Package 'ribiosUtils'

Description

Given several objects, the function tests whether all of them are identical.

Usage

allIdentical(...)

Arguments

Value

Logical, whether all objects are the same

Author(s)

Jitao David Zhang <[email protected]>

See Also

identical

Examples

test1 <- test2 <- test3 <- LETTERS[1:3]
allIdentical(test1, test2, test3)
allIdentical(list(test1, test2, test3))

num1 <- num2 <- num3 <- num4 <- sqrt(3)
allIdentical(num1, num2, num3, num4)

Description

Apply isTopOrIncAndNotExcl filter to a matrix

Usage

applyTopOrIncAndNotExclFilter(matrix, MARGIN, top = 1, falseValue = 0, ...)

Arguments

Value

A matrix with the same dimnames but with elements not satisfying isTopOrIncAndNotExcl replaced by falseValue.

Examples

myMat <- matrix(c(1,2,3,4,8,7,6,5,12,9,11,10), nrow=3, byrow=TRUE,
   dimnames=list(c("A", "B", "C"), c("Alpha", "Beta", "Gamma", "Delta")))
print(myMat)
applyTopOrIncAndNotExclFilter(myMat, 1, top=2, falseValue=-1)
applyTopOrIncAndNotExclFilter(myMat, 2, top=2, falseValue=-1)
applyTopOrIncAndNotExclFilter(myMat, 2, top=2, falseValue=-1, decreasing=FALSE)
applyTopOrIncAndNotExclFilter(myMat, 1, top=2, falseValue=-1, incFunc=function(x) x%%2==0)
applyTopOrIncAndNotExclFilter(myMat, 1, top=2, falseValue=-1,
   incFunc=function(x) x%%2==0, excFunc=function(x) x<5)

Description

Convert string-valued data frame or matrix into a numeric matrix

Usage

asNumMatrix(x)

Arguments

Value

A numeric matrix with the same dimension

Author(s)

Jitao David Zhang <[email protected]>

Examples

testDf <- data.frame(a=c("2.34", "4.55"), b=c("7.33", "9.10"))
asNumMatrix(testDf)

testMatrix <- matrix(c("2.34", "4.55", "9E-3","-2.44", "7.33", "9.10"), nrow=2)
asNumMatrix(testMatrix)

Description

The function calls matchColumnName internally to match the column names.

Usage

assertColumnName(data.frame.cols, reqCols, ignore.case = FALSE)

Arguments

Value

If all required column names are present, their indices are returned *invisibly*. Otherwise an error message is printed.

Examples

myTestDf <- data.frame(HBV=1:3, VFB=0:2, BVB=4:6, FCB=2:4)
myFavTeams <- c("HBV", "BVB")
assertColumnName(myTestDf, myFavTeams)
myFavTeamsCase <- c("hbv", "bVb")
assertColumnName(myTestDf, myFavTeamsCase, ignore.case=TRUE)

Description

Check dimensionality of contrast matrix

Usage

assertContrast(design, contrast)

Arguments

Value

Side effect is used: the function stops if the ncol(design) does not equal nrow(contrast)

Examples

design <- matrix(1:20, ncol=5)
contrast <- matrix(c(-1,1,0,0,0, 0,1,0,-1,0), nrow=5)
assertContrast(design, contrast)

Description

Check dimensionality of design matrix

Usage

assertDesign(nsample, design)

Arguments

Value

Side effect is used: the function stops if sample size does not equal ncol(matrix)

Examples

nsample <- 4
design <- matrix(1:20, ncol=5)
assertDesign(nsample, design)

Description

Check dimensionality of both design and contrast matrix

Usage

assertDesignContrast(nsample, design, contrast)

Arguments

Value

Side effect is used: the function stops if there are errors in the dimensionalities

See Also

assertDesign, assertContrast

Examples

nsample <- 4
design <- matrix(1:20, ncol=5)
contrast <- matrix(c(-1,1,0,0,0, 0,1,0,-1,0), nrow=5)
assertDesignContrast(nsample, design, contrast)

Description

Print BEDA project information

Usage

bedaInfo()

Value

A list, including pstore path, URL, git address, and user id The function is used at the end of the Rmarkdown report to print relevant information to help other colleagues finding relevant resources

Examples

bedaInfo()

Description

Translate BiOmics-Pathology pstore path to URL

Usage

biomicsPstorePath2URL(path)

Arguments

Value

Character string of biomics pstore path The URL is only visible inside Roche

Examples

biomicsPstorePath2URL("/pstore/data/biomics/")

Description

The basic concepts of these functions are borrowed from the bound function in the Qt framework.

Usage

bound(x, low, high)

boundNorm(x, low = min(x, na.rm = TRUE), high = max(x, na.rm = TRUE))

Arguments

Details

bound sets the values smaller than low, or larger than high, to the value of low and high respectively.If no such values exist, the vector or matrix is returned unchanged.

boundNorm performs a 0-1 normalization. Input vector or matrix is transformed linearly onto the region defined between low and high, which has the unit length (1).

Value

A numeric vector or matrix, the same type as input.

