Package 'nucim' reference manual

Title:	Nucleome Imaging Toolbox
Description:	Tools for 4D nucleome imaging. Quantitative analysis of the 3D nuclear landscape recorded with super-resolved fluorescence microscopy. See Volker J. Schmid, Marion Cremer, Thomas Cremer (2017) <doi:10.1016/j.ymeth.2017.03.013>.
Authors:	Volker Schmid [aut, cre]
Maintainer:	Volker Schmid <[email protected]>
License:	GPL-3
Version:	1.0.12
Built:	2024-11-11 05:15:35 UTC
Source:	https://github.com/bioimaginggroup/nucim

Barplot with Intervals for two or three bars beside

Description

Barplot with Intervals for two or three bars beside

Usage

barplot_with_interval_23(x, method = "minmax", qu = c(0, 1), ylim = NULL, ...)
barplot_with_interval_23(x, method = "minmax", qu = c(0, 1), ylim = NULL, ...)

Arguments

`x`	array
`method`	method for intervals: "minmax" (default), "quantile" or "sd"
`qu`	vector of two quantiles for method="quantile
`ylim`	limits for y axis. Default:NULL is ylim=c(0,max(interval))
`...`	additional parameters forwarded to barplot

Value

plot

Class neighbourhood distribution

Description

Class neighbourhood distribution

Usage

class.neighbours(img, N, N.max = 7, cores = 1)
class.neighbours(img, N, N.max = 7, cores = 1)

Arguments

`img`	Class image
`N`	which class
`N.max`	maximum class (default: 7)
`cores`	number of cores used in parallel (needs parallel package)

Value

vector of length N.max

Classify DAPI

Description

These functions are provided for compatibility with older version of the nucim package. They may eventually be completely removed.

Usage

classify.single(...)
classify.single(...)

Arguments

...

parameters for classify

Value

image with classes

Count classes in classified image

Description

Count classes in classified image

Usage

classify.table(class, N)
classify.table(class, N)

Arguments

`class`	classes image
`N`	number of classes

Value

table with number of voxels per class

Compute colors in classes distribution

Description

Compute colors in classes distribution

Usage

colors.in.classes(
  classes,
  color1,
  color2 = NULL,
  mask = array(TRUE, dim(classes)),
  N = max(classes, na.rm = TRUE),
  type = "tresh",
  thresh1 = NULL,
  thresh2 = NULL,
  sd1 = 2,
  sd2 = 2,
  col1 = "green",
  col2 = "red",
  test = FALSE,
  plot = TRUE,
  beside = TRUE,
  ylim = NULL,
  verbose = FALSE,
  ...
)
colors.in.classes(
  classes,
  color1,
  color2 = NULL,
  mask = array(TRUE, dim(classes)),
  N = max(classes, na.rm = TRUE),
  type = "tresh",
  thresh1 = NULL,
  thresh2 = NULL,
  sd1 = 2,
  sd2 = 2,
  col1 = "green",
  col2 = "red",
  test = FALSE,
  plot = TRUE,
  beside = TRUE,
  ylim = NULL,
  verbose = FALSE,
  ...
)

Arguments

`classes`	Image of classes
`color1`	Image of first color
`color2`	Image of second color
`mask`	Image mask
`N`	Maximum number of classes
`type`	Type of spot definition, see details
`thresh1`	Threshold for first color image
`thresh2`	Threshold for second color image
`sd1`	For automatic threshold, that is: mean(color1)+sd1*sd(color1)
`sd2`	For automatic threshold of color2
`col1`	Name of color 1
`col2`	Name of color 2
`test`	Compute tests: "Wilcoxon" for Wilcoxon rank-sum (Mann-Whitney U), chisq for Chi-squared test
`plot`	Plot barplots
`beside`	a logical value. If FALSE, the columns of height are portrayed as stacked bars, and if TRUE the columns are portrayed as juxtaposed bars.
`ylim`	limits for the y axis (plot)
`verbose`	verbose mode
`...`	additional plotting parameters

Details

Type of spot definitions: "thresh" or "t": Threshold based (threshold can be given by thresh1/2 or automatically derived) "voxel" or "v": Spots are given as binary voxel mask "intensity" or "i": Voxels are weighted with voxel intensity. Intensity is scaled to [0,1] after subtracting thresh1/2 (or automatic threshold)

