Package 'seqsetvis'

Description

2 steps ssvOverlapIntervalSets. ssvFetchBigwig. Otherwise refer to the vignettes to see

Author(s)

Maintainer: Joseph R Boyd [email protected] (ORCID)

Description

see sam specs for cigar details

Usage

.expand_cigar_dt(cigar_dt, op_2count = c("M", "D", "=", "X"))

Arguments

Value

data.table with cigar entries expanded

Description

see sam specs for cigar details

Usage

.expand_cigar_dt_recursive(cigar_dt)

Arguments

Value

data.table with cigar entries expanded

Description

flag = scanBamFlag(isDuplicate = FALSE)

Usage

.rm_dupes(reads_dt, max_dupes)

Arguments

Value

reads_dt with duplicated reads over max_dupes removed

Description

flag = scanBamFlag(isDuplicate = FALSE)

Usage

.rm_dupesPE(reads_dt, max_dupes)

Arguments

Value

reads_dt with duplicated reads over max_dupes removed

Description

adds rectangle boxes proportional to cluster sizes of heatmap with optional labels.

Usage

add_cluster_annotation(
  cluster_ids,
  p = NULL,
  xleft = 0,
  xright = 1,
  rect_colors = c("black", "gray"),
  text_colors = rev(rect_colors),
  show_labels = TRUE,
  label_angle = 0,
  row_ = "id",
  cluster_ = "cluster_id"
)

Arguments

Value

A ggplot with cluster annotations added.

Examples

data(CTCF_in_10a_profiles_dt)
#simplest uses
add_cluster_annotation(factor(c(rep("A", 3), "B")))
p = ggplot() + coord_cartesian(xlim = c(0,10))
add_cluster_annotation(factor(c(rep("A", 3), "B")), p)

#intended use with ssvSignalHeatmap
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 3)
assign_dt = unique(clust_dt[, .(id, cluster_id)])[order(id)]
p_heat = ssvSignalHeatmap(clust_dt, show_cluster_bars = FALSE)
add_cluster_annotation(assign_dt$cluster_id, p_heat,
  xleft = -500, xright = -360, rect_colors = rainbow(3), text_colors = "gray")

#when colors are named, the names are used rather that just the order
rect_colors = safeBrew(assign_dt$cluster_id)
text_colors = safeBrew(assign_dt$cluster_id, "greys")
p_clusters = add_cluster_annotation(assign_dt$cluster_id,
  rect_colors = rect_colors, text_colors = text_colors)
#specialized use as plot outside of heatmap
p1 = assemble_heatmap_cluster_bars(plots = list(p_clusters, p_heat), rel_widths = c(1, 3))

#when colors are named, the names are used rather that just the order
#these plots will be identical even though order of colors changes.
rect_colors = rect_colors[c(2, 3, 1)]
text_colors = text_colors[c(3, 1, 2)]
p_clusters = add_cluster_annotation(assign_dt$cluster_id,
  rect_colors = rect_colors, text_colors = text_colors)
#specialized use as plot outside of heatmap
p2 = assemble_heatmap_cluster_bars(plots = list(p_clusters, p_heat), rel_widths = c(1, 3))

cowplot::plot_grid(p1, p2, ncol = 1)

Description

see calc_norm_factors for normalization details.

Usage

append_ynorm(
  full_dt,
  value_ = "y",
  cap_value_ = "y_cap_value",
  norm_value_ = "y_norm",
  by1 = "id",
  by2 = "sample",
  aggFUN1 = max,
  aggFUN2 = function(x) quantile(x, 0.95),
  cap_dt = NULL,
  do_not_cap = FALSE,
  do_not_scaleTo1 = FALSE,
  force_append = FALSE
)

Arguments

Value

data.table, full_dt with cap_value_ and norm_value_ values appended.

Examples

data(CTCF_in_10a_profiles_dt)
append_ynorm(CTCF_in_10a_profiles_dt)
append_ynorm(CTCF_in_10a_profiles_dt,
  aggFUN1 = mean, aggFUN2 = function(x)quantile(x, .5))

Description

http://www.cookbook-r.com/Manipulating_data/Calculating_a_moving_average/

Usage

applyMovingAverage(
  dt,
  n,
  centered = TRUE,
  x_ = "x",
  y_ = "y",
  by_ = c("id", "sample"),
  maFun = movingAverage
)

Arguments

Value

a newly derived data.table where a movingAverage has been applied.

Examples

data(CTCF_in_10a_profiles_dt)
agg_dt = CTCF_in_10a_profiles_dt[, list(y = mean(y)), by = list(sample, x)]
ggplot(agg_dt) +
    geom_line(aes(x = x, y = y, color = sample))

ma_smooth = applyMovingAverage(agg_dt, n = 5,
    y_ = 'y', by_ = c('sample'))
ggplot(ma_smooth) +
    geom_line(aes(x = x, y = y, color = sample))

ma_smooth$method = "moving_average"
agg_dt$method = "none"
ggplot(rbind(ma_smooth, agg_dt)) +
    geom_line(aes(x = x, y = y, color = method)) +
    facet_wrap(~sample)

Description

applySpline Is intended for two-dimensional tidy data.tables, as retured by ssvFetchBigwig

Usage

applySpline(
  dt,
  n,
  x_ = "x",
  y_ = "y",
  by_ = c("id", "sample"),
  splineFun = stats::spline
)

Arguments

Details

by_ is quite powerful. If by_ = c('gene_id', 'sample_id'), splines will be calculated individually for each gene in each sample. alternatively if by_ = c('gene_id')

Value

a newly derived data.table that is n times longer than original.

See Also

ssvFetchBigwig

Examples

data(CTCF_in_10a_profiles_dt)
#data may be blockier than we'd like
ggplot(CTCF_in_10a_profiles_dt[, list(y = mean(y)), by = list(sample, x)]) +
    geom_line(aes(x = x, y = y, color = sample))

#can be smoothed by applying a spline  (think twice about doing so,
#it may look prettier but may also be deceptive or misleading)

splined_smooth = applySpline(CTCF_in_10a_profiles_dt, n = 10,
    y_ = 'y', by_ = c('id', 'sample'))
ggplot(splined_smooth[, list(y = mean(y)), by = list(sample, x)]) +
    geom_line(aes(x = x, y = y, color = sample))

Description

assemble_heatmap_cluster_bars

Usage

assemble_heatmap_cluster_bars(plots, ...)

Arguments

Value

A grob produced by cowplot::plot_grid

Examples

data(CTCF_in_10a_profiles_gr)
plots = ssvSignalHeatmap.ClusterBars(CTCF_in_10a_profiles_gr, return_unassembled_plots = TRUE)
assemble_heatmap_cluster_bars(plots)

Description

matches system.file("extdata/Bcell_PE.mm10.bam", package = "seqsetvis")

Format

GRanges length 4

Details

this is included only for testing ssvFetchBamPE functions.

Value

GRanges length 4

Description

Calculate normalization factors in a two step process:

Usage

calc_norm_factors(
  full_dt,
  value_ = "y",
  cap_value_ = "y_cap_value",
  by1 = "id",
  by2 = "sample",
  aggFUN1 = max,
  aggFUN2 = function(x) quantile(x, 0.95)
)

Arguments

Details

1. summarize every region for each sample (default summary function is max)

2. caclulate a value to cap each sample to based on regions (default is 95th quantile).