Author(s)

Jitao David Zhang <[email protected]>

Examples

myVec <- c(2,4,3,-1,9,5,3,4)
bound(myVec, 0, 8)
boundNorm(myVec)
## boundNorm returns negative values if input values lie out of the
## given region between low and high
boundNorm(myVec, 0, 8)

myMat <- matrix(myVec, nrow=2)
myMat
bound(myMat, 0, 8)
boundNorm(myMat)
boundNorm(myMat, 0, 8)

Description

Column bind by rownames

Usage

cbindByRownames(..., type = c("intersect", "union"))

rbindByColnames(..., type = c("intersect", "union"))

Arguments

Value

A matrix

Examples

mat1 <- matrix(1:9, nrow=3, byrow=FALSE,
               dimnames=list(LETTERS[1:3], LETTERS[1:3]))
mat2 <- matrix(1:9, nrow=3, byrow=FALSE,
               dimnames=list(LETTERS[2:4], LETTERS[4:6]))
mat3 <- matrix(1:9, nrow=3, byrow=FALSE,
               dimnames=list(LETTERS[c(2,4,5)], LETTERS[7:9]))
cbindByRownames(mat1, mat2, mat3, type="intersect")
cbindByRownames(mat1, mat2, mat3, type="union")
## it is also possible to pass a list
cbindByRownames(list(mat1, mat2, mat3), type="union")
mat4 <- matrix(1:9, nrow=3, byrow=FALSE,
               dimnames=list(LETTERS[1:3], LETTERS[1:3]))
mat5 <- matrix(1:9, nrow=3, byrow=FALSE,
               dimnames=list(LETTERS[4:6], LETTERS[2:4]))
mat6 <- matrix(1:9, nrow=3, byrow=TRUE,
               dimnames=list(LETTERS[7:9], LETTERS[c(2,4,6)]))
rbindByColnames(mat4, mat5, mat6, type="intersect")
rbindByColnames(mat4, mat5, mat6,  type="union")
## it is also possible to pass a list
rbindByColnames(list(mat4, mat5, mat6), type="union")

Description

checkFile checks whether file exists, assertFile stops the program if files do not exist

Usage

checkFile(...)

assertFile(...)

Arguments

Details

assertFile is often used in scripts where missing a file would cause the script fail.

Value

checkFile returns logical vector. assertFile returns an invisible TRUE if files exist, otherwise halts and prints error messages.

Author(s)

Jitao David Zhang <[email protected]>

See Also

isDir and assertDir

Examples

myDesc <- system.file("DESCRIPTION", package="ribiosUtils")
myNEWS <- system.file("NEWS", package="ribiosUtils")
checkFile(myDesc, myNEWS)
assertFile(myDesc, myNEWS)

Description

Print the chosen few (the first and the last) items of a long vector

Usage

chosenFew(vec, start = 3, end = 1, collapse = ",")

Arguments

Value

A character string ready to be printed

Note

In case the vector is shorter than the sum of start and end, the whole vector is printed.

Author(s)

Jitao David Zhang <[email protected]>

Examples

lvec1 <- 1:100
chosenFew(lvec1)
chosenFew(lvec1, start=5, end=3)

svec <- 1:8
chosenFew(svec)
chosenFew(svec, start=5, end=4)

Close connections to all loggers This function closes all open connections set up by loggers It is automatically run at the end of the R session (setup by registerLog)

Description

Close connections to all loggers This function closes all open connections set up by loggers It is automatically run at the end of the R session (setup by registerLog)

Usage

closeLoggerConnections()

Value

Invisible NULL. Only side effect is used.

See Also

registerLog

Description

Pairwise jaccard/overlap coefficient can be calculated efficiently using matrix Pairwise overlap coefficient of binary matrix by column

Usage

columnOverlapCoefficient(x, y = NULL)

Arguments

Value

A matrix of column-wise pairwise overlap coefficients of the binary matrix. NaN is reported when neither of the columns have any non-zero element.

Examples

set.seed(1887)
testMatrix1 <- matrix(rbinom(120, 1, 0.2), nrow=15)
columnOverlapCoefficient(testMatrix1)

testMatrix2 <- matrix(rbinom(150, 1, 0.2), nrow=15)
testMatrix12Poe <- columnOverlapCoefficient(testMatrix1, 
  testMatrix2)

Description

Basic set operations are used to compare two vectors

Usage

compTwoVecs(vec1, vec2)

Arguments

Value

A vector of six integer elements

Author(s)

Jitao David Zhang <[email protected]>

Examples

year1 <- c("HSV", "FCB", "BVB", "S04", "FCN")
year2 <- c("HSV", "FCK", "S04")
compTwoVecs(year1, year2)

Description

Calculate correlation coefficients using common rows of the two matrices

Usage

corByRownames(mat1, mat2, ...)

Arguments

Value

A matrix of the dimension $m \times n$, where m and n are number of columns in mat1 and mat2, respectively. The matrix has an attribute, commonRownames, giving the common rownames shared by the two matrices.

Examples

myMat1 <- matrix(rnorm(24), nrow=6, byrow=TRUE,
  dimnames=list(sprintf("R%d", 1:6), sprintf("C%d", 1:4)))
myMat2 <- matrix(rnorm(35), nrow=7, byrow=TRUE,
  dimnames=list(sprintf("R%d", 7:1), sprintf("C%d", 1:5)))
corByRownames(myMat1, myMat2)

Description

Count tokens by splitting strings

Usage

countTokens(str, split = "\t", ...)