Value

Table of classes with color 1 (and 2)

Compute colors in classes distribution for folders

Description

Compute colors in classes distribution for folders

Usage

colors.in.classes.folder(
  path,
  color1,
  color2 = NULL,
  N = 7,
  type = "intensity",
  thresh1 = NULL,
  thresh2 = NULL,
  sd1 = 2,
  sd2 = 2,
  col1 = "green",
  col2 = "red",
  cores = 1,
  verbose = FALSE
)
colors.in.classes.folder(
  path,
  color1,
  color2 = NULL,
  N = 7,
  type = "intensity",
  thresh1 = NULL,
  thresh2 = NULL,
  sd1 = 2,
  sd2 = 2,
  col1 = "green",
  col2 = "red",
  cores = 1,
  verbose = FALSE
)

Arguments

`path`	Path to root folder
`color1`	Image of first color
`color2`	Image of second color
`N`	Maximum number of classes
`type`	Type of spot definition, see details
`thresh1`	Threshold for first color image
`thresh2`	Threshold for second color image
`sd1`	For automatic threshold, that is: mean(color1)+sd1*sd(color1)
`sd2`	For automatic threshold of color2
`col1`	Name of color 1
`col2`	Name of color 2
`cores`	Number of cores used in parallel, cores=1 implies no parallelization
`verbose`	verbose mode

Value

Results are in folder colorsinclasses

Compute distance to border of classes

Description

Compute distance to border of classes

Usage

compute.distance2border(
  f,
  color,
  N,
  from.spots = FALSE,
  output = "dist2border",
  cores = 1
)
compute.distance2border(
  f,
  color,
  N,
  from.spots = FALSE,
  output = "dist2border",
  cores = 1
)

Arguments

`f`	folder of classes images
`color`	folder of color images ("spots-"color for spots images)
`N`	which class
`from.spots`	Logical.
`output`	output folder
`cores`	number of parallel cores which can be used

Value

images in output"-"color"-"N

Mask DAPI in kernel

Description

Mask DAPI in kernel

Usage

dapimask(
  img,
  size = NULL,
  voxelsize = NULL,
  thresh = "auto",
  silent = TRUE,
  cores = 1
)
dapimask(
  img,
  size = NULL,
  voxelsize = NULL,
  thresh = "auto",
  silent = TRUE,
  cores = 1
)

Arguments

`img`	DAPI channel image (3d)
`size`	size of img in microns
`voxelsize`	size of voxel in microns
`thresh`	threshold for intensity. Can be "auto": function will try to find automatic threshold
`silent`	Keep silent?
`cores`	number of cores available for parallel computing

Value

mask image, array with same dimension as img.

Automatic DAPI mask segmentation for files

Description

Automatic DAPI mask segmentation for files

Usage

dapimask.file(
  file,
  folder = "blue",
  voxelsize = NULL,
  size = NULL,
  silent = FALSE,
  cores = 1
)
dapimask.file(
  file,
  folder = "blue",
  voxelsize = NULL,
  size = NULL,
  silent = FALSE,
  cores = 1
)

Arguments

`file`	file to read
`folder`	with
`voxelsize`	real size of voxel (in microns), if NULL (default), look in folder XYZmic
`size`	real size of image (in microns), if NULL (default), look in folder XYZmic
`silent`	Keep silent?
`cores`	Number of cores available for parallel computing

Value

nothing, DAPI mask image will be saved to dapimask/

Automatic DAPI mask segmentation for folder

Description

Automatic DAPI mask segmentation for folder

Usage

dapimask.folder(
  path,
  folder = "blue",
  voxelsize = NULL,
  size = NULL,
  cores = 1
)
dapimask.folder(
  path,
  folder = "blue",
  voxelsize = NULL,
  size = NULL,
  cores = 1
)

Arguments

`path`	path to folder with DAPI
`folder`	folder with DAPI images
`voxelsize`	real size of voxel (in microns), if NULL (default), look in folder XYZmic
`size`	real size of image (in microns), if NULL (default), look in folder XYZmic
`cores`	number of cores to use in parallel (need parallel package)