The uderlying assumption here is that meaningful enrichment is present at the majority of regions provided. If prevalence varies by a specific factor, say ChIP-seq targets with different characteristics - ie. when analyzing TSSes for H3K4me3 and an infrequent transcription factor it is more appropriate to specify appropriate quantile cutoffs per factor.

Value

data.table mapping by2 to cap_value_.

Examples

data(CTCF_in_10a_profiles_dt)
calc_norm_factors(CTCF_in_10a_profiles_dt)
calc_norm_factors(CTCF_in_10a_profiles_dt,
  aggFUN1 = mean, aggFUN2 = function(x)quantile(x, .5))

Description

centerAtMax locates the coordinate x of the maximum in y and shifts x such that it is zero at max y.

Usage

centerAtMax(
  dt,
  x_ = "x",
  y_ = "y",
  by_ = "id",
  view_size = NULL,
  trim_to_valid = TRUE,
  check_by_dupes = TRUE,
  x_precision = 3,
  replace_x = TRUE
)

Arguments

Details

character. by_ controls at the level of the data centering is applied. If by_ is "" or NULL, a single max position will be determined for the entire dataset. If by is "id" (the default) then each region will be centered individually across all samples.

Value

data.table with x (or xnew if replace_x is FALSE) shifted such that x = 0 matches the maximum y-value define by by_ grouping

Examples

data(CTCF_in_10a_profiles_gr)
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', by_ = 'id',
  check_by_dupes = FALSE)
#it's a bit clearer what's happening with trimming disabled
#but results are less useful for heatmaps etc.
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', by_ = 'id',
  check_by_dupes = FALSE, trim_to_valid = FALSE)
#specify view_size to limit range of x values considered, prevents
#excessive data trimming.
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', view_size = 100, by_ = 'id',
check_by_dupes = FALSE)

Description

centerFixedSizeGRanges First calculates the central coordinate of each GRange in grs and extends in both direction by half of fixed_size

Usage

centerFixedSizeGRanges(grs, fixed_size = 2000)

Arguments

Value

Set of GRanges after resizing all input GRanges, either shortened or lengthened as required to match fixed_size

Examples

library(GenomicRanges)
grs = GRanges("chr1", IRanges(1:10+100, 1:10*3+100))
centered_grs = centerFixedSizeGRanges(grs, 10)
width(centered_grs)

Description

Centers query GRanges at maximum signal in prof_dt.

Usage

centerGRangesAtMax(
  prof_dt,
  qgr,
  x_ = "x",
  y_ = "y",
  by_ = "id",
  width = 1,
  view_size = NULL
)

Arguments

Value

a GRanges with same mcols as qgr that has been centered based on signal in prof_dt and with regions of specified width.

Examples

data(CTCF_in_10a_overlaps_gr)
data(CTCF_in_10a_profiles_gr)
data(CTCF_in_10a_profiles_dt)
centerGRangesAtMax(CTCF_in_10a_profiles_dt, CTCF_in_10a_overlaps_gr)
centerGRangesAtMax(CTCF_in_10a_profiles_gr, CTCF_in_10a_overlaps_gr)

centerGRangesAtMax(CTCF_in_10a_profiles_gr, CTCF_in_10a_overlaps_gr, view_size = 50)

Description

MCF10A CTCF profiles at 20 windows per chromHMM state, hg38.

Format

a GRanges object of length 4000 with 5 metadata columns sufficient for use with ggplot2

Details

part of chromHMM_demo_data

the result of ssvFetchBigwig() on the MCF10A_CTCF_FE.bw near 20 randomly selected windows per chromHMM state.

Value

a GRanges object of length 4000 with 5 metadata columns sufficient for use with ggplot2

Description

URL to download hg19ToHg38 liftover chain from UCSC

Format

a character containing a URL

Details

file is gzipped .txt

part of chromHMM_demo_data

Value

a character containing a URL

Description

Vignette data for seqsetvis was downloaded directly from GEO series GSE57498. This data is the state segmentation by chromHMM in the MCF7 cell line. chromHMM creates a hidden markov model by integrating several ChIP-seq samples, in this case:

• MCF7_H3K27ac_ChIP-Seq

• MCF7_H3K27me3_ChIP-Seq

• MCF7_H3K4me1_ChIP-Seq

• MCF7_H3K4me3_ChIP-Seq

• MCF7_RNApolIIp_ChIP-Seq

Data from GEO series GSE57498 is from the publication Taberlay PC et al. 2014

Details

Contains:

• chromHMM_demo_overlaps_gr

• chromHMM_demo_bw_states_gr

• chromHMM_demo_state_total_widths

• chromHMM_demo_state_colors

• chromHMM_demo_segmentation_url

• chromHMM_demo_chain_url

Description

overlap of MCF10A CTCF with MCF7 chromHMM states, hg38.

Format

a GRanges object of length 98 with 10 logical metadata columns, 1 per state.

Details

part of chromHMM_demo_data

the result of ssvOverlapIntervalSets() on MCF10A CTCF peaks and MCF7 chromHMM states with use_first = TRUE

first (the MCF10A peaks) and no_hit columns have been removed each remaining column represents MCF10A peaks overlapping with a state.

Value

a GRanges object of length 98 with 10 logical metadata columns, 1 per state.

Description

URL to download hg19 MCF7 chromHMM segmentation

Format

a character containing a URL

Details

file is gzipped bed with name, score, itemRgb and thick meta columns

part of chromHMM_demo_data

Value

a character containing a URL

Description

original state name to color mappings stored in segmentation bed

Format

a named character vector mapping states to hex colors

Details

part of chromHMM_demo_data

Value

a named character vector mapping states to hex colors

Description

state name to total width mappings, hg38

Format

named numeric of total widths per state

Details

part of chromHMM_demo_data

Value

named numeric of total widths per state

Description

perform kmeans clustering on matrix rows and return reordered matrix along with order matched cluster assignments. clusters are sorted using hclust on centers

Usage

clusteringKmeans(mat, nclust, centroids = NULL, iter.max = 30)

Arguments

Value

data.table with group__ variable indicating cluster membership and id__ variable that is a factor indicating order based on within cluster similarity

Examples

data(CTCF_in_10a_profiles_dt)
dt = data.table::copy(CTCF_in_10a_profiles_dt)
mat = data.table::dcast(dt, id ~ sample + x, value.var = "y" )
rn = mat$id
mat = as.matrix(mat[,-1])
rownames(mat) = rn
clust_dt = clusteringKmeans(mat, nclust = 3)
dt = merge(dt, clust_dt[, .(id = id__, group = group__)])
dt$id = factor(dt$id, levels = clust_dt$id)
dt[order(id)]

Description

perform kmeans clustering on matrix rows and return reordered matrix along with order matched cluster assignments clusters are sorted using hclust on centers the contents of each cluster are sorted using hclust

Usage

clusteringKmeansNestedHclust(
  mat,
  nclust,
  within_order_strategy = valid_sort_strategies[2],
  centroids = NULL,
  manual_mapping = NULL,
  iter.max = 30
)

Arguments

Value

data.table with 2 columns of cluster info. id__ column corresponds with input matrix rownames and is sorted within each cluster using hierarchical clusering group__ column indicates cluster assignment

Examples

data(CTCF_in_10a_profiles_dt)
dt = data.table::copy(CTCF_in_10a_profiles_dt)
mat = data.table::dcast(dt, id ~ sample + x, value.var = "y" )
rn = mat$id
mat = as.matrix(mat[,-1])
rownames(mat) = rn
clust_dt = clusteringKmeansNestedHclust(mat, nclust = 3)
clust_dt

Description

converts a valid r color name ("black", "red", "white", etc.) to a hex value

Usage

col2hex(color_name)

Arguments

Value

hex value of colors coded by colors()

Examples

col2hex(c("red", "green", "blue"))
col2hex(c("lightgray", "gray", "darkgray"))

Description

collapse non-contiguous regions (i.e. exons) into a contiguous coordinate starting at 1. this is strand sensitive and intended for use with all exons of a single gene.