Arguments

Value

Integer vector: count of tokens in the strings

Author(s)

Jitao David Zhang <[email protected]>

See Also

strsplit to split strings, or a convenient wrapper strtoken in this package.

Examples

myStrings <- c("HSV\t1887\tFavorite", "FCB\t1900", "FCK\t1948")
countTokens(myStrings)

## the function deals with factors as well
countTokens(factor(myStrings))

Description

The function is particularly useful for scripting.

Usage

createDir(dir, showWarnings = FALSE, recursive = TRUE, mode = "0777")

Arguments

Value

Directory name (invisible)

Author(s)

Jitao David Zhang <[email protected]>

Examples

tempdir <- tempdir()
createDir(tempdir)

Description

This function copies the skeleton RMarkdown template shipped with ribiosUtils to the current working directory (or any specified path) under a user-given file name. When no file name is provided, a default name based on the current date is used.

Usage

createRmdTemplate(filename = NULL, overwrite = FALSE)

Arguments

Value

The path of the copied file (invisible).

Author(s)

Jitao David Zhang <[email protected]>

Examples

## Not run: 
## copy with default file name
createRmdTemplate()

## copy with a custom file name
createRmdTemplate("my_analysis.Rmd")

## End(Not run)

## copy to a temporary directory
dest <- createRmdTemplate(file.path(tempdir(), "test_report.Rmd"))
file.exists(dest)
unlink(dest)

Description

Cumulative Jaccard Index

Usage

cumJaccardIndex(list)

cumJaccardDistance(list)

Arguments

Value

The cumulative Jaccard Index, a vector of values between 0 and 1, of the same length as the input list

The cumulative Jaccard Index is calculated by calculating the Jaccard Index of element i and the union of elements between 1 and i-1. The cumulative Jaccard Index of the first element is set as 0.0.

The cumulative Jaccard distance is defined in almost the same way, with the only difference the distance is returned. The value of the first element is 1.0.

Note

An advantage of using cumulative overlap coefficient over cumulative Jaccard Index is that it is monotonic: the value is garanteed to decrease from 1 to 0, whereas the cumulative Jaccard Index may not be monotic.

See Also

cumOverlapCoefficient

Examples

myList <- list(first=LETTERS[1:5], second=LETTERS[6:10], third=LETTERS[8:12], fourth=LETTERS[1:12])
cumJaccardIndex(myList)
cumJaccardDistance(myList)

Description

Cumulative overlap coefficient

Usage

cumOverlapCoefficient(list)

cumOverlapDistance(list)

Arguments

Value

The cumulative overlap coefficients, a vector of values between 0 and 1, of the same length as the input list

The cumulative overlap coefficient is calculated by calculating the overlap coefficient of element i and the union of elements between 1 and i-1. The cumulative overlap coefficient of the first element is set as 0.0.

The cumulative overlap distance is defined in almost the same way, with the only difference the distance is returned. The value of the first element is 1.0. Pratically it is calculated by 1-cumOverlapCoefficient.

Since the denominator of the overlap coefficient is the size of the smaller set of the two, which is bound to be the size of element i, the cumulative overlap distance can be interpreted as the proportion of new items in each new element that are unseen in previous elements. Similarly, the cumulative overlap coefficient can be interpreted as the proportion of items in each new element that have been seen in previous elements. See examples below.

Note

An advantage of using cumulative overlap coefficient over cumulative Jaccard Index is that it is monotonic: the value is garanteed to decrease from 1 to 0, whereas the cumulative Jaccard Index may not be monotic.

Examples

myList <- list(first=LETTERS[1:5], second=LETTERS[6:10], third=LETTERS[8:12], fourth=LETTERS[1:12])
cumOverlapCoefficient(myList)
cumOverlapDistance(myList)

Description

Proportion of cumulative sum over sum

Usage

cumsumprop(x)

Arguments

Value

the proportion cumulative sum over sum

Examples

x <- 1:4
cumsumprop(x) ## 0.1, 0.3, 0.6, 1

Description

Three types of labels (levels) are supported: “cut.default” (Interval labels returned by cut as default), “left” (Left boundary of intervals), and “right” (Right boundary of intervals).

Usage

cutInterval(
  x,
  step = 1,
  labelOption = c("cut.default", "left", "right"),
  include.lowest = FALSE,
  right = TRUE,
  dig.lab = 3,
  ordered_result = FALSE,
  ...
)

Arguments

Value

A vector of factors

Author(s)

Jitao David Zhang <[email protected]>

See Also

cut

Examples

testNum <- rnorm(100)
(testFac <- cutInterval(testNum, step=1, labelOption="cut.default"))

## compare the result to
(testFacCut <- cut(testNum, 10))

Description

Cut a tree into groups of ordered sizes

Usage

cutreeIntoOrderedGroups(tree, k = NULL, h = NULL, decreasing = TRUE)

Arguments

Value

A named integer vector of cluster assignments, ordered by cluster size (largest first by default). If multiple values of k or h are provided, a matrix with one column per value.

