Package 'blaseRtools'

Description

This is a repository of R tools for Single Cell RNA seq and other lab data analysis functions.

Author(s)

Maintainer: Brad Blaser [email protected] (ORCID)

See Also

Useful links:

• https://github.com/blaserlab/blaseRtools

• Report bugs at https://github.com/blaserlab/blaseRtools/issues

Description

Operators

Usage

... %notin% NA

Description

Add predefined sample-level cell metadata to cell data sets while importing

Usage

add_cds_factor_columns(cds, columns_to_add)

Arguments

Description

Add significance annotations to an existing ggplot

Usage

add_sig_annotations(
  plot,
  p_table,
  y_npc,
  group1_col = "group1",
  group2_col = "group2",
  label_col = "significance",
  x_levels = NULL,
  facet_cols = NULL,
  draw_brackets = TRUE,
  bracket_tip_npc = 0.015,
  bracket_margin_npc = 0.02,
  text_size_pt = NULL,
  star_y_npc_offset = 0.01,
  text_family = NULL,
  text_face = NULL,
  text_colour = "black",
  bracket_colour = "black",
  bracket_linewidth = 0.4,
  bracket_linetype = 1,
  bracket_lineend = "round",
  vjust = 0
)

Arguments

Value

A ggplot object with annotation layers appended.

Description

Similar to aggregate. Splits the matrix into groups as specified by groupings, which can be one or more variables. Aggregation function will be applied to all columns in data, or as specified in formula. Warning: groupings will be made dense if it is sparse, though data will not.

Usage

## S3 method for class 'Matrix'
aggregate(x, groupings = NULL, form = NULL, fun = "sum", ...)

Arguments

Details

aggregate.Matrix uses its own implementations of functions and should be passed a string in the fun argument.

Value

A sparse Matrix. The rownames correspond to the values of the groupings or the interactions of groupings joined by a _.

There is an attribute crosswalk that includes the groupings as a data frame. This is necessary because it is not possible to include character or data frame groupings in a sparse Matrix. If needed, one can cbind(attr(x,"crosswalk"),x) to combine the groupings and the aggregates.

See Also

summarise

summarise

aggregate

Description

An instance of this class is best created by calling "bb_parseape()" on a genebank or APE-formatted file. That function will parse the file, correctly format the sections and place them in the slots of the Ape Object. Technically only "LOCUS" is a required slot for the Ape object, however there is no point without having "ORIGIN" (sequence data), and so bb_parseape() will fail without an "ORIGIN" section. Other slots are optional. Additional slots will be ignored by the constructor function. DNA sequence will be stored in a DNAStringSet object and features in a GRanges object. See https://www.ncbi.nlm.nih.gov/Sitemap/samplerecord.html for genebank file specification.

Slots

LOCUS

The LOCUS line of the genebank formatted as a character string.

DEFINITION

The DEFINITION line of the genebank file formatted as a character string.

ACCESSION

The ACCESSION section of the genebank file formatted as a character string.

VERSION

The VERSION section of the genebank file formatted as a character string.

SOURCE

The SOURCE section of the genebank file formatted as a character string.

COMMENT

The COMMENT section of the genebank file formatted as a character string.

FEATURES

The FEATURES section of the genebank file formatted as a character string. Created internally from the GRanges object. Caution: some FEATURE attributes may be lost in conversion.

ORIGIN

The DNA sequence

end_of_file

The end of the file signal.

dna_biostring

The entire ORIGIN sequence formatted as a DNAStringSet of length 1.

granges

Genebank features formatted as a GRanges object.

Description

Get the DNASringSet Slot from an Ape Object

Usage

Ape.DNA(ape)

Description

Save an Ape Instance as a Fasta File

Usage

Ape.fasta(ape, feature = NULL, out)

Arguments

Description

For the supplied Ape object, run FIMO to identify putative transcription factor binding sites in a DNA subsequence.

Usage

Ape.fimo(ape, fimo_feature, out = NULL)

Arguments

Description

Get the GRanges Slot from an Ape Object

Usage

Ape.granges(ape)

Description

Save an Ape Instance as a Genebank Format File

Usage

Ape.save(ape, out)

Arguments

Description

Set the FEATURES Slot of a GRanges Object

Usage

Ape.setFeatures(ape, gr)

Arguments

Description

Convert objects to Monocle3 cell_data_set objects

Usage

as.cell_data_set(x, ...)

## S3 method for class 'Seurat'
as.cell_data_set(
  x,
  assay = DefaultAssay(object = x),
  reductions = AssociatedDimReducs(object = x, assay = assay),
  default.reduction = DefaultDimReduc(object = x, assay = assay),
  graph = paste0(assay, "_snn"),
  group.by = NULL,
  ...
)

Arguments

Details

The Seurat method utilizes as.SingleCellExperiment to transfer over expression and cell-level metadata. The following additional information is also transferred over:

• Cell emebeddings are transferred over to the reducedDims slot. Dimensional reduction names are converted to upper-case (eg. “umap” to “UMAP”) to match Monocle 3 style

• Feature loadings are transfered to cds@reduce_dim_aux$gene_loadings if present. NOTE: only the feature loadings of the last dimensional reduction are transferred over

• Standard deviations are added to cds@reduce_dim_aux$prop_var_expl if present. NOTE: only the standard deviations of the last dimensional reduction are transferred over

• Clustering information is transferred over in the following manner: if cell-level metadata entries “monocle3_clusters” and “monocle3_partitions” exist, then these will be set as the clusters and partitions, with no nearest neighbor graph being added to the object; otherwise, Seurat's nearest-neighbor graph will be converted to an igraph object and added to the cell_data_set object along with Seurat's clusters. No partition information is added when using Seurat's clsuters

Value

A cell_data_set object

See Also

as.SingleCellExperiment

Description

Convert single cell experiment to Seurat

Usage

## S3 method for class 'cell_data_set'
as.Seurat(
  x,
  counts = "counts",
  data = NULL,
  assay = "RNA",
  project = "cell_data_set",
  loadings = NULL,
  clusters = NULL,
  ...
)

Arguments

Details

The cell_data_set method for as.Seurat utilizes the SingleCellExperiment method of as.Seurat to handle moving over expression data, cell embeddings, and cell-level metadata. The following additional information will also be transfered over:

• Feature loadings from cds@reduce_dim_aux$gene_loadings will be added to the dimensional reduction specified by loadings or the name of the first dimensional reduction that contains "pca" (case-insensitive) if loadings is not set

• Monocle 3 clustering will be set as the default identity class. In addition, the Monocle 3 clustering will be added to cell-level metadata as “monocle3_clusters”, if present

• Monocle 3 partitions will be added to cell-level metadata as “monocle3_partitions”, if present

• Monocle 3 pseudotime calculations will be added to “monocle3_pseudotime”, if present

• The nearest-neighbor graph, if present, will be converted to a Graph object, and stored as “assay_monocle3_graph”

See Also

as.Seurat.SingleCellExperiment

Description

Generates a matrix of counts aggregated by gene and/or cell group.

Usage

bb_aggregate(
  obj,
  assay = "RNA",
  experiment_type = "Gene Expression",
  gene_group_df = NULL,
  cell_group_df = NULL,
  norm_method = c("log", "binary", "size_only"),
  pseudocount = 1,
  scale_agg_values = TRUE,
  max_agg_value = 3,
  min_agg_value = -3,
  binary_min = 0,
  exclude.na = TRUE
)

Arguments

Details

The best way to group genes or cells is by using bb_*meta and then select cell_id or feature_id plus one metadata column with your group labels.

Value

A dense or sparse matrix.