Usage

collapse_gr(genome_gr)

Arguments

Value

a new GRanges object with same mcols as input with all intervals starting at 1 and no empty space between syntenic regions.

Examples

library(data.table)
library(GenomicRanges)
dev_dat = data.table(seqnames = "chrTest",
                     transcript_id = c(1, 1,  2, 2, 3, 3, 3),
                     start = c(5,  30,  8, 30, 2, 30, 40),
                     end = c(10, 35, 15, 38, 7, 35, 45),
                     strand = "+")

genome_gr = GRanges(dev_dat)
collapse_gr(genome_gr)

neg_gr = genome_gr
strand(neg_gr) = "-"
collapse_gr(neg_gr)

Description

(preliminary implementation, sub-optimal)

Usage

convert_collapsed_coord(genome_gr, x)

Arguments

Details

see collapse_gr for explanation of intended uses. this function translates all values of x from original genomic coordinates to new coordinate space created by collapse_gr.

Value

numeric, positions of every value of x within collapse coordinates. values outside of collapsed regions (an intron or outside range) will be NA.

Examples

library(data.table)
library(GenomicRanges)
dev_dat = data.table(seqnames = "chrTest",
                     transcript_id = c(1, 1,  2, 2, 3, 3, 3),
                     start = c(5,  30,  8, 30, 2, 30, 40),
                     end = c(10, 35, 15, 38, 7, 35, 45),
                     strand = "+")

genome_gr = GRanges(dev_dat)
convert_collapsed_coord(genome_gr, start(genome_gr))
convert_collapsed_coord(genome_gr, end(genome_gr))

Description

copy_clust_info

Usage

copy_clust_info(target, to_copy, row_ = "id", cluster_ = "cluster_id")

Arguments

Value

data.table or GRanges (whichever target is) containing row order and cluster assignment derived from to_copy. Suitable for ssvSignalHeatmap and related functions.

Examples

data(CTCF_in_10a_narrowPeak_grs)
data(CTCF_in_10a_overlaps_gr)
data(CTCF_in_10a_profiles_dt)
#this takes cluster info from signal and applies to peak hits to
#create a heatmap of peak hits clustered by signal.
clust_dt1 = ssvSignalClustering(CTCF_in_10a_profiles_dt)
peak_hit_gr = ssvFetchGRanges(
  CTCF_in_10a_narrowPeak_grs,
  qgr = CTCF_in_10a_overlaps_gr
)
peak_hit_gr.clust = copy_clust_info(peak_hit_gr, clust_dt1)
peak_hit_gr.clust$hit = peak_hit_gr.clust$y > 0
ssvSignalHeatmap(peak_hit_gr.clust, fill_ = "hit") +
  scale_fill_manual(values = c("FALSE" = "gray90", "TRUE" = "black"))

Description

Calculate cross correlation by using shiftApply on read coverage Rle

Usage

crossCorrByRle(
  bam_file,
  query_gr,
  max_dupes = 1,
  fragment_sizes = 50:300,
  read_length = NULL,
  flip_strand = FALSE,
  ...
)

Arguments

Value

named list of results

Examples

data(CTCF_in_10a_overlaps_gr)
bam_f = system.file("extdata/test.bam",
    package = "seqsetvis", mustWork = TRUE)
query_gr = CTCF_in_10a_overlaps_gr[1:2]
crossCorrByRle(bam_f, query_gr[1:2], fragment_sizes = seq(50, 300, 50))

Description

FE bigWig tracks for CTCF ChIP-seq in a MCF10A progression model. See GEO series GSE98551 for details.

Format

named character vector of length 3

Details

part of CTCF_in_10a_data

Description

Vignette data for seqsetvis was downloaded directly from GEO series GSE98551. This data is CTCF ChIP-seq from a model of breast cancer progression derived from the MCF10A cell line.

Data from GEO series GSE98551 is from the publication Fritz AJ et al. 2018

Details

Contains:

• CTCF_in_10a_overlaps_gr

• CTCF_in_10a_profiles_dt

• CTCF_in_10a_bigWig_urls

• CTCF_in_10a_narrowPeak_urls

Description

list of GRanges that results in 100 random subset when overlapped

Format

named list of GRanges of length 3

Details

part of CTCF_in_10a_data

Value

named list of GRanges of length 3

Description

macs2 peak calls for CTCF ChIP-seq in a MCF10A progression model. See GEO series GSE98551 for details.

Format

named character vector of length 3

Details

part of CTCF_in_10a_data

Description

MACS2 narrowPeak calls on pooled biological replicates at pval 1e-5 and then 0.05 IDR filtered. IDR cutoffs determined by comparing top 150,000 pvalue sorted peak in replicates.

Format

GenomicRanges with 3 metadata columns of membership table

Details

See GEO series GSE98551 for details.

part of CTCF_in_10a_data

Description

A tidy data.table at window size 50 bp within 350 bp of peak center The variables are as follows:

Format

A tidy data.table of 2100 rows and 9 columns

Details

part of CTCF_in_10a_data

1. seqnames. chromosome for GRanges compatibility

2. start. start of interval

3. end. end of interval

4. width. width of interval

5. strand. leftover from GRanges.

6. id. unique identifier

7. y. fold-enrichment over input.

8. x. bp relative to center

9. sample. name of originating sample

Description

A tidy GRanges at window size 50 bp within 350 bp of peak center The variables are as follows:

Format

A tidy GRanges of 2100 rows and 4 metadata columns

Details

part of CTCF_in_10a_data

1. id. unique identifier

2. y. fold-enrichment over input.

3. x. bp relative to center

4. sample. name of originating sample

Description

Mainly a utility function for loading MACS2 narrowPeak and broadPeak.

Usage

easyLoad_bed(
  file_paths,
  file_names = NULL,
  extraCols = character(),
  n_cores = getOption("mc.cores", 1)
)

Arguments

Value

a named list of GRanges loaded from file_paths

Examples

bed_f = system.file("extdata/test_loading.bed",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_bed(bed_f, "my_bed")

Description

easyLoad_broadPeak takes a character vector of file paths to narrowPeak files from MACS2 and returns a named list of GRanges.