Value

nothing, results are in folder dapimask

Detects spots for one file

Description

Detects spots for one file

Usage

find.spots.file(
  file,
  dir,
  color,
  thresh = NULL,
  thresh.auto = FALSE,
  thresh.quantile = 0.9,
  filter = NULL,
  cores = 1
)
find.spots.file(
  file,
  dir,
  color,
  thresh = NULL,
  thresh.auto = FALSE,
  thresh.quantile = 0.9,
  filter = NULL,
  cores = 1
)

Arguments

`file`	file
`dir`	directory for results
`color`	which color, images have to be in folder with color name
`thresh`	threshold
`thresh.auto`	Logical. Find threshold automatically?
`thresh.quantile`	numeric. use simple
`filter`	2d-filter to use before spot detection
`cores`	number of cores to use in parallel (with parallel package only)

Value

spot images in spot-color/, number of spots as txt files in spot-color/

Detects spots

Description

Detects spots

Usage

find.spots.folder(
  f,
  color,
  thresh = 1,
  thresh.auto = TRUE,
  filter = NULL,
  cores = 1
)
find.spots.folder(
  f,
  color,
  thresh = 1,
  thresh.auto = TRUE,
  filter = NULL,
  cores = 1
)

Arguments

`f`	path to folder
`color`	which color, images have to be in folder with color name
`thresh`	threshold
`thresh.auto`	Logical. Find threshold automatically?
`filter`	2d-filter to use before spot detection
`cores`	number of cores to use in parallel (with parallel package only)

Value

spot images in spot-color/, number of spots as txt files in spot-color/

Heatmap colors for n classes

Description

Heatmap colors for n classes

Usage

heatmap.color(n)
heatmap.color(n)

Arguments

`n`	number of colors.

Examples

barplot(8:1,col=heatmap.color(8))
barplot(8:1,col=heatmap.color(8))

Heatmap colors for 7 classes

Description

Heatmap colors for 7 classes

Usage

heatmap7(...)
heatmap7(...)

Arguments

...

parameters are ignored.

Examples

barplot(7:1,col=heatmap7())
barplot(7:1,col=heatmap7())

Find all distances to next neighbour of all classes for folders

Description

Find all distances to next neighbour of all classes for folders

Usage

nearestClassDistances.folder(
  path,
  N = 7,
  voxelsize = NULL,
  add = FALSE,
  cores = 1
)
nearestClassDistances.folder(
  path,
  N = 7,
  voxelsize = NULL,
  add = FALSE,
  cores = 1
)

Arguments

`path`	path to folder
`N`	number of classes, default: 7
`voxelsize`	real size of voxels (in microns), if NULL (default), look in folder XYZmic
`add`	if TRUE, only process images which have not been processed before (i.e. have been added to classN)
`cores`	number of cores to use in parallel (needs parallel package if cores>1)

Value

nothing, results are in folder distances in RData format

Plot barplot for classified images in a folder

Description

Plot barplot for classified images in a folder

Usage

plot_classify.folder(
  path,
  N = 7,
  cores = 1,
  col = grDevices::grey(0.7),
  method = "sd"
)
plot_classify.folder(
  path,
  N = 7,
  cores = 1,
  col = grDevices::grey(0.7),
  method = "sd"
)

Arguments

`path`	path to folder
`N`	number of classes, default: 7
`cores`	number of cores to use in parallel (needs parallel package if cores>1)
`col`	color of bars, either one or a vector of hex RGB characters
`method`	method for error bars ("sd", "minmax", "quartile")

Value

plots

Plot for colors in classes distribution for folders

Description

Plot for colors in classes distribution for folders

Usage

plot_colors.in.classes.folder(path, col1 = "green", col2 = "red")
plot_colors.in.classes.folder(path, col1 = "green", col2 = "red")

Arguments

`path`	path to folder
`col1`	color of channel 1
`col2`	color of channel 2

Value

plot

Plots all distances to next neighbour of all classes for folders

Description

Plots all distances to next neighbour of all classes for folders

Usage

plot_nearestClassDistances.folder(
  path,
  N = 7,
  cores = 1,
  method = "quantile",
  qu = 0.01
)
plot_nearestClassDistances.folder(
  path,
  N = 7,
  cores = 1,
  method = "quantile",
  qu = 0.01
)

Arguments

`path`	path to folder
`N`	number of classes, default: 7
`cores`	number of cores to use in parallel (needs parallel package if cores>1)
`method`	method for summarizing distances, either "min" or "quantile"
`qu`	quantile for method="quantile", default: 0.01

Value

plots

Split RGB channels

Description

Split RGB channels

Usage

splitchannel(img, preprocess = TRUE)
splitchannel(img, preprocess = TRUE)

Arguments

`img`	rgb image
`preprocess`	logical. Should preprocessing be applied?