See Also

cli_div, cli_alert normalized_counts, my.aggregate.Matrix character(0)

Description

Align a CDS object according to a metadata variable

Usage

bb_align(cds, align_by, n_cores = 8)

Arguments

Value

A modified cell data set object with aligned dimensions and new metadata columns holding prealignment umap coordinates.

Description

Annotate a Plot using NPC Coordinates

Usage

bb_annotate_npc(label, x, y, ...)

Arguments

Description

Will copy and rename the files and generate two files: "blinding_key.csv" with the original and blinded file names, and "scoresheet.csv" with just the blinded filenames. Add columns as needed to scoresheet, for example, runx_count. Then run bb_unblind to rejoin scoresheet to the key and generate an unblinded result file.

Usage

bb_blind_images(analysis_file, file_column, output_dir)

Arguments

Value

nothing

Description

This function does several things. It removes ranges with non-standard chromosomes and drops their levels. It will optionally set the genome to the user-provided value. Typically we would use "hg38" or "danRer11". This is the exported version because it is so useful.

Usage

bb_buff_granges(x, gen)

Arguments

Value

A GRanges object

Description

Use this function to transfer cell labels from one single cell dataset to another. If a cds is provided for either reference or query, it is converted to a seurat object and the labels are transferred to the query by anchor finding. The assignments will be in the form of a new cds column with name predicted."column name from reference". This should be unique on the first application of the function to a query dataset. However, if running queries against more than 1 reference data set it is possible that you will unintentionally generate the same column name which would overwrite the first assignment column. The function checks for this and aborts with the recommendation to supply a unique id to the unique_id parameter.

Usage

bb_cds_anno(query_cds, ref, transfer_col, unique_id = NULL)

Arguments

Value

a CDS with two new cell metadata columns

See Also

cli_abort components, SCTransform, RunPCA, RunUMAP, FindTransferAnchors, MapQuery exprs character(0) reexports select, mutate-joins as_tibble

Description

Supply a cds subset and aggregation values and get a heatmap. Plot the return value using cowplot::plot_grid(). This function wraps several complicated functions from ComplexHeatmap and tries to apply default values for the most common use case: plotting top markers from a cds with cells grouped arbitrarily (usually by cluster of some type). This function provides the option to aggregate by genes as well in order to plot gene modules. For cell and gene aggregation, provide a name (in the form of a string) of the corresponding metadata column and the aggregation will be performed. There are many aesthetic parameters for this function and even more available for the internal ComplexHeatmap::Heatmap() function. The best way to adjust parameters not provided here is to scroll the popup box that appears when you type ComplexHeatmap::Heatmap() in RStudio and pass them in via the ellipsis. The default is to put genes in columns and cells in rows. You can flip this behavior by setting flip_axis = TRUE. This will require adjustment of some aesthetic parameters. Complex annotations (beyond labeling cherry-picked genes) are not currently supported.

Usage

bb_cds_heatmap(
  cds_subset,
  cellmeta_col = NULL,
  rowmeta_col = NULL,
  heatmap_highlights = NULL,
  three_colors = c("blue4", "ivory", "red3"),
  flip_axis = FALSE,
  name = NULL,
  heatmap_legend_param = list(title_gp = gpar(fontface = "plain", fontsize = 9),
    grid_width = unit(0.14, "in"), labels_gp = gpar(fontsize = 8)),
  row_dend_width = unit(5, "mm"),
  column_dend_height = unit(5, "mm"),
  column_dend_side = "bottom",
  show_row_names = T,
  row_names_gp = gpar(fontsize = 9),
  show_column_names = F,
  row_dend_gp = gpar(lwd = 0.5),
  column_dend_gp = gpar(lwd = 0.5),
  row_title = NULL,
  column_title = NULL,
  padding = 1.5,
  labels_rot = 45,
  ...
)

Arguments

Value

A complex heatmap in the form of a gtree.

Description

Use this function to identify ligand/receptor pairs expressed by cell clusters in human, mouse or zebrafish single cell data. A CellChat object is generated which can be used to visualize these connections using bb_cellchat_heatmap or other tools from package CellChat.

Usage

bb_cellchat(
  cds,
  group_var,
  n_cores = 12,
  species = c("human", "mouse", "zebrafish"),
  min_cells = 10,
  prob_type = c("triMean", "truncatedMean", "median"),
  prob_trim = NULL,
  project = TRUE,
  pop_size_arg = TRUE,
  ask = TRUE
)

Arguments

Details

see github::sqjin/CellChat

Value

A CellChat object

See Also

normalized_counts mutate-joins,pull tibble createCellChat,CellChatDB.human,CellChatDB.mouse,character(0),subsetData,identifyOverExpressedGenes,identifyOverExpressedInteractions,projectData,computeCommunProb,filterCommunication,computeCommunProbPathway,aggregateNet character(0) plan

Description

This will generate heatmap from a CellChat object using ComplexHeatmap::Heatmap. Options are provided to filter for sender and receiver cells, to generate simple marginal annotations and for aesthetic control.

Usage

bb_cellchat_heatmap(
  object,
  source_filter = NULL,
  target_filter = NULL,
  interaction_filter = NULL,
  interaction_threshold = 0,
  colors = c("transparent", "red3"),
  rowanno = c(NULL, "Annotation", "Pathway"),
  rowanno_colors = NULL,
  colanno = c(NULL, "Source", "Target"),
  colanno_colors = NULL,
  pval_filter = 0.05,
  heatmap_name = "Interaction\nScore",
  heatmap_show_row_dend = TRUE,
  heatmap_row_dend_width = unit(5, "mm"),
  heatmap_show_column_dend = TRUE,
  heatmap_column_dend_height = unit(5, "mm"),
  heatmap_row_names_gp = gpar(fontsize = 10),
  heatmap_column_names_gp = gpar(fontsize = 10),
  heatmap_column_names_rot = 90,
  heatmap_column_title = NULL,
  heatmap_column_title_gp = gpar(fontsize = 12, fontface = "bold"),
  col_anno_name_gp = gpar(fontsize = 10, fontface = "bold"),
  row_anno_name_gp = gpar(fontsize = 10, fontface = "bold"),
  return_value = c("heatmap", "plot", "matrix")
)

Arguments

Details

see github::sqjin/CellChat

Value

a heatmap as a grid object; plot using cowplot::plot_grid

See Also

subsetCommunication as_tibble,c("tibble", "tibble"),rownames filter,mutate,select,mutate-joins,group_by,summarise pivot_wider rowAnnotation,columnAnnotation,draw-dispatch,Heatmap colorRamp2 grid.grab

Description

Take a cell_data_set object or a Seurat object and return the cell metadata in the form of a tibble. The unique cell identifier column is labeled cell_id by default. Prior versions of this function would only accept a cell_data_set. The input argument has been changed from cds to obj to reflect the fact that Seurat objects are now also accepted.

Usage

bb_cellmeta(obj, row_name = "cell_id", cds = NULL)

Arguments

Details

If a value is supplied for cds, a warning will be issued and the function will pass the value of cds to obj.

Value

A tibble

Description

Requires a cds with an alt experiment established. Use bb_split_citeseq to generate this and to normalize binding data using the CLR method. Returns a ggplot.

Usage

bb_cite_umap(
  cds,
  antibody,
  assay = "CLR_counts",
  cell_size = 1,
  alpha = 1,
  alt_dim_x = NULL,
  alt_dim_y = NULL,
  plot_title = NULL,
  color_legend_title = NULL,
  order = TRUE,
  rescale = NULL,
  ncol = NULL
)

Arguments

Value

a ggplot

See Also

reducedDims

Description

Use this function to determine the differential representation of cells in clusters. It will determine fold change in a single experimental class over a single control or reference class. This value is normalized to the number of cells captured in all clusters from the class. Significance is determined using Fisher's exact test. This test may overestimate significance in large data sets. In this case, bb_cluster_representation2 may be more robust.