Usage

easyLoad_broadPeak(
  file_paths,
  file_names = NULL,
  n_cores = getOption("mc.cores", 1)
)

Arguments

Value

a named list of GRanges loaded from file_paths

Examples

bp_f = system.file("extdata/test_loading.broadPeak",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_broadPeak(bp_f, "my_broadPeak")

Description

easyLoad_FUN takes a character vector of file paths run an arbitrary function defined in load_FUN

Usage

easyLoad_FUN(
  file_paths,
  load_FUN,
  file_names = NULL,
  n_cores = getOption("mc.cores", 1),
  ...
)

Arguments

Value

a named list of results from load_FUN

Examples

bed_f = system.file("extdata/test_loading.bed",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_bed(bed_f, "my_bed")

Description

easyLoad_IDRmerged loads "overlapped-peaks.txt" from IDR.

Usage

easyLoad_IDRmerged(
  file_paths,
  file_names = NULL,
  n_cores = getOption("mc.cores", 1),
  max_idr = 0.05
)

Arguments

Value

named list of GRanges

Examples

idr_file = system.file("extdata/test_idr.overlapped-peaks.txt",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_IDRmerged(idr_file)
easyLoad_IDRmerged(idr_file, max_idr = .01)

Description

easyLoad_narrowPeak takes a character vector of file paths to narrowPeak files from MACS2 and returns a named list of GRanges.

Usage

easyLoad_narrowPeak(
  file_paths,
  file_names = NULL,
  n_cores = getOption("mc.cores", 1)
)

Arguments

Value

a named list of GRanges loaded from file_paths

Examples

np_f = system.file("extdata/test_loading.narrowPeak",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_narrowPeak(np_f, "my_narrowPeak")

Description

easyLoad_seacr takes a character vector of file paths to seacr output bed files and returns a named list of GRanges.

Usage

easyLoad_seacr(
  file_paths,
  file_names = NULL,
  n_cores = getOption("mc.cores", 1)
)

Arguments

Value

a named list of GRanges loaded from file_paths

Examples

bed_f = system.file("extdata/test_loading.seacr.bed",
    package = "seqsetvis", mustWork = TRUE)
easyLoad_seacr(bed_f, "my_seacr")

Description

see sam specs for cigar details

Usage

expandCigar(
  cigar_dt,
  op_2count = c("M", "D", "=", "X"),
  return_data.table = FALSE
)

Arguments

Value

data.table with cigar entries expanded

Examples

data(CTCF_in_10a_overlaps_gr)
qgr = CTCF_in_10a_overlaps_gr[1:5]
bam_file = system.file("extdata/test.bam", package = "seqsetvis", mustWork = TRUE)
raw_dt = ssvFetchBam(bam_file, qgr, return_unprocessed = TRUE)
expandCigar(raw_dt)

Description

fetch a bam file pileup with the ability to consider read extension to fragment size (fragLen)

Usage

fetchBam(
  bam_f,
  qgr,
  fragLen = NULL,
  target_strand = c("*", "+", "-")[1],
  max_dupes = Inf,
  splice_strategy = c("none", "ignore", "add", "only", "splice_count")[1],
  flip_strand = FALSE,
  return_unprocessed = FALSE,
  ...
)

Arguments

Value

GRanges containing tag pileup values in score meta column. tags are optionally extended to fragment length (fragLen) prior to pile up.

Description

findMaxPos

Usage

findMaxPos(prof_dt, qgr, x_ = "x", y_ = "y", by_ = "id", width = 1)

Arguments

Value

data.table of relative x position from center per id

Examples

data(CTCF_in_10a_overlaps_gr)
data(CTCF_in_10a_profiles_gr)
data(CTCF_in_10a_profiles_dt)
findMaxPos(CTCF_in_10a_profiles_dt, CTCF_in_10a_overlaps_gr)
findMaxPos(CTCF_in_10a_profiles_gr, CTCF_in_10a_overlaps_gr)

Description

calculate fragLen from a bam file for specified regions

Usage

fragLen_calcStranded(
  bam_f,
  qgr,
  n_regions = 100,
  include_plot_in_output = FALSE,
  test_fragLen = seq(100, 400, 5),
  flip_strand = FALSE,
  ...
)

Arguments

Value

numeric fragment length

Examples

data(CTCF_in_10a_overlaps_gr)
bam_file = system.file("extdata/test.bam",
    package = "seqsetvis")
qgr = CTCF_in_10a_overlaps_gr[1:5]
fragLen_calcStranded(bam_file, qgr)
#if plot is included, a list is returned, item 2 is the plot
fragLen_calcStranded(bam_file, qgr,
  include_plot_in_output = TRUE)[[2]]

Description

parse fragLen from MACS2 output

Usage

fragLen_fromMacs2Xls(macs2xls_file)

Arguments

Value

numeric fragment length

Examples

xls_file = system.file("extdata/test_peaks.xls",
    package = "seqsetvis")
fragLen_fromMacs2Xls(xls_file)

Description

get_mapped_reads

Usage

get_mapped_reads(bam_files)

Arguments

Value

the total mapped reads in each bam file as a named numeric vector.

Examples

bam_file = system.file("extdata/test.bam", package = "seqsetvis", mustWork = TRUE)
get_mapped_reads(bam_file)

Description

determine the most common read length for input bam_file. uses 50 randomly selected regions from query_gr. If fewer than 20 reads are present, loads all of query_gr.

Usage

getReadLength(bam_file, query_gr)

Arguments

Value

numeric of most common read length.

Examples

data(CTCF_in_10a_overlaps_gr)
qgr = CTCF_in_10a_overlaps_gr[1:5]
bam_file = system.file("extdata/test.bam", package = "seqsetvis", mustWork = TRUE)
getReadLength(bam_file, qgr)

Description

returns a ggplot with ellipses drawn using specified parameters used by ssvFeatureVenn and ssvFeatureEuler

Usage

ggellipse(
  xcentres,
  ycentres,
  r,
  r2 = r,
  phi = rep(0, length(xcentres)),
  circle_colors = NULL,
  group_names = LETTERS[seq_along(xcentres)],
  line_alpha = 1,
  fill_alpha = 0.3,
  line_width = 2,
  n_points = 200
)

Arguments

Details

uses eulerr's non-exported ellipse drawing coordinate function

Value

a ggplot containing ellipses

Examples

ggellipse(xcentres = c(1, 1, 2),
    ycentres = c(2, 1, 1),
    r = c(1, 2, 1))
ggellipse(xcentres = c(1, 1, 2),
    ycentres = c(2, 1, 1),
    r = c(1, 2, 1),
    fill_alpha = 0,
    group_names = paste("set", 1:3))
ggellipse(xcentres = c(1, 1, 2),
    ycentres = c(2, 1, 1),
    r = c(1, 2, 1),
    circle_colors = c("red", "orange", "yellow"),
    line_alpha = 0,
    group_names = paste("set", 1:3))

Description

ensures compatibility between seqlength of gr and bam_file based on header

Usage

harmonize_seqlengths(query_gr, bam_file, force_fix = FALSE)