Value

list with red, green, blue channels and size in microns.

Split RGB images into channels and pixel size information

Description

These functions are provided for compatibility with older version of the nucim package. They may eventually be completely removed.

Usage

splitchannels(...)
splitchannels(...)

Arguments

...

parameters for splitchannels.folder

Value

Nothing, folders red, green, blue and XYZmic include separate channels and pixel size information

Split channels into files and extracts size in microns

Description

Split channels into files and extracts size in microns

Usage

splitchannels.file(file, channels, rgb.folder, normalize = FALSE)
splitchannels.file(file, channels, rgb.folder, normalize = FALSE)

Arguments

`file`	file name
`channels`	e.g. c("red","green","blue")
`rgb.folder`	folder with file
`normalize`	boolean. Should we try to do normalization?

Value

files in "./red/", "./green", "./blue" and "./XYZmic"

Split RGB images into channels and pixel size information

Description

Split RGB images into channels and pixel size information

Usage

splitchannels.folder(
  path,
  channels = c("red", "green", "blue"),
  rgb.folder = "rgb",
  normalize = FALSE,
  cores = 1
)
splitchannels.folder(
  path,
  channels = c("red", "green", "blue"),
  rgb.folder = "rgb",
  normalize = FALSE,
  cores = 1
)

Arguments

`path`	Path to root folder
`channels`	Vector of channels in images
`rgb.folder`	Folder with RGB images
`normalize`	boolean. Should we try to do normalization
`cores`	Number of cores used in parallel, cores=1 implies no parallelization

Value

Nothing, folders red, green, blue and XYZmic include separate channels and pixel size information

Examples

splitchannels.folder("./")

splitchannels.folder("./")

Find spots using information from two channels

Description

Find spots using information from two channels

Usage

spots.combined(
  red,
  green,
  mask,
  size = NULL,
  voxelsize = NULL,
  thresh.offset = 0.1,
  window = c(5, 5),
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = NA,
  full.voxel = FALSE
)
spots.combined(
  red,
  green,
  mask,
  size = NULL,
  voxelsize = NULL,
  thresh.offset = 0.1,
  window = c(5, 5),
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = NA,
  full.voxel = FALSE
)

Arguments

`red`	image
`green`	image
`mask`	image mask
`size`	size of img in microns
`voxelsize`	size of voxel in microns
`thresh.offset`	Thresh offset used in EBImage::thresh()
`window`	Half width and height of the moving rectangular window.
`min.sum.intensity`	spots smaller than min.sum.intensity are ignored
`max.distance`	use only spots with distance to other color spot smaller than max.distance
`use.brightest`	Logical; use only brightest in max.distance?
`max.spots`	maximum of spots (per channel), only when use brightest=TRUE
`full.voxel`	Logical; output contains full voxel instead of rgb intensities

Value

RGB image with spots will be written to output folder

Find spots using information from two channels

Description

Find spots using information from two channels

Usage

spots.combined.file(
  file,
  size = NULL,
  voxelsize = NULL,
  folder = "./",
  thresh.offset = 0.1,
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = 2,
  full.voxel = FALSE,
  output = "markers"
)
spots.combined.file(
  file,
  size = NULL,
  voxelsize = NULL,
  folder = "./",
  thresh.offset = 0.1,
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = 2,
  full.voxel = FALSE,
  output = "markers"
)