Usage

bb_cluster_representation(
  cds,
  cluster_var,
  class_var,
  experimental_class,
  control_class,
  pseudocount = 1,
  return_value = c("table", "plot")
)

Arguments

Value

A ggplot or a table of data for plotting

Description

Use this function to determine the differential representation of cells in clusters. It uses a regression method to determine fold change between groups of biological samples. It can only compare two sample groups, e.g. control vs experimental at this point. See parameter descriptions for how to identify these properly.

Usage

bb_cluster_representation2(
  obj,
  sample_var,
  cluster_var,
  comparison_var,
  comparison_levels = NULL,
  color_pal = c("red3", "blue4"),
  sig_val = c("FDR", "PValue"),
  return_val = c("plot", "data")
)

Arguments

Details

DETAILS

Value

OUTPUT_DESCRIPTION

Source

http://bioconductor.org/books/3.13/OSCA.multisample/differential-abundance.html

Description

This function takes a cell data set object and a cell metadata variable as input. The latter is specified as a named list. The name is the cell metadata variable name and each element must be a character vector of length 2 specifying the levels of that variable to compare. For example, for a column named "genotype" with levels "WT", "heterozygote", and "homozygote", you would compare differential abundance between WT and homozygote like this: list(genotype = c("WT", "homozygote")). Cells with these classifications will get a differential abundance score. Negative values will be assigned to the first element (WT in this example) and positive values to the second element (homozygote in this example). Cells with other values for this variable ("heterozygote" in this example) will get an NA. Multiple comparisons can be performed for different levels within the same variable or for different variables altogether by providing additional elements to the list. E.g. list(genotype = c("WT", "heterozygote), genotype = c("WT", "homozygote")). For each comparison, a new cell metadata column will be added in the form of da_score_name_level1_level2. E.g. da_score_genotype_WT_homozygote.

Usage

bb_daseq(obj, comparison_list, sample_var = "sample")

Arguments

Value

a cell data set

See Also

cli_div, cli_alert map2, reexports, pmap, reduce filter, pull, mutate-joins, select, join_by, rename reducedDims getDAcells as_tibble

Description

Use doubletfinder to model and mark doublets

Usage

bb_doubletfinder(cds, doublet_prediction, qc_table, ncores = 1)

Arguments

Value

A tibble of low- and high-confidence doublet calls by barcode

Description

Use this to extract gene sets from the MSIGDB. Most gene sets are known by "STANDARD_NAME". You can filter the gene set list by supplying a named filter list to the bb_extract_msig function. The name of each list element should be one of the metadata column names and the list element contents should be the values to filter for. Filtering works in an additive way, meaning if you supply a filter list with two elements it will extract gene sets passing filters 1 AND 2.

Usage

bb_extract_msig(
  filter_list = NULL,
  return_form = c("id_list", "name_list", "tibble")
)

Arguments

Value

Gene set as a list or tibble.

Description

This will only work if the backslashes have already been escaped to double backslashes. For example, readr automatically does this when reading in windows file paths from csv files. If you are using as a standalone function, run this in the terminal: scan(what = "character, n = 1), paste the unmodified filepath at the blank line, and copy the modified file path as the argument, x.

Usage

bb_fix_file_path(x)

Arguments

Value

A linux-compatible filepath

Description

Often you will wish to package a Signac object or share the object to someone who does not have access to the same directories as you. This is a problem because one of the internal sub-objects, the Fragment object, holds a file path to a directory containing an atac fragments file and its .tbi intex. Often this file will only be accessbile to you. This function allows you to replace the fragments file path that is held internally within the Signac object. Just copy the files to a shared location and provide that file location to the function.

For Signac objects with multiple internal Fragments objects, provide vector of file paths.

For packaged fragments files, use system.file or fs::path_package to access this, usually from extdata.

Usage

bb_fragment_replacement(obj, new_paths)

Arguments

Value

A modified Signac object.

See Also

c("Cells.Fragment", "CountFragments", "CreateFragmentObject", "FilterCells", "Fragment-class", "Fragments", "Fragments", "SplitFragments", "UpdatePath", "ValidateCells", "ValidateFragments", "ValidateHash", "head.Fragment", "subset.Fragment"), CreateFragmentObject cli_abort file_access map2

Description

Make a dotplot of gene expression by cell population

Usage

bb_gene_dotplot(
  cds,
  markers,
  group_cells_by,
  reduction_method = "UMAP",
  norm_method = c("size_log", "log_only"),
  scale_expression_by_gene = FALSE,
  lower_threshold = 0,
  max.size = 10,
  group_ordering = "bicluster",
  gene_ordering = NULL,
  pseudocount = 1,
  scale_max = 3,
  scale_min = -3,
  colorscale_name = NULL,
  sizescale_name = NULL,
  ...
)

Arguments

Value

A ggplot

Description

Based on Monocle3's gene module functions. Implemented with default values. Will convert a Seurat object to a cell data set using SeuratWrappers and then calculate modules. The function returns an object of the same type.

Usage

bb_gene_modules(obj, n_cores = 8, cds = NULL)

Arguments

Details

see https://cole-trapnell-lab.github.io/monocle3/docs/differential/#gene-modules

Value

An object of the same type: Seurat or cell_data_set

See Also

graph_test, find_gene_modules rename, mutate-joins, mutate, select fct_shift

Description

Plot expression of a gene or genes in pseudotime.

Usage

bb_gene_pseudotime(
  cds_subset,
  min_expr = NULL,
  cell_size = 0.75,
  nrow = NULL,
  ncol = 1,
  panel_order = NULL,
  color_cells_by = "pseudotime",
  trend_formula = "~ splines::ns(pseudotime, df=3)",
  label_by_short_name = TRUE,
  vertical_jitter = NULL,
  horizontal_jitter = NULL
)

Arguments

Value

A ggplot

Description

Takes in a Seurat or cell_dat_set object, extracts UMAP dimensions and gene expression values. For Seurat, default assay is "RNA"; can be changed to if necessary. For cell_data_set, the assay parameter does nothing; the function extracts log and size-factor normalized counts which are similar but not identical to the Seurat "RNA" assay. If a vector of genes is supplied to gene_or_genes, a faceted plot will be generated. If a dataframe is supplied, an aggregated plot will be generated with a facet for each gene group. The dataframe must be of 2 colums: the first containing feature ids and the second containing grouping information. This is best generated using bb_rowmeta.

Usage

bb_gene_umap(
  obj,
  gene_or_genes,
  assay = "RNA",
  order = TRUE,
  cell_size = 1,
  alpha = 1,
  ncol = NULL,
  plot_title = NULL,
  color_legend_title = "Expression",
  max_expr_val = NULL,
  alt_dim_x = NULL,
  alt_dim_y = NULL,
  rasterize = FALSE,
  raster_dpi = 300,
  cds = NULL
)

Arguments

Value

A ggplot

See Also

normalized_counts as_tibble pivot_longer mutate-joins, mutate, select, arrange ggplot, aes, geom_point, scale_colour_viridis_d, labs, facet_wrap, vars, theme, margin

Description

Make a plot of gene expression in UMAP form

Usage

bb_gene_violinplot(
  cds,
  variable,
  genes_to_plot,
  experiment_type = "Gene Expression",
  pseudocount = 1,
  include_jitter = FALSE,
  ytitle = "Expression",
  plot_title = NULL,
  rows = 1,
  show_x_label = TRUE,
  legend_pos = "none",
  comparison_list = NULL,
  palette = NULL,
  violin_alpha = 1,
  jitter_alpha = 1,
  jitter_color = "black",
  jitter_fill = "transparent",
  jitter_size = 0.5,
  facet_scales = "fixed",
  order_genes = TRUE,
  jitter_match = FALSE,
  rasterize = FALSE,
  raster_dpi = 300
)

Arguments

Value

A ggplot

Description

This is a very data-dense plot and is the recommended way for showing expression of single markers/genes by cell group. By default, this function will return an unfaceted ggplot with cell groups on the X axis and genes on the Y axis with dot size representing proportion of cells in the cell group expressing a gene and color scale representing per-cell expression.