See Also

cutree

Examples

hc <- hclust(dist(USArrests))
hck5 <- cutreeIntoOrderedGroups(hc, k = 5)
table(hck5)
## compare with cutree, which does not order the groups
table(cutree(hc, k=5))

hck25 <- cutreeIntoOrderedGroups(hc, k = 2:5)
apply(hck25, 2, table)

Description

The function try to assign a factor vector for a data.frame object. See details below.

Usage

dfFactor(df, sample.group)

Arguments

Details

The function tries to get a factor vector of the same length as the number of rows in the data.frame. The determination is done in the following order: Step 1: It tries to find a column in the data.frame with the name as given by sample.group. If found, this column is transformed into a factor if not and returned. Step 2: It tries to interpret the sample.group as an integer, as the index of the column in the data.frame giving the factor. Step 3: When sample.group itself is a vector of the same length as the data.frame, it is cast to factor when it is still not and returned.

Otherwise the program stops with error.

Value

A factor vector with the same length as the data.frame

Author(s)

Jitao David Zhang <[email protected]>

Examples

df <- data.frame(gender=c("M", "M", "F", "F", "M"),
age=c(12,12,14,12,14), score=c("A", "B-", "C", "B-", "A"))
dfFactor(df, "gender")
dfFactor(df, "score")
dfFactor(df, 1L)
dfFactor(df, 2L)
dfFactor(df, df$score)

Description

Convert factor columns in a data.frame into character strings

Usage

dfFactor2Str(df)

Arguments

Value

A data.frame with factor columns coereced into character strings

Examples

exampleDf <- data.frame(Teams=c("HSV", "FCB", "FCB", "HSV"), 
             Player=c("Mueller", "Mueller", "Robben", "Holtby"), 
             scores=c(3.5, 1.5, 1.5, 1.0), stringsAsFactors=TRUE)
strDf <- dfFactor2Str(exampleDf)
stopifnot(identical(strDf[,1], c("HSV", "FCB", "FCB", "HSV")))
stopifnot(identical(exampleDf[,1], factor(c("HSV", "FCB", "FCB", "HSV"))))

Description

Format labels for wells in microwell plates to equal widths

Usage

equateWellLabelWidth(wells)

Arguments

Value

A vector of the same length, with all labels adjusted to the equal width, with left-padding zeros added whenever it makes sense. If the input labels are already of the same length, no change is applied.

Examples

equateWellLabelWidth(c("A1", "C10", "D12"))

Description

Many files have base and extensions in their names, for instance for the file mybook.pdf, the base is mybook and the extension is pdf. basefilename extname functions extract these information from one or more file names.

Usage

extname(x, ifnotfound = NA, lower.case = FALSE)

Arguments

Value

The base file name or the extension as characters, of the same length as the input file name character. In case that a file name does not contain a extension, NA will be returned.

Note

In case there are multiple dots in the input file name, the last field will be taken as the extension, and the rest as the base name. For instance for file test.out.txt, returned base name is test.out and extension is txt.

Author(s)

Jitao David Zhang <[email protected]>

Examples

extname("mybook.pdf")
extname("sequence.in.fasta")
extname(c("/path/mybook.pdf", "test.doc"))
extname("README")
extname("README", ifnotfound="")
extname("/path/my\ home/Holiday Plan.txt")

basefilename("mybook.pdf")
basefilename("sequence.in.fasta")
basefilename(c("/path/mybook.pdf", "test.doc"))
basefilename("README")
basefilename("/path/my\ home/Holiday Plan.txt")

basefilename("myBook.pdf", lower.case=TRUE)
extname("myBook.PDF", lower.case=TRUE)

Description

Make the first alphabet of strings uppercase

Usage

firstUp(str)

Arguments

Value

A vector of the same length, with the first alphabet in uppercase

See Also

toTitleCase

Examples

firstUp('test string')
firstUp(strsplit('many many years ago', ' ')[[1]])

Description

Shorten strings to strings with a fix width of characters

Usage

fixWidthStr(str, nchar = 8, align = c("left", "right"))

Arguments

Value

A vector of strings with fixed widths

Note

NA will be converted to characters and the same fixed width will be applied. The behavior is different from shortenStr, where NA is kept as it is.

See Also

shortenStr

Examples

inputStrs <- c("abc", "abcd", "abcde", "abcdefg", "NA", NA) 
outputStrs <- fixWidthStr(inputStrs, nchar=4)
stopifnot(all(nchar(outputStrs)==4))

Description

If any of the expressions in ‘...’ are not all TRUE, stop is called, producing an error message defined by the msg parameter.

Usage

haltifnot(..., msg = "Error undefined. Please contact the developer")

Arguments

Details

The function is adapted from the stopifnot function, with the difference that the error message can be defined the programmer. With haltifnot error message can be more informative, which is desired for diagnostic and user-interation purposes.