Arguments

Value

GRanges with seqlengths matching bam_file

Examples

library(GenomicRanges)
query_gr = GRanges("chr1", IRanges(1, 100))
#seqlengths has not been set
seqlengths(query_gr)
bam = system.file("extdata/test.bam", package = "seqsetvis")
gr2 = harmonize_seqlengths(query_gr, bam)
#seqlengths now set
seqlengths(gr2)

Description

Create a wide matrix from a tidy data.table more suitable for clustering methods

Usage

make_clustering_matrix(
  tidy_dt,
  row_ = "id",
  column_ = "x",
  fill_ = "y",
  facet_ = "sample",
  max_rows = 500,
  max_cols = 100,
  clustering_col_min = -Inf,
  clustering_col_max = Inf,
  dcast_fill = NA,
  fun.aggregate = "mean"
)

Arguments

Value

A wide matrix version of input tidy data.table

Examples

data(CTCF_in_10a_profiles_dt)
mat = make_clustering_matrix(CTCF_in_10a_profiles_dt)
mat[1:5, 1:5]

Description

merge_clusters

Usage

merge_clusters(
  clust_dt,
  to_merge,
  row_ = "id",
  cluster_ = "cluster_id",
  reapply_cluster_names = TRUE
)

Arguments

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
set.seed(0)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 6)
ssvSignalHeatmap(clust_dt)
agg_dt = clust_dt[, list(y = mean(y)), list(x, cluster_id, sample)]
ggplot(agg_dt, aes(x = x, y = y, color = sample)) +
  geom_path() +
  facet_grid(cluster_id~.)

to_merge = c(2, 3, 5)
# debug(merge_clusters)
new_dt = merge_clusters(clust_dt, c(2, 3, 5), reapply_cluster_names = FALSE)
new_dt.relabel = merge_clusters(clust_dt, c(2, 3, 5), reapply_cluster_names = TRUE)
new_dt.relabel.sort = within_clust_sort(new_dt.relabel, within_order_strategy = "sort")

table(clust_dt$cluster_id)
table(new_dt$cluster_id)

cowplot::plot_grid(
  ssvSignalHeatmap(clust_dt) + labs(title = "original"),
  ssvSignalHeatmap(new_dt) + labs(title = "2,3,5 merged"),
  ssvSignalHeatmap(new_dt.relabel) + labs(title = "2,3,5 merged, renumbered"),
  ssvSignalHeatmap(new_dt.relabel.sort) + labs(title = "2,3,5 merged, renumbered and sorted")
)

Description

Deprecated and renamed as prepare_fetch_GRanges_width

Usage

prepare_fetch_GRanges(
  qgr,
  win_size,
  min_quantile = 0.75,
  target_size = NULL,
  skip_centerFix = FALSE
)

Arguments

Details

output GRanges parallels input with consistent width evenly divisible by win_size. Has warning if GRanges needed resizing, otherwise no warning and input GRanges is returned unchanged.

Value

GRanges, either identical to qgr or with suitable consistent width applied.

Examples

data(CTCF_in_10a_overlaps_gr)
#use prepare_fetch_GRanges_width instead:
qgr = prepare_fetch_GRanges_width(CTCF_in_10a_overlaps_gr, win_size = 50)
#no warning if qgr is already valid for windowed fetching
prepare_fetch_GRanges_width(qgr, win_size = 50)

Description

If $id is set, that value is used as name and duplicates are checked for.

Usage

prepare_fetch_GRanges_names(qgr, include_id = FALSE)

Arguments

Value

and named GRanges based on input qgr.

Examples

data(CTCF_in_10a_overlaps_gr)
qgr = CTCF_in_10a_overlaps_gr
names(qgr) = NULL
#default is to paste "region_" and iteration along length of qgr
prepare_fetch_GRanges_names(qgr)
#id gets used is already set
qgr$id = paste0("peak_", rev(seq_along(qgr)), "_of_", length(qgr))
prepare_fetch_GRanges_names(qgr)

Description

output GRanges parallels input with consistent width evenly divisible by win_size. Has warning if GRanges needed resizing, otherwise no warning and input GRanges is returned unchanged.

Usage

prepare_fetch_GRanges_width(
  qgr,
  win_size,
  min_quantile = 0.75,
  target_size = NULL,
  skip_centerFix = FALSE
)

Arguments

Value

GRanges, either identical to qgr or with suitable consistent width applied.

Examples

data(CTCF_in_10a_overlaps_gr)
qgr = prepare_fetch_GRanges_width(CTCF_in_10a_overlaps_gr, win_size = 50)
#no warning if qgr is already valid for windowed fetching
prepare_fetch_GRanges_width(qgr, win_size = 50)

Description

Returns the lowest multiple of win_size greater than min_quantile quantile of width(qgr)

Usage

quantileGRangesWidth(qgr, min_quantile = 0.75, win_size = 1)

Arguments

Value

numeric that is >= min_quantile and evenly divisible by win_size

Examples

data(CTCF_in_10a_overlaps_gr)
gr = CTCF_in_10a_overlaps_gr
quantileGRangesWidth(gr)
quantileGRangesWidth(gr, min_quantile = .5, win_size = 100)

Description

Applies hierarchical clustering to centroids of clusters to reorder.

Usage

reorder_clusters_hclust(
  clust_dt,
  hclust_result = NULL,
  row_ = "id",
  column_ = "x",
  fill_ = "y",
  facet_ = "sample",
  cluster_ = "cluster_id",
  reapply_cluster_names = TRUE,
  return_hclust = FALSE
)

Arguments

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 10)
new_dt = reorder_clusters_hclust(clust_dt)
cowplot::plot_grid(
    ssvSignalHeatmap(clust_dt),
    ssvSignalHeatmap(new_dt)
)

Description

Manually applies a new order (top to bottom) for cluster using the result of ssvSignalClustering.

Usage

reorder_clusters_manual(
  clust_dt,
  manual_order,
  row_ = "id",
  cluster_ = "cluster_id",
  reapply_cluster_names = TRUE
)

Arguments

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 3)
new_dt = reorder_clusters_manual(clust_dt = clust_dt, manual_order = 2)
cowplot::plot_grid(
    ssvSignalHeatmap(clust_dt),
    ssvSignalHeatmap(new_dt)
)

Description

Attempts to reorder clusters so that rows with highest signal on the left relative to the right appear at the top. Signal should have a roughly diagonal pattern in a "stepdown" pattern.

Usage

reorder_clusters_stepdown(
  clust_dt,
  row_ = "id",
  column_ = "x",
  fill_ = "y",
  facet_ = "sample",
  cluster_ = "cluster_id",
  reapply_cluster_names = TRUE,
  step_by_column = TRUE,
  step_by_facet = FALSE
)

Arguments

Details

This can be down by column (step_by_column = TRUE) which averages across facets. By facet (step_by_column = FALSE, step_by_facet = TRUE) which averages all columns per facet. Or both column and facet (step_by_column = TRUE, step_by_facet = TRUE), which does no averaging so it looks at the full matrix as plotted.