But it also may be of interest to add aesthetic variables such as facets or additional color scales. There are two ways this function will facilitate that. First, you can supply a vector of cell groups to the cell_grouping argument and a the cells will be grouped by the composite value of these factors. Usually if you are doing this, you also will want to have access to the components of this composite variable to facet by. So you can supply "data" to the return_value argument to get a tibble. From there you can modify as necessary and generate a ggplot assigning aesthetics and scales as desired and using geom_point.

This function also supports visualizing citeseq data. These data should be allocated to an alternative experiment in the cds object. To show these data, set experiment_type to "Antibody Capture" or the name of the alternate experiment with citeseq data. The genes parameter should be the name assigned to the antibody derived tag. Expression threshold is particularly useful in this case because of the background binding observed with antibodies. The default is 0 and so by default any cell with more than 0 counts will be considered an expressor of that marker. This threshold is applied before scaling across markers. The best way to set this threshold is to visualize your markers of interest and isotypes with expression_threshold = 0 and scale_expr = FALSE. Then pick a threshold value based on the color scale and rerun with scale_expr either TRUE or FALSE.

Usage

bb_genebubbles(
  obj,
  genes,
  cell_grouping,
  experiment_type = "Gene Expression",
  scale_expr = TRUE,
  expression_threshold = 0,
  gene_ordering = c("bicluster", "as_supplied"),
  group_ordering = c("bicluster", "as_supplied"),
  return_value = c("plot", "data")
)

Arguments

Value

A ggplot or a tibble

Description

A function to find enriched go terms from a query list of gene names relative to a reference list of gene names.

Usage

bb_goenrichment(
  query,
  reference,
  group_pval = 0.01,
  go_db = c("org.Hs.eg.db", "org.Dr.eg.db", "org.Mm.eg.db")
)

Arguments

Value

A list of items including the enrichment results.

Description

Make a scatter plot of GO term associations

Usage

bb_goscatter(
  simMatrix,
  reducedTerms,
  size = "score",
  addLabel = TRUE,
  labelSize = 4
)

Arguments

Value

A ggplot

Description

A function to reduce go terms by semantic similarity

Usage

bb_gosummary(
  x,
  reduce_threshold = 0.8,
  go_db = c("org.Hs.eg.db", "org.Dr.eg.db", "org.Mm.eg.db")
)

Arguments

Value

A list of items for downstream plotting

Description

This is a thin wrapper around rtracklayer::import.bw. The purpose is to serve as a helper function for making Trace objects from bigwig files. This function converts the name of the numeric value metadata column to "coverage" for consistency with Trace objects made from Seurat/Signac objects. The numeric column it will look for is "score" by default. This appears to be the default name applied by import.bw and so this is the default value for this function. The option is provided to change it if necessary. The group variable must be supplied. Typically this will be something informative, like "ATAC" or some Histone mark that the data come from.

Usage

bb_import_bw(path, group, coverage_column = "score")

Arguments

Value

a granges object

Description

The function reads peaks in .bed file format produced by SEACR. Optionally add a group variable and value for later filtering or faceting when combined with other peak files.

The function reads peak files produced by MACS. Optionally add a group variable and value for later filtering or faceting when combined with other peak files.

Usage

bb_import_seacr_peaks(file, group_variable = NULL, group_value = NULL)

bb_import_macs_narrowpeaks(file, group_variable = NULL, group_value = NULL)

Arguments

Value

A GRanges object

A GRanges object

See Also

read_delim select, mutate makeGRangesFromDataFrame

read_delim select, mutate makeGRangesFromDataFrame

Description

This function reads a 10X Genomics H5 file and returns a cell_data_set or CDS. The option to split citeseq data .

Usage

bb_load_tenx_h5(filename, sample_metadata_tbl = NULL, split_citeseq = FALSE)

Arguments

Value

A cell data set.

Description

Converts a cell data set into a loupe file.

Usage

bb_loupeR(cds, output_dir = ".", output_file = "loupe")

Arguments

Value

Nothing

See Also

cli_abort, cli_alert create select, mutate, across, reexports starts_with reducedDims exprs compare create_loupe

Description

This function takes either a gene name or sequence coordinates and returns an ape object with DNA sequence from the the selected genome reference. You can choose from hg38 and GRCz11. Features are generated from ensembl GFF files. Features overlapping the query range (gene or sequence) are returned as a GRanges object and as features within the Ape object. The features included can optionally be filtered using the "include_type" argument to the function. Using the "additional_granges" argument, you can provide additional features not present in the standard gene model which will be added to the Ape object GRanges slot and to the features slot.

Usage

bb_make_ape_genomic(
  query,
  genome = c("hg38", "GRCz11"),
  extend_left = 0,
  extend_right = 0,
  include_type = c("ncRNA_gene", "rRNA", "exon", "pseudogene", "pseudogenic_transcript",
    "ncRNA", "gene", "CDS", "lnc_RNA", "mRNA", "three_prime_UTR", "five_prime_UTR",
    "unconfirmed_transcript", "scRNA", "C_gene_segment", "D_gene_segment",
    "J_gene_segment", "V_gene_segment", "miRNA", "tRNA", "snRNA", "snoRNA",
    "lincRNA_gene", "lncRNA_gene", "unconfirmed_transcript"),
  additional_granges = NULL
)

Arguments

Value

An APE object

Description

This function takes a specific ensembl transcript identifier, such as ENST00000348343.11, and gets the cDNA sequence from the corresponding transcriptomic reference. This is returned as an Ape object with the UTR's and the CDS annotated as features.

Usage

bb_make_ape_transcript(query, transcriptome = c("hg38", "GRCz11"))

Arguments

Details

DETAILS

Value

OUTPUT_DESCRIPTION

See Also

TxDb.Hsapiens.UCSC.hg38.knownGene BSgenome.Hsapiens.UCSC.hg38 org.Hs.eg.db character(0) BSgenome.Drerio.UCSC.danRer11 org.Dr.eg.db cli_abort matchPattern

Examples

## Not run: 
if(interactive()){
 #EXAMPLE1
 }

## End(Not run)

Description

Problem 1: Signac objects and GRanges made from bigwigs are large and it is computationally expensive to get data from them when tweaking plots. Problem 2: For the most part we like how Signac plots genomic coverage of track-like data and we would like to show bulk data in a similar way with a similar compuatational interface. The trace object is a small intermediate object that holds the minimal amount of data you need to make a coverage plot showing accessibility or binding to a specific genomic region. Then you can use the trace object to quickly and easily generate tracks as needed for your plot. These tracks are all ggplots and are easy to configure for legible graphics using add-on layers. There are options for displaying groups by color or facet which are built in with good graphical defaults. If these are not suitable, they can be changed post hoc like any other ggplot.