Value

NULL if all statements in ... are TRUE

Author(s)

Jitao David Zhang <[email protected]>

See Also

stop, warning and stopifnot

Examples

haltifnot(1==1, all.equal(pi, 3.14159265), 1<2) ## all TRUE
m <- matrix(c(1,3,3,1), 2,2)
haltifnot(m == t(m), diag(m) == rep(1,2)) ## all TRUE

op <- options(error = expression(NULL))
# "disable stop(.)"  << Use with CARE! >>
     
haltifnot(all.equal(pi, 3.141593),  2 < 2, all(1:10 < 12), "a" < "b",
          msg="not all conditions are TRUE. Please contact the devleoper")
options(op)# revert to previous error handler

Description

These two functions reassembles head and tail, showing the first rows and columns of 2D data structures, e.g. matrix or data.frame.

Usage

headhead(x, m = 6L, n = 6L)

Arguments

Details

While head and tail can be applied to data.frame or matrix as well, they show all columns of the first (last) rows even if the matrix has a large number of columns. These two function, headhead and tailtail, circumvent this problem by showing only the first rows AND the first columns.

Value

The first rows/columns of the input object

Author(s)

Jitao David Zhang <[email protected]>

See Also

head, tail

Examples

myMat <- matrix(rnorm(10000), nrow=10L)
head(myMat)
headhead(myMat)
tailtail(myMat)

Description

This function prints head and tail elements of a vector for visualization purposes. See examples for its usage.

Usage

headtail(vec, head = 2, tail = 1, collapse = ", ")

Arguments

Details

Head and tail elements are concatenated with ellipsis, if there are any elements that are not shown in the vector.

Value

A character string representing the vector

Author(s)

Jitao David Zhang <[email protected]>

See Also

head, tail

Examples

testVec1 <- LETTERS[1:10]
headtail(testVec1)
headtail(testVec1, head=3, tail=3)
headtail(testVec1, head=3, tail=3, collapse="|")

testVec2 <- letters[1:3]
headtail(testVec2, head=1, tail=1)
headtail(testVec2, head=2, tail=1)

Description

Test whether two matrices are identical by values and by dim names

Usage

identicalMatrix(x, y, epsilon = 1e-12)

Arguments

Value

Logical

Examples

set.seed(1887); x <- matrix(rnorm(1000), nrow=10, dimnames=list(LETTERS[1:10],NULL))
set.seed(1887); y <- matrix(rnorm(1000), nrow=10, dimnames=list(LETTERS[1:10],NULL))
set.seed(1887); z <- matrix(rnorm(1000), nrow=10, dimnames=list(letters[1:10],NULL))
stopifnot(identicalMatrix(x,y))
stopifnot(!identicalMatrix(x,z))

Description

Test whether two matrices have the same numerica values given certain accuracy

Usage

identicalMatrixValue(x, y, epsilon = 1e-12)

Arguments

Value

Logical

Examples

set.seed(1887); x <- matrix(rnorm(1000), nrow=10)
set.seed(1887); y <- matrix(rnorm(1000), nrow=10)
set.seed(1882); z <- matrix(rnorm(1000), nrow=10)
stopifnot(identicalMatrixValue(x,y))
stopifnot(!identicalMatrixValue(x,y+1E-5))
stopifnot(!identicalMatrixValue(x,y-1E-5))
stopifnot(!identicalMatrixValue(x,z))

Description

Case-insensitive match and pmatch functions, especially useful in parsing user inputs, e.g. from command line.

Usage

imatch(x, table, ...)

Arguments

Details

imatch and ipmatch works similar as match and pmatch, except that they are case-insensitive.

matchv, imatchv and ipmatchv are shot-cuts to get the matched value (therefore the ‘v’) if the match succeeded, or NA if not. match(x, table) is equivalent to table[match(x, table)]. See examples.

Value

imatch and ipmatch returns matching indices, or NA (by default) if the match failed.

matchv, imatchv and ipmatchv returns the matching element in table, or NA if the match failed. Note that when cases are different in x and table, the one in table will be returned. This is especially useful for cases where user's input has different cases as the internal options.

Author(s)

Jitao David Zhang <[email protected]>

See Also

match and pmatch

Examples

user.input <- c("hsv", "BvB")
user.input2 <- c("HS", "BV")
internal.options <- c("HSV", "FCB", "BVB", "FCN")

match(user.input, internal.options)
imatch(user.input, internal.options)
ipmatch(user.input, internal.options)
ipmatch(user.input2, internal.options)

matchv(user.input, internal.options)
matchv(tolower(user.input), tolower(internal.options))
imatchv(user.input, internal.options)
ipmatchv(user.input, internal.options)
ipmatchv(user.input2, internal.options)

Description

Invert the names and elements of a list

Usage

invertList(inputList, simplify = FALSE)

Arguments

Value

A list with values from the input becoming names and vice versa. When simplify=TRUE and there are no duplicated names, a named character vector is returned instead.

Examples

myList <- list("A"=c("a", "alpha"), "B"=c("b", "Beta"), "C"="c")
invertList(myList)
invertList(myList, simplify=TRUE)

Description

Checks whether given character strings point to valid directories

Usage

isDir(...)

checkDir(...)

assertDir(...)

Arguments

Details

isDir tests whether the given string represent a valid, existing directory. assertDir performs a logical test, and stops the program if the given string does not point to a given directory.

checkDir is synonymous to isDir

Value

isDir returns logical vector.

assertDir returns an invisible TRUE if directories exist, otherwise halts and prints error messages.