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 10)
new_dt = reorder_clusters_stepdown(clust_dt)
cowplot::plot_grid(
    ssvSignalHeatmap(clust_dt),
    ssvSignalHeatmap(new_dt)
)

Description

reverse_clusters

Usage

reverse_clusters(
  clust_dt,
  row_ = "id",
  column_ = "x",
  fill_ = "y",
  facet_ = "sample",
  cluster_ = "cluster_id",
  reverse_rows_within = TRUE,
  reapply_cluster_names = TRUE
)

Arguments

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
set.seed(0)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 3)
rev_dt = reverse_clusters(clust_dt)
rev_dt.no_relabel = reverse_clusters(clust_dt, reapply_cluster_names = FALSE)
rev_dt.not_rows = reverse_clusters(clust_dt, reverse_rows_within = FALSE)
cowplot::plot_grid(nrow = 1,
  ssvSignalHeatmap(clust_dt) + labs(title = "original"),
  ssvSignalHeatmap(rev_dt) + labs(title = "reversed"),
  ssvSignalHeatmap(rev_dt.no_relabel) + labs(title = "reversed, no relabel"),
  ssvSignalHeatmap(rev_dt.not_rows) + labs(title = "reversed, not rows")
)

Description

Allows RColorBrew to handle n values less than 3 and greater than 8 without warnings and return expected number of colors.

Usage

safeBrew(n, pal = "Dark2")

Arguments

Details

For convenience, instead of the number n requested, n may be a character or factor vector and outputs will be appropriately named for use with scale_color/fill_manual.

Additionally, accepts pal as "gg", "ggplot", or "ggplot2" to reproduce default ggplot colors in the same way.

Value

a character vector of hex coded colors of length n from the color brewer palette pal. If n is supplied as character or factor, output will be named accordingly.

Examples

plot(1:2, rep(0, 2),  col = safeBrew(2, "dark2"), pch = 16, cex = 6)
plot(1:12, rep(0, 12),  col = safeBrew(12, "set1"), pch = 16, cex = 6)
plot(1:12, rep(0, 12),  col = safeBrew(12, "set2"), pch = 16, cex = 6)
plot(1:12, rep(0, 12),  col = safeBrew(12, "set3"), pch = 16, cex = 6)

Description

convert a list of sets, each list item should be a character vector denoting items in sets

Usage

set_list2memb(set_list)

Arguments

Value

converts list of characters/numeric to membership table matrix

Description

orients the relative position of x's zero value and extends ranges to be contiguous

Usage

shift_anchor(score_dt, window_size, anchor)

Arguments

Value

score_dt with x values shifted appropriately and start and end extended to make ranges contiguous

Description

Splits one specified cluster in number of new clusters determined by nclust

Usage

split_cluster(
  clust_dt,
  to_split,
  nclust = 2,
  row_ = "id",
  column_ = "x",
  fill_ = "y",
  facet_ = "sample",
  cluster_ = "cluster_id",
  reapply_cluster_names = TRUE
)

Arguments

Value

data.table as output from ssvSignalClustering

Examples

data(CTCF_in_10a_profiles_dt)
set.seed(0)
clust_dt = ssvSignalClustering(CTCF_in_10a_profiles_dt, nclust = 3)
split_dt = split_cluster(clust_dt, to_split = 2, nclust = 3)
split_dt.no_rename = split_cluster(
  clust_dt,
  to_split = 2,
  nclust = 3,
  reapply_cluster_names = FALSE
)
cowplot::plot_grid(nrow = 1,
  ssvSignalHeatmap(clust_dt),
  ssvSignalHeatmap(split_dt),
  ssvSignalHeatmap(split_dt.no_rename)
)

Description

ssv_mclapply

Usage

ssv_mclapply(X, FUN, mc.cores = getOption("mc.cores", 1), ...)

Arguments

Value

result of either pblapply or pbmclapply

Description

ssvAnnotateSubjectGRanges

Usage

ssvAnnotateSubjectGRanges(
  annotation_source,
  subject_gr,
  annotation_name = NULL,
  multi_resolver_FUN = "default"
)

## S4 method for signature 'GRanges'
ssvAnnotateSubjectGRanges(
  annotation_source,
  subject_gr,
  annotation_name = NULL,
  multi_resolver_FUN = "default"
)

## S4 method for signature 'list'
ssvAnnotateSubjectGRanges(
  annotation_source,
  subject_gr,
  annotation_name = NULL,
  multi_resolver_FUN = "default"
)

## S4 method for signature 'GRangesList'
ssvAnnotateSubjectGRanges(
  annotation_source,
  subject_gr,
  annotation_name = NULL,
  multi_resolver_FUN = "default"
)

Arguments

Value

GRanges with the same regions as subject_gr but with addtional mcols added from annotation_source.

Examples

library(GenomicRanges)
data(CTCF_in_10a_narrowPeak_grs)
np_grs = CTCF_in_10a_narrowPeak_grs
olap_gr = ssvOverlapIntervalSets(np_grs)
# annotating with a signle GRanges is OK
ssvAnnotateSubjectGRanges(np_grs$MCF10A_CTCF, olap_gr)
# provide a name if that's useful
ssvAnnotateSubjectGRanges(np_grs$MCF10A_CTCF, olap_gr,
  annotation_name = "MCF10A")
# a named list adds each annotation
ssvAnnotateSubjectGRanges(np_grs, olap_gr)
# overriding list names is an option
ssvAnnotateSubjectGRanges(np_grs, olap_gr, LETTERS[1:3])
# GRangeList are handled like a standard list
ssvAnnotateSubjectGRanges(GRangesList(np_grs), olap_gr, LETTERS[1:3])

Description

In constrast to ssvOverlapIntervalSets, only regions where a consensus of input grs are present are preserved and annotated.

Usage

ssvConsensusIntervalSets(
  grs,
  ext = 0,
  min_number = 2,
  min_fraction = 0.5,
  preserve_mcols = FALSE,
  ...
)

Arguments

Details

Only the most stringent of min_number or min_fraction will be applied.

Value

GRanges with metadata columns describing consensus overlap of input grs.

Examples

library(GenomicRanges)
a = GRanges("chr1", IRanges(1:7*10, 1:7*10))
b = GRanges("chr1", IRanges(5:10*10, 5:10*10))
ssvConsensusIntervalSets(list(a, b))

Description

see ssvMakeMembTable

Usage

ssvFactorizeMembTable(object)

Arguments

Value

a 2 column ("id" and "group") data.frame. "id" is factor of item names if any or simply order of items. "group" is a factor of set combinations

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFactorizeMembTable(CTCF_in_10a_overlaps_gr)
ssvFactorizeMembTable(list(1:4, 2:3, 4:6))

Description

bar plots of set sizes

Usage

ssvFeatureBars(
  object,
  show_counts = TRUE,
  bar_colors = NULL,
  counts_text_colors = NULL,
  return_data = FALSE,
  count_label_size = 8
)

Arguments

Value

ggplot of bar plot of set sizes

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeatureBars(list(1:3, 2:6))
ssvFeatureBars(CTCF_in_10a_overlaps_gr, count_label_size = 10)
ssvFeatureBars(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])

Description

Outputs a ggplot binary heatmap, where color indicates TRUE and the other indicates FALSE in a membership table. The heatmap is sorted, TRUE at the top, by column left to right. Changes to column order can reveal different patterns.