Usage

bb_makeTrace(
  obj,
  gene_to_plot = NULL,
  genome = c("hg38", "danRer11"),
  extend_left = 0,
  extend_right = 0,
  peaks = NULL,
  bulk_group_col = "group",
  bulk_coverage_col = "coverage",
  fill_in = FALSE,
  fixed_width = 100
)

Arguments

Description

@description This function merges replicate peaks GRanges objects into 1. @param peaks_gr A regular list of GRanges objects (Not a GRangesList). @return A GRanges object @export @import IRanges GenomicRanges

Usage

bb_merge_narrowpeaks(peaks_gr)

Description

Use this function to generate data for making TSS enrichment plots or other metafeature plots that are centered on a single genomic locus. This function returns the data you need for the plot. Use the tibble element that is returned to plot the enrichment plot and the matrix for the heatmap. The problem currently is that the binwidths for the enrichment plot need to be smaller than the binwidths for the heatmap to look good. If you use good binwidths for the enrichment plot, the heatmap will crash. So either reduce the size of the heatmap matrix before plotting that or rerun the function with a different bin size. This function allows sample names to be added, so several samples can be column-bound together for comparison. Each gene is normalized to its own outer flanks so this should account for differences in sequencing depth to some degree. You also have the option to include all possible TSS in the plot (i.e. including zeros) which you may want to do if comparing several samples. To do this, set select_hits to FALSE.

Usage

bb_metafeature(
  query,
  targets,
  select_hits = TRUE,
  width = 2000,
  binwidth = 10,
  sample_id = NULL
)

Arguments

Value

A list including a matrix and a tibble.

Description

This function wraps the monocle3 method for data projection and label transfer into a function. This function takes in reference and query cds objects plus a character vector of label column names and returns the query CDS with the new labels.

bb_monocle_anno and bb_monocle_project use similar inputs and methods and both return cds objects, but the cds objects are different. Both wrap around the monocle3 mehtod for data projection and label transfer.

bb_monocle_anno takes in reference and query cds objects plus a character vector of label column names to transfer. It returns the query CDS with new cell metadata contining one or more columns with the labels corresponding to the nearest neighbor in the reference data. A suffix is appended to the new column name in the result. By default this is "_ref" but it can be changed using the suffix parameter.

bb_monocle_project also takes in reference and query cds objects and character vector of label column names. In this case, a combined cds object is returned carrying the query and reference data projected int the reference data space. New column names are added indicating the reference and query data. The query cells are given the label of their nearest neighbor in the reference. These label column(s) are prepended with "merged_".

Importantly, for both of these to work, the reference and query genes must have a shared namespace.

Usage

bb_monocle_(cds_qry, cds_ref, labels, suffix, use_aligned)

bb_monocle_anno(cds_qry, cds_ref, labels, suffix = "_ref", use_aligned = TRUE)

bb_monocle_project(
  cds_qry,
  cds_ref,
  labels,
  suffix = "_ref",
  use_aligned = TRUE
)

Arguments

Details

see https://cole-trapnell-lab.github.io/monocle3/docs/projection/

Value

A cell data set

See Also

cli_abort, cli_alert sets estimate_size_factors, preprocess_cds, reduce_dimension, save_transform_models, load_transform_models, preprocess_transform, reduce_dimension_transform, transfer_cell_labels, fix_missing_cell_labels reducedDims path map, reduce SummarizedExperiment-class select, mutate-joins, join_by

Examples

## Not run: 
if(interactive()){
 #EXAMPLE1
 }

## End(Not run)

Description

Note: this function is deprecated in favor of bb_monocle_regression_better which returns log2FoldChange instead of estimates. The log2FoldChange value is the normalized_effect value returned from the monocle regression functions.

Usage

bb_monocle_regression(
  cds,
  gene_or_genes,
  stratification_variable = NULL,
  stratification_value = NULL,
  form,
  linking_function = "negbinomial"
)

Arguments

Value

A tibble containing the regression results.

Description

This function replaces bb_monocle_regression. It returns log2FoldChange instead of estimates. The log2FoldChange value is the normalized_effect value returned from the monocle regression functions.

Usage

bb_monocle_regression_better(
  cds,
  gene_or_genes,
  stratification_variable = NULL,
  stratification_value = NULL,
  form,
  linking_function = "negbinomial"
)

Arguments

Value

A tibble containing the regression results.

Description

This is the main function for reading file in genebank/ape/equivalent format and generating an instance of the Ape class. String manipulations are used to parse the input ape file. Biostrings and GRanges functions are called to generate DNAStringSet and GRanges objects to store sequence and feature data, respectively. The Ape constructor function is called internally at the end.

Usage

bb_parseape(input_file)

Arguments

Value

An Ape object

Description

Plots expression for one or more genes as a function of pseudotime

Usage

bb_plot_genes_in_pseudotime(
  cds,
  gene_or_genes,
  pseudotime_dim,
  min_expr = NULL,
  cell_size = 0.75,
  nrow = NULL,
  ncol = 1,
  panel_order = NULL,
  color_cells_by = pseudotime_dim,
  trend_formula_df = 3,
  label_by_short_name = TRUE,
  vertical_jitter = NULL,
  horizontal_jitter = NULL,
  legend_title = NULL
)

Arguments

Value

a ggplot2 plot object

Description

Will generate a ggplot from the colData of a SummarizedHeatmap object. Typically this will be placed at the top or bottom of a plot. If flipped, use the side argument to put the colData on the left or right.

Usage

bb_plot_heatmap_colData(
  obj,
  tile_color = "white",
  vars = colnames(colData(obj)),
  side = c("top", "bottom", "right", "left"),
  manual_pal = NULL
)

Arguments

Value

a ggplot

Description

Takes in a SummarizedHeatmap object and returns a ggplot of the rowDendro slot. This can be positioned with the side parameter. Default is to position it on the left. If flipping the heatmap so that the rows run vertically, you will need to change the side argument to top or bottom.

If row order is set explicitly when creating this object, the dendrogram slot will be NULL, and this function will abort.

Usage

bb_plot_heatmap_colDendro(
  obj,
  side = c("top", "bottom", "left", "right"),
  linewidth = 0.5
)

Arguments

Value

a ggplot

Description

Use geom_text_repel to selectively highlight some column names. Useful when there are too many to highlight to be able to use the axis directly.

Usage

bb_plot_heatmap_colHighlight(
  obj,
  highlights = character(0),
  side = c("top", "bottom", "right", "left"),
  ...
)

Arguments

Value

a ggplot

Description

Takes in a Summarized Heatmap object and returns a ggplot of the matrix data.

Usage

bb_plot_heatmap_main(
  obj,
  tile_color = "white",
  high = "red3",
  mid = "white",
  low = "blue4",
  flip = FALSE
)

Arguments

Value

a ggplot

Description

Use this function to create a plot annotating SummarizedHeatmap rowDAta

Usage

bb_plot_heatmap_rowData(
  obj,
  tile_color = "white",
  vars = colnames(rowData(obj)),
  side = c("right", "left", "top", "bottom"),
  manual_pal = NULL
)

Arguments

Value

a ggplot

Description

Takes in a SummarizedHeatmap object and returns a ggplot of the rowDendro slot. This can be positioned with the side parameter. Default is to position it on the left. If flipping the heatmap so that the rows run vertically, you will need to change the side argument to top or bottom.

If row order is set explicitly when creating this object, the dendrogram slot will be NULL and this function will abort.

Usage

bb_plot_heatmap_rowDendro(
  obj,
  side = c("left", "right", "top", "bottom"),
  linewidth = 0.5
)

Arguments

Value

A ggplot

Description

Plots a selected row name from a SummarizedHeatmap object. Useful when there are too many to highlight so you can't alter the plot axis.

Usage

bb_plot_heatmap_rowHighlight(
  obj,
  highlights = character(0),
  side = c("right", "left", "top", "bottom"),
  ...
)

Arguments

Value

a ggplot

Description

Use as argument for the "gene_or_genes" parameter for bb_gene_umap.

Usage

bb_plot_rowData_col(cds, rowData_col, filter_in = NULL, filter_out = NULL)

Arguments

Value

A data frame in the format needed to pass into bb_gene_umap.