Author(s)

Jitao David Zhang <[email protected]>

See Also

file.info, checkFile and assertFile

Examples

dir1 <- tempdir()
dir2 <- tempdir()

isDir(dir1, dir2)
assertDir(dir1, dir2)

Description

Determines whether an object is of class try-error

Usage

isError(x)

Arguments

Value

Logical value, TRUE if x inherits the try-error class.

Author(s)

Jitao David Zhang <[email protected]>

Examples

if(exists("nonExistObj")) rm(nonExistsObj)
myObj <- try(nonExistObj/5, silent=TRUE)
isError(myObj)

Description

Whether an integer is odd (or even)

Usage

isOdd(x)

isEven(x)

Arguments

Value

Logical, whether the input number is odd or even.

isOdd and isEven returns whether an integer is odd or even, respectively.

Examples

isOdd(3)
isEven(4)

Description

Tell whether a character string is a Roche compound ID

Usage

isRocheCompoundID(str)

Arguments

Value

A logical vector of the same length as str, indicating whether each element is a Roche compound ID or not.

Note

Short versions (RO[1-9]{2,7}) are supported.

Examples

isRocheCompoundID(c("RO1234567", "RO-1234567", 
                   "RO1234567-000", "RO1234567-000-000",
                   "ROnoise-000-000"))

Description

Logical vector of being top or included and not excluded

Usage

isTopOrIncAndNotExcl(x, top = 1, incFunc, excFunc, decreasing = TRUE)

Arguments

Value

A logical vector of the same length as the input x, indicating whether each element is being either top or included, and not excluded. The function can be used to keep top elements of a vector while considering both inclusion and exclusion criteria.

Examples

myVal <- c(2, 4, 8, 7, 1)
isTopOrIncAndNotExcl(myVal, top=1)
isTopOrIncAndNotExcl(myVal, top=3)
isTopOrIncAndNotExcl(myVal, top=3, incFunc=function(x) x>=2)
isTopOrIncAndNotExcl(myVal, top=3, excFunc=function(x) x%%2==1)
isTopOrIncAndNotExcl(myVal, top=3, incFunc=function(x) x>=2, excFunc=function(x) x%%2==1)
myVal2 <- c("a", "A", "a", "A", "A")
isTopOrIncAndNotExcl(myVal2, 2)
isTopOrIncAndNotExcl(myVal2, 2, incFunc=function(x) x=="A")
isTopOrIncAndNotExcl(myVal2, 4)
isTopOrIncAndNotExcl(myVal2, 4, excFunc=function(x) x=="a")

## the function returns all TRUEs if top is larger than the length of the vector
isTopOrIncAndNotExcl(myVal, top=9)

Description

Calculate the Jaccard Index between two vectors

Usage

jaccardIndex(x, y)

jaccardDistance(x, y)

Arguments

Value

The Jaccard Index, a number between 0 and 1

JaccardDistance is defined as 1-JaccardIndex.

Examples

myX <- 1:6
myY <- 4:9
jaccardIndex(myX, myY)
jaccardDistance(myX, myY)

myX <- LETTERS[1:5]
myY <- LETTERS[6:10]
jaccardIndex(myX, myY)
jaccardDistance(myX, myY)

Description

A common task in expression analysis is to collapse multiple features that are mapped to the same gene by some statistic. This function does this job by keeping the matrix row (normally features) with the higheest statistic specified by the user.

Usage

keepMaxStatRow(
  matrix,
  keys,
  keepNArows = TRUE,
  stat = function(x) mean(x, na.rm = TRUE),
  levels = c("rownames", "attribute", "discard"),
  ...
)

Arguments

Details

isMaxStatRow returns a logical vector, with rows with maximal statistics each key as TRUE and otherwise as FALSE. keepMaxStatRowInd returns the integer indices of such rows. Finally keepMaxStatRow returns the resulting matrices.

For use see examples

Value

A numeric matrix with rows mapped to unique keys, selected by the maximum statistics. See examples below

Author(s)

Jitao David Zhang <[email protected]>

Examples

myFun1 <- function(x) mean(x, na.rm=TRUE)
myFun2 <- function(x) sd(x, na.rm=TRUE)
mat1 <- matrix(c(1,3,4,-5,
                 0,1,2,3,
                 7,9,5,3,
                 0,1,4,3), ncol=4, byrow=TRUE)
keys1 <- c("A", "B", "A", "B")

isMaxStatRow(mat1, keys1, stat=myFun1)
isMaxStatRow(mat1, keys1, stat=myFun2)

keepMaxStatRowInd(mat1, keys1, stat=myFun1)
keepMaxStatRowInd(mat1, keys1, stat=myFun2)

keepMaxStatRow(mat1, keys1, stat=myFun1)
keepMaxStatRow(mat1, keys1, stat="myFun2")
keepMaxStatRow(mat1, keys1, stat="myFun2", levels="discard")
keepMaxStatRow(mat1, keys1, stat="myFun2", levels="attribute")