Usage

ssvFeatureBinaryHeatmap(
  object,
  raster_approximation = TRUE,
  true_color = "black",
  false_color = "#EFEFEF",
  raster_width_min = 1000,
  raster_height_min = 1000,
  return_data = FALSE
)

Arguments

Details

As a svg output, the final plot can be unwieldy. The default of raster_approximation = TRUE is easier to work with, especially for larger membership tables.

Value

ggplot using geom_tile of membership table sorted from left to right.

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeatureBinaryHeatmap(list(1:3, 2:6))
# horizontal version
ssvFeatureBinaryHeatmap(list(1:3, 2:6)) + coord_flip() +
  theme(axis.text.x = element_blank(), axis.text.y = element_text())
ssvFeatureBinaryHeatmap(CTCF_in_10a_overlaps_gr)
ssvFeatureBinaryHeatmap(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])
ssvFeatureBinaryHeatmap(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,3:2])

Description

Try to load a bed-like file and convert it to a GRanges object

Usage

ssvFeatureEuler(
  object,
  line_width = 2,
  shape = c("circle", "ellipse")[1],
  n_points = 200,
  fill_alpha = 0.3,
  line_alpha = 1,
  circle_colors = NULL,
  return_data = FALSE
)

Arguments

Value

ggplot of venneuler results

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeatureEuler(list(1:3, 2:6))
ssvFeatureEuler(CTCF_in_10a_overlaps_gr)
ssvFeatureEuler(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])

Description

Generate a ggplot pie plot of set sizes.

Usage

ssvFeaturePie(object, slice_colors = NULL, return_data = FALSE)

Arguments

Value

ggplot pie graph of set sizes

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeaturePie(list(1:3, 2:6))
ssvFeaturePie(CTCF_in_10a_overlaps_gr)
ssvFeaturePie(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])

Description

Uses the UpSetR package to create an UpSetR::upset plot of region overlaps.

Usage

ssvFeatureUpset(
  object,
  return_UpSetR = FALSE,
  nsets = NULL,
  nintersects = 15,
  order.by = "freq",
  ...
)

Arguments

Value

ggplot version of UpSetR plot

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeatureUpset(list(1:3, 2:6))
ssvFeatureUpset(CTCF_in_10a_overlaps_gr)
ssvFeatureUpset(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])

Description

ggplot implementation of vennDiagram from limma package. Currently limited at 3 sets. ssvFeatureUpset and ssvFeatureBinaryHeatmap are good options for more than 3 sets. ssvFeatureEuler can work too but can take a very long time to run for more than 5 or so.

Usage

ssvFeatureVenn(
  object,
  group_names = NULL,
  counts_txt_size = 5,
  counts_as_labels = FALSE,
  show_outside_count = FALSE,
  line_width = 3,
  circle_colors = NULL,
  fill_alpha = 0.3,
  line_alpha = 1,
  counts_color = NULL,
  counts_as_percent = FALSE,
  percentage_digits = 1,
  percentage_suffix = "%",
  n_points = 200,
  return_data = FALSE
)

Arguments

Value

ggplot venn diagram

Examples

data(CTCF_in_10a_overlaps_gr)
ssvFeatureVenn(list(1:3, 2:6))
ssvFeatureVenn(CTCF_in_10a_overlaps_gr)
ssvFeatureVenn(S4Vectors::mcols(CTCF_in_10a_overlaps_gr)[,2:3])

ssvFeatureVenn(list(1:3, 2:6),
  counts_as_percent = TRUE,
  percentage_digits = 2)

ssvFeatureVenn(list(1:3, 2:6),
  counts_as_percent = TRUE,
  percentage_digits = 0,
  percentage_suffix = " %",
  counts_txt_size = 12)

Description

ssvFetchBam iteratively calls fetchWindowedBam.single. See ssvFetchBam.single for more info.

Usage

ssvFetchBam(
  file_paths,
  qgr,
  unique_names = NULL,
  names_variable = "sample",
  file_attribs = NULL,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  fragLens = "auto",
  target_strand = c("*", "+", "-", "both")[1],
  flip_strand = FALSE,
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  max_dupes = Inf,
  splice_strategy = c("none", "ignore", "add", "only", "splice_count")[1],
  n_cores = getOption("mc.cores", 1),
  n_region_splits = 1,
  return_unprocessed = FALSE,
  force_skip_centerFix = FALSE,
  ...
)

Arguments

Details

if qgr contains the range chr1:1-100 and win_size is 10, values from positions chr1 5,15,25...85, and 95 will be retrieved from bw_file

Value

A tidy formatted GRanges (or data.table if specified) containing fetched values.

Examples

if(Sys.info()['sysname'] != "Windows"){
data(CTCF_in_10a_overlaps_gr)
library(GenomicRanges)
bam_f = system.file("extdata/test.bam",
    package = "seqsetvis", mustWork = TRUE)
bam_files = c("a" = bam_f, "b" = bam_f)
qgr = CTCF_in_10a_overlaps_gr[1:5]
bw_gr = ssvFetchBam(bam_files, qgr, win_size = 10)
bw_gr2 = ssvFetchBam(as.list(bam_files), qgr, win_size = 10)

bw_dt = ssvFetchBam(bam_files, qgr, win_size = 10,
    return_data.table = TRUE)
}

Description

fetch a windowed version of a bam file, returns GRanges

Usage

ssvFetchBam.single(
  bam_f,
  qgr,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  fragLen = NULL,
  target_strand = c("*", "+", "-", "both")[1],
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  max_dupes = Inf,
  splice_strategy = c("none", "ignore", "add", "only", "splice_count")[1],
  flip_strand = FALSE,
  return_unprocessed = FALSE,
  force_skip_centerFix = FALSE,
  ...
)

Arguments

Value

tidy GRanges (or data.table if specified) with pileups from bam file. pileup is calculated only every win_size bp.

Description

Iterates a character vector (ideally named) and calls ssvFetchBamPE.single on each. Appends grouping variable to each resulting data.table and uses rbindlist to efficiently combine results

Usage

ssvFetchBamPE(
  file_paths,
  qgr,
  unique_names = NULL,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  fragLens = "not_used",
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  names_variable = "sample",
  return_data.table = FALSE,
  max_dupes = Inf,
  n_cores = getOption("mc.cores", 1),
  n_region_splits = 1,
  min_isize = 1,
  max_isize = Inf,
  return_unprocessed = FALSE,
  return_fragSizes = FALSE,
  force_skip_centerFix = FALSE,
  ...
)

Arguments

Details

#' In contrast to ssvFetchBam, extension of reads to estimated fragment size is not an issue as each read pair represents a fragment of exact size.

ssvFetchBamPE iteratively calls fetchWindowedBam.single. See ssvFetchBamPE.single for more info.

if qgr contains the range chr1:1-100 and win_size is 10, values from positions chr1 5,15,25...85, and 95 will be retrieved from bw_file

Value

A tidy formatted GRanges (or data.table if specified) containing fetched values.