Description

Generates the X axis for stacking other track plots on top of.

Usage

bb_plot_trace_axis(trace, xtitle = NULL)

Arguments

Description

A function to generate a line plot from tracklike genomic data. This pulls data from the Trace object trace_data slot. Check what numeric and categorical variables are available for plotting using Trace.data.

Usage

bb_plot_trace_data(
  trace,
  yvar = "coverage",
  yvar_label = "Coverage",
  facet_var = "group",
  color_var = "group",
  pal = NULL,
  legend_pos = "none",
  group_filter = NULL,
  group_variable = "group"
)

Arguments

Description

A function to generate a link plot from tracklike genomic data. Links will automatically be trimmed to lie entirely within the plot range. An additional, optional score cutoff can be provided.

Usage

bb_plot_trace_links(
  trace,
  cutoff = 0,
  link_low_color = "grey80",
  link_high_color = "red3",
  link_range = c(0, 1)
)

Arguments

Description

A function to generate a plot of the underlying gene model. The genes to be plotted are automatically selected according to the genome build and the plot range. The function automatically picks the longest principle transcript to show. Optionally, alternative transcripts can be shown by specifying the select_transcript argument. This must be an ensembl transcript identifier lying within the plot range.

Usage

bb_plot_trace_model(
  trace,
  font_face = "italic",
  line_width = 0.5,
  select_transcript = NULL,
  icon_fill = "cornsilk",
  icon_alpha = 0.5,
  arrow_scale = 1,
  segment_length_bp = 1000,
  debug = FALSE
)

Arguments

Description

A function to generate a peak plot from tracklike genomic data.

Usage

bb_plot_trace_peaks(
  trace,
  group_filter = NULL,
  group_variable = "group",
  pal = NULL
)

Arguments

Description

Plot motif footprinting results

Usage

bb_plotfootprint(
  object,
  features,
  alt_main_title = NULL,
  alt_color_title = NULL,
  legend_pos = "right",
  colorscale = NULL,
  assay = NULL,
  group.by = NULL,
  idents = NULL,
  label = TRUE,
  repel = TRUE,
  show.expected = TRUE,
  normalization = "subtract",
  label.top = 3,
  label.idents = NULL,
  fontsize = 14,
  linesize = 0.2
)

Arguments

Description

Print out a stats report

Usage

bb_print_full_stats(
  data,
  classification_variable,
  numeric_variable,
  test_type = c("Student", "Welch", "Wilcox"),
  output = NULL
)

Arguments

Value

A text file

Description

For a given GRanges object containing peaks, determine how many peaks overlap promoters, how many promoters are overlapped by peaks and the significance of enrichment of query peaks relative to promoters.

Usage

bb_promoter_overlap(query, tss = c("hg38_tss", "dr11_tss"), width = 200)

Arguments

Value

A list including overlap information and binomal test results.

Description

Use this function to perform Pseudbulk DGE analysis.

Usage

bb_pseudobulk_mf(
  cds,
  pseudosample_table,
  design_formula,
  count_filter = 10,
  result_recipe = "default",
  test = "Wald",
  reduced = NULL
)

Arguments

Value

A list of results from pseudobulk analysis

Description

This function allows you to simulate single cell data from bulk RNA-seq data. It requires a TPM count matrix. The rationale is that TPM quantifies transcript counts per million reads, so you can think of this like 1 million UMI counts from a scRNA-seq experiment distributed in a certain way across the transcriptome. This function samples n_pseudocells with transcripts_per_pseudocell from this distribution. Then it creates a cell data set based on a matrix of these samples.

Importantly, this function cannot accurately identify transcriptional heterogenity within the bulk data. The sampling effect may reveal some potential heterogeneity but there is no method here for determining whether this is due to randomness or heterogeneity within the data.

The intended use of this function is to project a pseudocell cds onto an actual single cell data set. This can be used to help identify regions of UMAP space that are shared between the pseudocells and the real cells.

Note: The matrix rownames and row metadata for the returned cds will contain only the rownames from the input tpm_matrix. So these should share the same namespace as the single cell cds that the pseudocell data will be projected onto.

Usage

bb_pseudocells(
  tpm_matrix,
  n_pseudocells,
  transcripts_per_pseudocell,
  remove_genes = NULL
)

Arguments

Value

A cell data set

See Also

map, reduce, map2, pmap tibble count, select components new_cell_data_set, combine_cds, preprocess_cds, reduce_dimension

Examples

## Not run: 
if(interactive()){
 #EXAMPLE1
 }

## End(Not run)

Description

This function determines a gene expression trajectory using learn_graph from monocle3 and then calculates pseudotime dimensions along this trajectory using order_cells. So it is 2 functions wrapped into 1. Usually we will not adjust the parameters for learn_graph with the possible exception of close_loop and use_partition which are also available in this function with the same defaults. If you need to fine-tune the trajectory, use monocle3::learn_graph on the cds object first and then run this function to calculate pseudotime. The graph learning will not be repeated on an object unless force_graph is set to TRUE.

If you just want to look at the trajectory graph and not calculate pseudotime, change calculate_pseudotime to FALSE, or run monocle3::learn_graph.

After the pseudotime values are calculated, they are handled differently than in monocle3. In this function, they are copied from the hidden CDS slot and made an explicit cell metadata column. Pseudotime needs a starting point or anchor. There is no interactive option here as in monocle3. To identify this starting point, you identify a cell metadata variable and provide it to cluster_variable. This should identify a cohesive group of cells in UMAP space such as a leiden cluster, louvain cluster or partition. Then provide a value corresponding to the cluster of interest to cluster_value. The function will start pseudotime at the cell closest to the graph node in that cluster. The pseudotime value column will be named automatically as a composite of the cluster_variable and cluster_value parameters.

Usage

bb_pseudotime(
  cds,
  calculate_pseudotime = TRUE,
  cluster_variable,
  cluster_value,
  use_partition = TRUE,
  close_loop = TRUE,
  force_graph = FALSE
)

Arguments

Value

A cell data set

See Also

cli_div, cli_alert learn_graph, order_cells, pseudotime SummarizedExperiment-class

Description

A function to run qc tests on cds objects.

Usage

bb_qc(
  cds,
  cds_name,
  genome = c("human", "mouse", "zfish"),
  nmad_mito = 2,
  nmad_detected = 2,
  max_mito = NULL,
  min_log_detected = NULL
)

Arguments

Value

A list of qc objects

Description

This function reads a single sorted, dedupliated paired end bam file and returns either a GRanges object or a GenomicAlignmentPairs object. The former requires much less memory but a the cost of retaining the outer boundaries of each read. If read 1 has start S1 and end E1 and read 2 has start S2 and end S2, the Granges object spans S1-E2.

Usage

bb_read_bam(
  sortedBam,
  genome = c("hg38", "danRer11"),
  return_type = c("GenomicAlignmentPairs", "GRanges")
)

Arguments

Value

An object according to return_type.

Description

This function reads the narrow peaks file (BED6 + 4 format) and turns it into a GRanges object.

Usage

bb_read_narrowpeak(file)

Arguments

Value

A GRanges object

Description

Rejoin qc and doubletfinder data to a cds object

Usage

bb_rejoin(cds, qc_data, doubletfinder_data)

Arguments

Value

A cell data set object with qc and doubletfinder data

Description

Remove rows that have duplicates in a given column

Usage

bb_remove_dupes(data, column)

Arguments

Value

A deduplicated tibble

Description

Take a cell_data_set or Seurat object and return the gene/feature metadata in the form of a tibble. RNA is used as the default assay.

