1-69*01,IGHV1-69*02',
'IGHV4-59*01,IGHV4-61*01','IGHV4-59*01,IGHV4-31*02','IGHV4-59*01'))
# Detect non reliable genes
nonReliableVGenes(clip_db)

Graphical output of double chromosome deletions

Description

The plotDeletionsByBinom function generates a graphical output of the double chromosome deletions in multiple samples.

Usage

plotDeletionsByBinom(
  GENE.usage.df,
  chain = c("IGH", "IGK", "IGL", "TRB", "TRA"),
  genes.low.cer = c("IGHV3-43", "IGHV3-20"),
  genes.dup = c("IGHD4-11", "IGHD5-18"),
  genes_order = NULL
)

Arguments

Details

A data.frame created by binom_test_deletion.

Value

A double chromosome deletion visualization.

Examples

# Load example data and germlines
data(samples_db)

# Infering haplotype
deletions_db = deletionsByBinom(samples_db);
plotDeletionsByBinom(deletions_db)

Graphical output for single chromosome D or J gene deletions according to V pooled method

Description

The plotDeletionsByVpooled function generates a graphical output for single chromosome D or J gene deletions (for heavy chain only).

Usage

plotDeletionsByVpooled(
  del.df,
  chain = c("IGH", "IGK", "IGL", "TRB", "TRA"),
  K_ranges = c(3, 7)
)

Arguments

Details

A data.frame created by deletionsByVpooled.

Value

A single chromosome deletion visualization.

Examples

# Load example data and germlines
data(samples_db)
del_db <- deletionsByVpooled(samples_db)
plotDeletionsByVpooled(del_db)

Graphical output of an inferred haplotype

Description

The plotHaplotype functions visualizes an inferred haplotype.

Usage

plotHaplotype(
  hap_table,
  html_output = FALSE,
  genes_order = NULL,
  text_size = 14,
  removeIGH = TRUE,
  plotYaxis = TRUE,
  chain = c("IGH", "IGK", "IGL", "TRB"),
  dir,
  width = 15,
  height = 20
)

Arguments

Details

A data.frame in a haplotype format created by createFullHaplotype function.

Value

A haplotype map visualization. If more than one subject is visualized, a pdf is created. If html_output is TRUE, a folder named html_output is created with individual graphs.

Examples

# Selecting a single individual from the haplotype samples data
haplo_db = samplesHaplotype[samplesHaplotype$subject=='I5', ]

# plot haplotype
plotHaplotype(haplo_db)

The RAbHIT package

Description

The rabhit package provides a robust novel method for determining antibody heavy and light chain haplotypes by adapting a Bayesian framework. The key functions in rabhit, broken down by topic, are described below.

Haplotype and deletions inference

rabhit provides tools to infer haplotypes based on given anchor genes, deletion detection based on relative gene usage, pooling v genes, and a single anchor gene.

• createFullHaplotype: Haplotypes inference and single chromosome deletions based on an anchor gene.

• deletionsByVpooled: Single chromosomal deletion detection by pooling V genes.

• deletionsByBinom: Double chromosomal deletion detection by relative gene usage.

• geneUsage: Relative gene usage.

• nonReliableVGenes: Non reliable gene assignment detection.

Haplotype and deletions visualization

Functions for visualization of the inferred haplotypes and deletions

• plotHaplotype: Haplotype inference map.

• deletionHeatmap: Single chromosome deletions heatmap.

• hapHeatmap: Chromosome comparison of multiple samples.

• hapDendo: Hierarchical clustering of multiple haplotypes based on Jaccard distance.

• plotDeletionsByVpooled: V pooled based single chromosome deletions heatmap.

• plotDeletionsByBinom: Double chromosome deletions heatmap.

References

1. Gidoni, M., Snir, O., Peres, A., Polak, P., Lindeman, I., Mikocziova, I., . . . Yaari, G. (2019). Mosaic deletion patterns of the human antibody heavy chain gene locus shown by Bayesian haplotyping. Nature Communications, 10(1). doi:10.1038/s41467-019-08489-3

Read a Change-O tab-delimited database file

Description

readHaplotypeDb reads a tab-delimited haplotype file created by a createFullHaplotype into a data.frame. Based on readChangeoDb function from alakazam.

Usage

readHaplotypeDb(file)

Arguments

Value

A data.frame of the haplotype file. Columns will be imported as is, except for the following columns which will be explicitly converted into character values:

• alleles

• subject

Example haplotype inference results

Description

A data.frame of example haplotype infrence results from createFullHaplotype after double chromosome deletion inference via deletionsByBinom and non reliable V genes detection via nonReliableVGenes. Source data is a colletion of IGH human naive b-cell repertiore data from five individuals (see references). Overall, the data set includes 6 samples. A single individual has two samples (Individual I5), one is short read sequences from BIOMED-2 protocol primers for framework 2 region (The sample is annotated I5_FR2).

Usage

samplesHaplotype

Format

A data.frame, in which each row is the haplotype inference summary of a gene of an individual, from the column selected to prefrom the haplptype infrence on.

References

Gidoni, Moriah, et al. Mosaic deletion patterns of the human antibody heavy chain gene locus shown by Bayesian haplotyping. Nature Communications. 10.1 (2019): 628.

See Also

See createFullHaplotype for detailed column descriptions.

Example IGH human naive b-cell repertiore

Description

A data.frame of example IGH human naive b-cell repertiore data from five individuals (see references). Overall, the data set includes 6 samples. A single individual has two samples (Individual I5), one is short read sequences from BIOMED-2 protocol primers for framework 2 region (The sample is annotated I5_FR2).

Usage

samples_db

Format

A data.frame in Change-O format (https://changeo.readthedocs.io/en/stable/standard.html) containing the following columns:

• 'SUBJECT': subject names

• 'V_CALL': V allele call(s) (in an IMGT format)

• 'D_CALL': D allele call(s) (in an IMGT format, only for heavy chains)

• 'J_CALL': J allele call(s) (in an IMGT format)

References

Gidoni, Moriah, et al. Mosaic deletion patterns of the human antibody heavy chain gene locus shown by Bayesian haplotyping. Nature Communications. 10.1 (2019): 628.