mat2 <- matrix(c(1,3,4,5,
                 0,1,2,3,
                 7,9,5,3,
                 0,1,4,3,
                 4,0,-1,3.1,
                 9,4,-3,2,
                 8,9,1,2,
                 0.1,0.2,0.5,NA,
                 NA, 4, 3,NA), ncol=4, byrow=TRUE,
               dimnames=list(LETTERS[1:9], NULL))
keys2 <- c("A", "B", "A", "B", NA, NA, "C", "A", "D")

isMaxStatRow(mat2, keys2, keepNArows=FALSE, stat=myFun1)
keepMaxStatRowInd(mat2, keys2, keepNArows=FALSE, stat=myFun1)

keepMaxStatRow(mat2, keys2, keepNArows=FALSE, stat=myFun1)
keepMaxStatRow(mat2, keys2, keepNArows=TRUE, stat=myFun1)
keepMaxStatRow(mat2, keys2, keepNArows=TRUE, stat=myFun1, levels="discard")
keepMaxStatRow(mat2, keys2, keepNArows=TRUE, stat=myFun1, levels="attribute")

Description

Return last characters from strings

Usage

lastChar(str)

Arguments

Value

A vector of the same length, containing last characters

Examples

lastChar("Go tell it on the mountain")
lastChar(c("HSV", "FCB", "BVB"))

Description

The specified library is loaded mutedly by suppressing all messages. If the library is not found, or its version under the specification of minVer, the R session dies with a message.

Usage

libordie(package, minVer, missing.quit.status = 1, ver.quit.status = 1)

Arguments

Details

Only one package should be tested once.

Value

NULL if success, otherwise the session will be killed.

Author(s)

Jitao David Zhang <[email protected]>

See Also

The function calls qqmsg internally to kill the session

Examples

libordie(stats)
libordie("methods")
libordie(base, minVer="2.15-1")

Description

Transform a list of character strings into a data.frame

Usage

list2df(list, names = NULL, col.names = c("Name", "Item"))

Arguments

Value

A data.frame

Examples

myList <- list(HSV=c("Mueller", "Papadopoulos", "Wood"), FCB=c("Lewandowski", "Robben", "Hummels"),
               BVB=c("Reus", "Goetze", "Kagawa"))
list2df(myList, col.names=c("Club", "Player"))

Description

Pairwise overlap coefficient of lists

Usage

listOverlapCoefficient(x, y = NULL, checkUniqueNonNA = TRUE)

Arguments

Value

A matrix of column-wise pairwise overlap coefficients.

Examples

set.seed(1887)
testSets1 <- sapply(rbinom(10, size=26, prob=0.3), 
  function(x) sample(LETTERS, x, replace=FALSE))
names(testSets1) <- sprintf("List%d", seq(along=testSets1))
testSets1Poe <- listOverlapCoefficient(testSets1)
testSets1PoeNoCheck <- listOverlapCoefficient(testSets1, checkUniqueNonNA=FALSE)
stopifnot(identical(testSets1Poe, testSets1PoeNoCheck))

testSets2 <- sapply(rbinom(15, size=26, prob=0.3),
  function(x) sample(LETTERS, x, replace=FALSE))
names(testSets2) <- sprintf("AnotherList%d", seq(along=testSets2))
testSets12Poe <- listOverlapCoefficient(testSets1, testSets2)

Description

Input data.frame must contain at least three columns: one contains row names (specified by row.col), one contains column names (column.col), and one contains values in matrix cells (value.col). The output is a 2D matrix.

Usage

longdf2matrix(
  df,
  row.col = 1L,
  column.col = 2L,
  value.col = 3L,
  missingValue = NULL
)

Arguments

Value

A 2D matrix equivalent to the long-format data frame

Author(s)

Jitao David Zhang <[email protected]>

See Also

matrix2longdf

Examples

test.df <- data.frame(H=c("HSV", "BVB", "HSV", "BVB"),
A=c("FCB", "S04", "S04", "FCB"),
score=c(3, 1, 1, 0))
longdf2matrix(test.df, row.col=1L, column.col=2L, value.col=3L)

data(Indometh)
longdf2matrix(Indometh, row.col="time", column.col="Subject",value.col="conc")
longdf2matrix(Indometh, row.col="Subject", column.col="time", value.col="conc")

Description

Given a vector known as master vcector, a data.frame and one column of the data.frame, the function matchColumnIndex matches the values in the column to the master vector, and returns the indices of each value in the column with respect to the vector. The function matchColumn returns whole or subset of the data.frame, with the matching column in the exact order of the vector.

Usage

matchColumn(vector, data.frame, column, multi = FALSE)

Arguments

Details

See more details below.

The function is used to address the following question: how can one order a data.frame by values of one of its columns, the order for which is given in a vector (known as “master vector”). matchColumnIndex and matchColumn provide thoroughly-tested implementation to address this question.

For one-to-one cases, where both the column and the vector have no duplicates and can be matched one-to-one, the question is straightforward to solve with the match function in R. In one-to-many or many-to-many matching cases, the parameter multi determines whether multiple rows matching the same value should be shown. If mutli=FALSE, then the sorted data.frame that are returned has exactly the same row number as the input vector; otherwise, the returned data.frame has more rows. See the examples below.