Usage

bb_rowmeta(
  obj,
  row_name = "feature_id",
  experiment_type = "Gene Expression",
  assay = "RNA",
  cds = NULL
)

Arguments

Details

If a value is supplied for cds, a warning will be issued and the function will pass the value of cds to obj.

Value

At tibble.

Description

A function to generate automated cell labelings with Seurat

Usage

bb_seurat_anno(cds, reference)

Arguments

Value

A modified cds with Seurat cell assignments.

Description

Reads standard Souporcell output files and aggregates reference/alternate allele counts across cells assigned to each Souporcell genotype. The returned matrix is designed for genotype-demultiplexing QC plots, especially heatmaps showing whether inferred genetic IDs have distinct SNP allele profiles.

Usage

bb_souporcell_matrix(
  souporcell_dir,
  return = c("continuous", "discrete"),
  orientation = c("snps_by_genotypes", "genotypes_by_snps"),
  variant_file = NULL,
  use_variant_names = TRUE,
  variant_name_style = c("chr_pos_ref_alt", "chr_pos", "id"),
  status_keep = "singlet",
  assignment_col = "assignment",
  status_col = "status",
  min_total_reads = 10,
  min_genotype_coverage = 0.7,
  require_finite = TRUE,
  min_range = 0.35,
  min_sd = 0.12,
  ref_threshold = 0.2,
  alt_threshold = 0.8,
  require_ref_like = 1L,
  require_alt_like = 1L,
  top_n = 75L,
  discrete_values = c(ref = 0, het = 0.5, alt = 1),
  genotype_prefix = "ID_",
  snp_prefix = "SNP_",
  sort_genotypes = TRUE,
  verbose = TRUE,
  return_metadata = FALSE
)

Arguments

Details

The function expects a Souporcell output directory containing alt.mtx, ref.mtx, and clusters.tsv. If present, common_variants_covered_tmp.vcf is used to label SNPs with genomic coordinates and alleles.

Value

If return_metadata = FALSE, a numeric matrix. By default, rows are SNPs and columns are Souporcell genotype IDs. If return_metadata = TRUE, a list with elements matrix, snp_summary, variant_table, alt_by_genotype, ref_by_genotype, depth_by_genotype, allele_fraction_by_genotype, and clusters.

Examples

mat <- build_souporcell_genotype_matrix(
  souporcell_dir = "results/g_pos_1/souporcell/K15",
  return = "discrete",
  orientation = "snps_by_genotypes",
  top_n = 75
)

pheatmap::pheatmap(
  mat,
  cluster_rows = TRUE,
  cluster_cols = TRUE,
  show_rownames = FALSE
)

Description

Extracts Peaks data from 10X counts matrix and feature metadata and saves it as an alternate experiment in the assigned CDS.

Usage

bb_split_atac(cds)

Arguments

Value

A cell data set

Description

If you have cite-seq data together with gene expression data, this function will move the cite seq data to a new separate experiment. It will use Seurat to normalize these data using the CLR method and store them in a new assay.

Usage

bb_split_citeseq(cds)

Arguments

Value

a new CDS

See Also

SummarizedExperiment-class

Description

A Function To Add Tibble Columns To Cell Metadata

Usage

bb_tbl_to_coldata(obj, min_tbl, join_col = "cell_id", cds = NULL)

Arguments

Value

An object of the same class

Description

Convert a wide-form tibble a matrix

Usage

bb_tbl_to_matrix(data)

Arguments

Value

A matrix

Description

A Function To Add Tibble Columns To Feature Metadata

Usage

bb_tbl_to_rowdata(
  obj,
  assay = "RNA",
  min_tbl,
  join_col = "feature_id",
  cds = NULL
)

Arguments

Value

An object of the same class

Description

Based on Monocle3's Partitions, Leiden, and Louvain clustering methods. Implemented mostly with default values. Seurat objects will be converted to cell_data_set objects for the clustering. The function produces a list of top markers for each cluster type and returns these assignments to the original object as new cell metadata columnts.

Usage

bb_triplecluster(
  obj,
  n_top_markers = 50,
  outfile = NULL,
  n_cores = 8,
  cds = NULL
)

Arguments

Value

A modified Seurat or cell_data_set object

Description

Will rejoin scoresheet and blinded key to produce unblinded results. If you change the names of either of those files, they have to be provided as arguments to the function. Otherwise keyfile and scorefile are optional.

Usage

bb_unblind_images(
  directory,
  keyfile = "blinding_key.csv",
  scorefile = "scoresheet.csv",
  analysis_file,
  file_column
)

Arguments

Value

nothing

Description

A function to generate a UMAP with colors mapped to colData variables

Usage

bb_var_umap(
  obj,
  var,
  assay = "RNA",
  value_to_highlight = NULL,
  foreground_alpha = 1,
  legend_pos = "right",
  cell_size = 0.5,
  alt_stroke_color = NULL,
  legend_title = NULL,
  plot_title = NULL,
  palette = NULL,
  alt_dim_x = NULL,
  alt_dim_y = NULL,
  overwrite_labels = FALSE,
  group_label_size = 3,
  alt_label_col = NULL,
  shape = 21,
  nbin = 100,
  facet_by = NULL,
  sample_equally = FALSE,
  rasterize = FALSE,
  raster_dpi = 300,
  show_trajectory_graph = FALSE,
  trajectory_graph_color = "grey28",
  trajectory_graph_segment_size = 0.75,
  label_root_node = FALSE,
  pseudotime_dim = var,
  label_principal_points = FALSE,
  graph_label_size = 2,
  cds = NULL,
  outline_cluster = FALSE,
  outline_color = "black",
  outline_size = 1,
  outline_type = "solid",
  outline_alpha = 1,
  ...,
  man_text_df = NULL,
  text_geom = "text",
  minimum_segment_length = 1,
  hexify = FALSE,
  n_hexbins = 100
)

Arguments

Value

a ggplot

Description

This function gets the comments slot from an Ape object.

Usage

COMMENTS(ape)

Description

Generate median +/- se stat object for jitter ggplot

Usage

data_median_se(x)

Arguments

Description

Generate mean +/- sd stat object for jitter ggplot

Usage

data_summary_mean_sd(x)

Arguments

Description

Generate mean +/- se stat object for jitter ggplot

Usage

data_summary_mean_se(x)

Arguments

Description

Generate median +/- iqr stat object for jitter ggplot

Usage

data_summary_median_iqr(x)

Arguments

Description

Generate median +/- mad stat object for jitter ggplot

Usage

data_summary_median_mad(x)

Arguments

Description

Similar in function to dcast, but produces a sparse Matrix as an output. Sparse matrices are beneficial for this application because such outputs are often very wide and sparse. Conceptually similar to a pivot operation.

Usage

dMcast(
  data,
  formula,
  fun.aggregate = "sum",
  value.var = NULL,
  as.factors = FALSE,
  factor.nas = TRUE,
  drop.unused.levels = TRUE
)

Arguments

Details

Casting formulas are slightly different than those in dcast and follow the conventions of model.matrix. See formula for details. Briefly, the left hand side of the ~ will be used as the grouping criteria. This can either be a single variable, or a group of variables linked using :. The right hand side specifies what the columns will be. Unlike dcast, using the + operator will append the values for each variable as additional columns. This is useful for things such as one-hot encoding. Using : will combine the columns as interactions.

Value

a sparse Matrix

See Also

cast

dcast

Description

This function cats the features slot from an Ape object.

Usage

FEATURES(ape)

Description

This function provides a pipe-friendly method to filter cds objects.

Usage

filter_cds(cds, cells = "all", genes = "all")

Arguments

Value

A filtered CDS

Description

Build significance annotation geoms

Usage

geom_sig_annotations(
  annotation_data,
  draw_brackets = TRUE,
  text_family = NULL,
  text_face = NULL,
  text_colour = "black",
  bracket_colour = "black",
  bracket_linewidth = 0.4,
  bracket_linetype = 1,
  bracket_lineend = "round",
  vjust = 0
)

Arguments

Value

A list of ggplot layers.