Examples

if(Sys.info()['sysname'] != "Windows"){
library(GenomicRanges)
bam_f = system.file("extdata/Bcell_PE.mm10.bam",
    package = "seqsetvis", mustWork = TRUE)
bam_files = c("a" = bam_f, "b" = bam_f)
data("Bcell_peaks")
qgr = Bcell_peaks
bw_gr = ssvFetchBamPE(bam_files, qgr, win_size = 10)
bw_gr2 = ssvFetchBamPE(as.list(bam_files), qgr, win_size = 10)

bw_dt = ssvFetchBamPE(bam_files, qgr, win_size = 10,
    return_data.table = TRUE)
}

Description

ssvFetchBamPE.RNA

Usage

ssvFetchBamPE.RNA(
  file_paths,
  qgr,
  unique_names = NULL,
  win_size = 50,
  target_strand = "both",
  absolute_strand = FALSE,
  splice_strategy = "ignore",
  return_data.table = FALSE,
  win_method = "sample",
  max_dupes = Inf,
  flip_strand = FALSE,
  sum_reads = TRUE,
  n_cores = getOption("mc.cores", 1),
  force_skip_centerFix = TRUE,
  n_region_splits = 1
)

Arguments

Value

A tidy formatted GRanges (or data.table if specified) containing fetched values.

Examples

library(GenomicRanges)
pkg_dir = system.file(package = "seqsetvis", "extdata", mustWork = TRUE)
bam_files_esr1 = dir(pkg_dir, pattern = "H1.+R1.ESR1_RNA.+bam$", full.names = TRUE)
names(bam_files_esr1) = sub("_R.+", "", basename(bam_files_esr1))
query_gr = GenomicRanges::GRanges("chr6:151656691-152129619:+")
query_gr = GenomicRanges::GRanges("chr6:152116691-152129619:+")

strand(query_gr) = "+"

prof_dt = ssvFetchBamPE.RNA(bam_files_esr1, query_gr, return_data.table = TRUE, win_size = 1)
prof_dt

Description

fetch a windowed version of a paired-end bam file, returns GRanges In contrast to ssvFetchBam, extension of reads to estimated fragment size is not an issue as each read pair represents a fragment of exact size.

Usage

ssvFetchBamPE.single(
  bam_f,
  qgr,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  max_dupes = Inf,
  min_isize = 1,
  max_isize = Inf,
  return_unprocessed = FALSE,
  return_fragSizes = FALSE,
  force_skip_centerFix = FALSE,
  ...
)

Arguments

Value

tidy GRanges (or data.table if specified) with pileups from bam file. pileup is calculated only every win_size bp.

Description

ssvFetchBigwig iteratively calls fetchWindowedBigwig.single. See ssvFetchBigwig.single for more info.

Usage

ssvFetchBigwig(
  file_paths,
  qgr,
  unique_names = NULL,
  names_variable = "sample",
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  fragLens = "not_used",
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  n_cores = getOption("mc.cores", 1),
  n_region_splits = 1,
  force_skip_centerFix = FALSE
)

Arguments

Details

if qgr contains the range chr1:1-100 and win_size is 10, values from positions chr1 5,15,25...85, and 95 will be retrieved from bw_file

Value

A tidy formatted GRanges (or data.table if specified) containing fetched values.

Examples

if(Sys.info()['sysname'] != "Windows"){
library(GenomicRanges)
bw_f = system.file("extdata/test_loading.bw",
    package = "seqsetvis", mustWork = TRUE)
bw_files = c("a" = bw_f, "b" = bw_f)
qgr = GRanges("chrTest", IRanges(1, 30))
bw_gr = ssvFetchBigwig(bw_files, qgr, win_size = 10)
bw_gr2 = ssvFetchBigwig(as.list(bw_files), qgr, win_size = 10)

bw_dt = ssvFetchBigwig(bw_files, qgr, win_size = 10,
    return_data.table = TRUE)
}

Description

ssvFetchBigwig.single Gets values for each region of the query GRanges (qgr). Values correspond to the center of each window of size win_size. A tidy formatted data.table object is returned suitable for plotting using ggplots.

Usage

ssvFetchBigwig.single(
  bw_file,
  qgr,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = stats::weighted.mean,
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  force_skip_centerFix = FALSE
)

Arguments

Details

if qgr contains the range chr1:1-100 and win_size is 10, values from positions chr1 5,15,25...85, and 95 will be retrieved from bw_file

Value

A GRanges (or data.table if specified) containing fetched values.

Description

ssvFetchGRanges Gets coverage values for each region of the query GRanges (qgr). Values correspond to the center of each window of size win_size. A tidy formatted data.table object is returned suitable for plotting using ggplots.

Usage

ssvFetchGRanges(
  grs,
  qgr,
  file_attribs = data.frame(matrix(0, nrow = length(grs), ncol = 0)),
  unique_names = names(grs),
  names_variable = "sample",
  win_size = 50,
  win_method = c("sample", "summary")[1],
  summary_FUN = function(x, w) max(x),
  target_strand = c("*", "+", "-", "both")[1],
  use_coverage = NULL,
  attrib_var = "score",
  fill_value = 0,
  anchor = c("left", "left_unstranded", "center", "center_unstranded")[3],
  return_data.table = FALSE,
  n_cores = getOption("mc.cores", 1),
  force_skip_centerFix = FALSE
)

Arguments

Value

A tidy formatted GRanges (or data.table if specified) containing fetched values.

Examples

data(CTCF_in_10a_narrowPeak_grs)
data(CTCF_in_10a_overlaps_gr)
ssvFetchGRanges(CTCF_in_10a_narrowPeak_grs, CTCF_in_10a_overlaps_gr, win_size = 200)

Description

Does nothing unless load_signal is overridden to carry out reading data from file_paths (likely via the appropriate ssvFetch* function, ie. ssvFetchBigwig or ssvFetchBam

Usage

ssvFetchSignal(
  file_paths,
  qgr,
  unique_names = NULL,
  names_variable = "sample",
  file_attribs = NULL,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  return_data.table = FALSE,
  load_signal = function(f, nam, qgr) {
     warning("nothing happened, ",
    "supply a function to", "load_signal parameter.")
 },
  n_cores = getOption("mc.cores", 1),
  n_region_splits = 1,
  force_skip_centerFix = FALSE
)

Arguments

Details

load_signal is passed f, nam, and qgr and is executed in the environment where load_signal is defined. See ssvFetchBigwig and ssvFetchBam for examples.

Value

A GRanges with values read from file_paths at intervals of win_size. Originating file is coded by unique_names and assigned to column of name names_variable. Output is data.table is return_data.table is TRUE.

Examples

library(GenomicRanges)
data(CTCF_in_10a_overlaps_gr)
bam_f = system.file("extdata/test.bam",
    package = "seqsetvis", mustWork = TRUE)
bam_files = c("a" = bam_f, "b" = bam_f)
qgr = CTCF_in_10a_overlaps_gr[1:2]
qgr = resize(qgr, 500, "center")
load_bam = function(f, nam, qgr) {
    message("loading ", f, " ...")
    dt = seqsetvis:::ssvFetchBam.single(bam_f = f,
                      qgr = qgr,
                      win_size = 50,
                      fragLen = NULL,
                      target_strand = "*",
                      return_data.table = TRUE)

    data.table::set(dt, j = "sample", value = nam)
    message("finished loading ", nam, ".")
    dt
}
ssvFetchSignal(bam_files, qgr, load_signal = load_bam)