Arguments

`file`	File name
`size`	size of img in microns, if size and voxelsize are NULL, size is determined from folder XYZmic
`voxelsize`	size of voxel in microns
`folder`	Folder
`thresh.offset`	Thresh offset used in EBImage::thresh()
`min.sum.intensity`	spots smaller than min.sum.intensity are ignored
`max.distance`	use only spots with distance to other color spot smaller than max.distance
`use.brightest`	Logical; use only brightest in max.distance?
`max.spots`	maximum of spots (per channel), only when use brightest=TRUE
`full.voxel`	Logical; output contains full voxel instead of rgb intensities
`output`	output folder

Value

RGB image with spots will be written to output folder

Find spots using information from two channels for folder

Description

Find spots using information from two channels for folder

Usage

spots.combined.folder(
  path,
  size = NULL,
  voxelsize = NULL,
  thresh.offset = 0.1,
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = 2,
  full.voxel = FALSE,
  output = "markers",
  cores = 1
)
spots.combined.folder(
  path,
  size = NULL,
  voxelsize = NULL,
  thresh.offset = 0.1,
  min.sum.intensity = 2,
  max.distance = 0.5,
  use.brightest = FALSE,
  max.spots = 2,
  full.voxel = FALSE,
  output = "markers",
  cores = 1
)

Arguments

`path`	path to folder
`size`	size of img in microns, if size and voxelsize are NULL, size is determined from folder XYZmic
`voxelsize`	size of voxel in microns
`thresh.offset`	Thresh offset used in EBImage::thresh()
`min.sum.intensity`	spots smaller than min.sum.intensity are ignored
`max.distance`	use only spots with distance to other color spot smaller than max.distance
`use.brightest`	Logical; use only brightest in max.distance?
`max.spots`	maximum of spots (per channel), only when use brightest=TRUE
`full.voxel`	Logical; output contains full voxel instead of rgb intensities
`output`	output folder
`cores`	number of cores we can use of parallel computing (needs parallel package if cores>1)

Value

RGB image with spots will be written to output folder

Test for colors in classes distribution for folders

Description

Test for colors in classes distribution for folders

Usage

t_colors.in.classes.folder(path, test = "Wilcoxon")
t_colors.in.classes.folder(path, test = "Wilcoxon")

Arguments

`path`	path to folder
`test`	"Wilcoxon", "wilcox" or "U" for Wilcoxon rank-sum (Mann-Whitney U), chisq for Chi-squared test

Value

test results

`inputfolder`	Input folder
`outputfolder`	Output folder
`N`	Max class #'

`blue`	DAPI channel (image)
`mask`	mask (image)
`N`	number of classes
`beta`	smoothing parameter used in potts model (default: 0.1)
`z`	scaling parameter: size of voxel in X-/Y-direction divided by the size of voxel in Z-direction (slice scaling parameter: size of voxel in X-/Y-direction divided by the size of voxel in Z-direction (slice thickness))
`silent`	boolean. Should algorithm be silent?

`x`	matrix
`method`	method for intervals: "minmax" (default), "quantile" or "sd"
`qu`	vector of two quantiles for method="quantile
`ylim`	limits for y axis. Default:NULL is ylim=c(0,max(interval))
`horiz`	boolean: horizontal bars?
`border`	border parameter forwarded to barplot, default: NA is nor border
`...`	additional parameters forwarded to barplot

Package 'nucim'

Help Index

Barplot with Intervals

Description

Usage

Arguments

Value

Barplot with Intervals for two or three bars beside

Description

Usage

Arguments

Value

Class neighbourhood distribution

Description

Usage

Arguments

Value

class.neighbours.folder

Description

Usage

Arguments

Value

Classify DAPI

Description

Usage

Arguments

Value

Classify DAPI

Description

Usage

Arguments

Value

Classify DAPI

Description

Usage

Arguments

Value

Count classes in classified image

Description

Usage

Arguments

Value

Compute colors in classes distribution

Description

Usage

Arguments

Details

Value

Compute colors in classes distribution for folders

Description

Usage

Arguments

Value

Compute distance to border of classes

Description

Usage

Arguments

Value

Mask DAPI in kernel

Description

Usage

Arguments

Value

Automatic DAPI mask segmentation for files

Description

Usage

Arguments

Value

Automatic DAPI mask segmentation for folder

Description

Usage

Arguments

Value

Detects spots for one file

Description

Usage

Arguments

Value

Detects spots

Description