Description

Helpers to add significance labels and optional brackets to ggplot2 plots from a user-supplied comparison table.

Returns an object that can be added directly to a ggplot with +.

Usage

geom_sig_table(
  p_table,
  y_npc,
  group1_col = "group1",
  group2_col = "group2",
  label_col = "significance",
  x_levels = NULL,
  facet_cols = NULL,
  draw_brackets = TRUE,
  bracket_tip_npc = 0.015,
  bracket_margin_npc = 0.02,
  text_size_pt = NULL,
  star_y_npc_offset = -0.05,
  text_family = NULL,
  text_face = NULL,
  text_colour = "black",
  bracket_colour = "black",
  bracket_linewidth = 0.2,
  bracket_linetype = 1,
  bracket_lineend = "round",
  vjust = 0
)

Arguments

Details

Notes:

• Facet-aware: if p_table contains facet columns matching the plot facets, annotations are placed only in matching panels. Otherwise they are repeated across all panels.

• PDF-safe significance stars: star-only labels like "", "", "" remain ASCII and are nudged upward slightly so they sit more like "ns".

• Bracket segment ends default to lineend = "round".

Value

An object that can be added to a ggplot with +.

Description

create a split violin geom

Usage

geom_split_violin(
  mapping = NULL,
  data = NULL,
  stat = "ydensity",
  position = "identity",
  ...,
  draw_quantiles = NULL,
  trim = TRUE,
  scale = "area",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

Description

This function takes a GRanges object and returns a character vector. Only 4 metadata fields from the GRanges object will be included: locus_tag, type, fwdcolor and revcolor. Locus_tag must be unique. This will be checked by the Ape constructor. This function should mostly be used internally in the construction and FEATURE-setting of instances of the Ape class.

Usage

granges_to_features(gr)

Arguments

Value

A character vector.

Description

This is an S4 Class for holding the relevant metadata we need for key images. The idea is to generate this object for each of the images we will use in a grant, paper or other important document. That way when we want to reuse these images we know where they are.

One can construct a single image object using the Image() constructor method. When called as such, it will open a file chooser to identify the file from the network drive. Then it will provide an interactive menu to add the needed metadata (see below).

The workflow is to use this to create a new Image object when you are using a new key image in a grant or other document. Then you will add the image to the image catalog using ImageCatalog.add.

The Image constructor provides several validation checks, including that the image file must be accessible.

Each slot has it's own getter and setter methods which are identical to the name of the slot.

It is critical that your images are stored in a common network drive. Ideally this is X/Labs/Blaser/staff/keyence imaging data

Avoid Duplication!! Please keep your all of your raw imaging data in X/Labs/Blaser/staff and subdirectories.

Usage

## S4 method for signature 'Image_'
show(object)

file_path(x)

file_path(x) <- value

species(x)

species(x) <- value

stage(x)

stage(x) <- value

genetics(x)

genetics(x) <- value

treatment(x)

treatment(x) <- value

microscope(x)

microscope(x) <- value

mag(x)

mag(x) <- value

filter(x)

filter(x) <- value

use(x)

use(x) <- value

note(x)

note(x) <- value

Slots

file_path

Path to file. Should start with ~/network/X/Labs/Blaser...

species

The species being imaged.

stage

The stage of the sample. Options include various times in hpf plus other for other timepoints.

genetics

Any genetic modifications.

treatment

Any treatments performed.

microscope

The microscope used.

mag

Magnification

filter

The filter or camera setup used.

use

The document the image is being used in.

note

Any additional notes.

Description

This object holds all of the individual Image objects.

Methods are provided for

• viewing as a tibble: ImageCatalog.as_tibble

• writing to tsv format: ImageCatalog.write

• adding an image: ImageCatalog.add

• deleting an image: ImageCatalog.delete

• subsetting and extracting: brackets and double brackets

An ImageCatalog object can be made from a list of Image objects using ImageCatalog(list = ).

More commonly, you will generate the ImageCatalog from a tsv catalog file. To make an image catalog this way, run ImageCatalog(catalog_path = ).

Usage

## S4 method for signature 'ImageCatalog'
show(object)

## S4 method for signature 'ImageCatalog,ANY,ANY,ANY'
x[i, j, ..., drop = TRUE]

## S4 method for signature 'ImageCatalog,ANY,ANY'
x[[i, j, ..., drop = TRUE]]

## S4 method for signature 'ImageCatalog'
x$name

ImageCatalog.as_tibble(image_catalog)

ImageCatalog.write(image_catalog, out)

ImageCatalog.add(image_catalog, image)

ImageCatalog.delete(image_catalog, hash)

Run link[monocle3]{learn_graph} on a Seurat object

Description

Run link[monocle3]{learn_graph} on a Seurat object

Usage

LearnGraph(object, reduction = DefaultDimReduc(object = object), ...)

Arguments

Value

A cell_data_set object with the pseudotime graph

See Also

learn_graph cell_data_set

Description

This function cats the Locus slot from an Ape object.

Usage

LOCUS(ape)

Description

Implementation of merge for Matrix. By explicitly calling merge.Matrix it will also work for matrix, for data.frame, and vector objects as a much faster alternative to the built-in merge.

Usage

merge.Matrix(
  x,
  y,
  by.x,
  by.y,
  all.x = TRUE,
  all.y = TRUE,
  out.class = class(x)[1],
  fill.x = ifelse(is(x, "sparseMatrix"), FALSE, NA),
  fill.y = fill.x,
  ...
)

join.Matrix(
  x,
  y,
  by.x,
  by.y,
  all.x = TRUE,
  all.y = TRUE,
  out.class = class(x)[1],
  fill.x = ifelse(is(x, "sparseMatrix"), FALSE, NA),
  fill.y = fill.x,
  ...
)

Arguments

Details

#' all.x/all.y correspond to the four types of database joins in the following way:

left

all.x=TRUE, all.y=FALSE

right

all.x=FALSE, all.y=TRUE

inner

all.x=FALSE, all.y=FALSE

full

all.x=TRUE, all.y=TRUE

Note that NA values will match other NA values.

Description

Often you will have a data table with repeats or batches of the same experiment. An effective way to control for batch effects is to normalize the data from each batch to a control group present in all of the experiments. To use this function, provide such a data table, identify the column holding the experimental group data, the identity of the control group to normalize by, the column holding the batch data, and the column holding the numerical data to normalize. Also select the function to average by (mean or median). The function will return the data table with three new columns: the average of the control group by batch, fold change of each observation relative to the batch average and the log2-transformed fold change. This function is pipe-friendly.

Usage

normalize_batch(
  data,
  group_col,
  norm_group,
  batch_col,
  data_col,
  fun = c("mean", "median")
)

Arguments

Value

A tibble with new columns indicating batch normalization group average, fold change for each observation relative to the batch average and log2 fold change

See Also

arg_match cli_abort filter, group_by, summarise, select, mutate-joins, mutate

Description

Facet-aware version. If p_table contains facet columns matching the plot, annotations are placed only in the matching panels. If not, annotations are repeated across all panels.

Usage

prepare_sig_annotations(
  plot,
  p_table,
  y_npc,
  group1_col = "group1",
  group2_col = "group2",
  label_col = "significance",
  x_levels = NULL,
  facet_cols = NULL,
  draw_brackets = TRUE,
  bracket_tip_npc = 0.015,
  bracket_margin_npc = 0.02,
  text_size_pt = NULL,
  star_y_npc_offset = 0.01
)

Arguments