In either case, in the returned data.frame object by matchColumn, values in the column used for matching are overwritten by the master vector.If multi=TRUE, the order of values in the column is also obeying the order of the master vector, with exceptions of repeating values casued by mutliple matching.

The column parameter can be either character string or non-negative integers. In the exceptional case, where column=0L (“L” indicates integer), the row names of the data.frame is used for matching instead of any of the columns.

Both functions are NA-friendly, since NAs in neither vector nor column should break the code.

Value

For matchColumnIndex, if multi is set to FALSE, an integer vector of the same length as the master vector, indicating the order of the data.frame rows by which the column can be re-organized into the master vector. When multi is TRUE, the returning object is a list of the same length as the master vector, each item containing the index (indices) of data.frame rows which match to the master vector.

For matchColumn, a data.frame is always returned. In case multi=FALSE, the returning data frame has the same number of rows as the length of the input master vector, and the column which was specified to match contains the master vector in its order. If multi=TRUE, returned data frame can contain equal or more numbers of rows than the master vector, and multiple-matched items are repeated.

Note

When multi=TRUE, the indices within each list element (for matchColumnIndex) are returned in ascending order.

Author(s)

Jitao David Zhang <[email protected]>

See Also

See match for basic matching operations.

Examples

df <- data.frame(Team=c("HSV", "BVB", "HSC", "FCB", "HSV"),
                 Pkt=c(25,23,12,18,21),
                 row.names=c("C", "B", "A", "F", "E"))
teams <- c("HSV", "BVB", "BRE", NA)
ind <- c("C", "A", "G", "F", "C", "B", "B", NA)

matchColumnIndex(teams, df, 1L, multi=FALSE)
matchColumnIndex(teams, df, 1L, multi=TRUE)
matchColumnIndex(teams, df, "Team", multi=FALSE)
matchColumnIndex(teams, df, "Team", multi=TRUE)
matchColumnIndex(teams, df, 0, multi=FALSE)
matchColumnIndex(ind, df, 0, multi=FALSE)
matchColumnIndex(ind, df, 0, multi=TRUE)

matchColumn(teams, df, 1L, multi=FALSE)
matchColumn(teams, df, 1L, multi=TRUE)
matchColumn(teams, df, "Team", multi=FALSE)
matchColumn(teams, df, "Team", multi=TRUE)
matchColumn(ind, df, 0, multi=FALSE)
matchColumn(ind, df, 0, multi=TRUE)

Description

Match a given vector to column names of a data.frame or matrix

Usage

matchColumnName(data.frame.cols, reqCols, ignore.case = FALSE)

Arguments

Value

A vector of integers as indices

Examples

myTestDf <- data.frame(HBV=1:3, VFB=0:2, BVB=4:6, FCB=2:4)
myFavTeams <- c("HBV", "BVB")
matchColumnName(myTestDf, myFavTeams)
myFavTeamsCase <- c("hbv", "bVb")
matchColumnName(myTestDf, myFavTeamsCase, ignore.case=TRUE)
## NA will be returned in this case if ignore.case is set to FALSE
matchColumnName(myTestDf, myFavTeamsCase, ignore.case=FALSE)

Description

The function converts a matrix into a long-format, three-column data.frame, containing row, columna nd value. Such ‘long’ data.frames can be useful in data visualization and modelling.

Usage

matrix2longdf(
  mat,
  row.names,
  col.names,
  longdf.colnames = c("row", "column", "value")
)

Arguments

Details

The function converts a matrix into a three-column, ‘long’ format data.frame containing row names, column names, and values of the matrix.

Value

A data.frame object with three columns: row, column and value. If the input matrix is of dimesion MxN, the returning data.frame is of the dimension MNx3.

Note

The length of row.names and col.names should be as the same as the matrix dimension. Otherwise the function raises warnings.

Author(s)

Jitao David Zhang <[email protected]>

Examples

test.mat <- matrix(1:12, ncol=4, nrow=3, dimnames=list(LETTERS[1:3],
LETTERS[1:4]))
print(test.mat)
print(matrix2longdf(test.mat))
print(matrix2longdf(test.mat, longdf.colnames=c("From", "To", "Time")))

Description

Merge infrequent levels by setting the threshold of the proportion of cumulative sum over sum a.k.a. cumsumprop

Usage

mergeInfreqLevelsByCumsumprop(
  classes,
  thr = 0.9,
  mergedLevel = "others",
  returnFactor = TRUE
)

Arguments

Value

A character string vector or a factor, of the same length as the input classes, but with potentially fewer levels.

Note

In case only one class is deemed as infrequent, its label is unchanged.

Examples

set.seed(1887)
myVals <- sample(c(rep("A", 4), rep("B", 3), rep("C", 2), "D"))
## in the example below, since A, B, C make up of 90% of the total,
## D is infrequent. Since it is alone, it is not merged
mergeInfreqLevelsByCumsumprop(myVals, 0.9) 
mergeInfreqLevelsByCumsumprop(myVals, 0.9, returnFactor=FALSE) ## return characters
## in the example below, since A and B make up 70% of the total, 
## and A, B, C 90%, they are all frequent and D is infrequent. 
## Following the logic above, no merging happens
mergeInfreqLevelsByCumsumprop(myVals, 0.8)
mergeInfreqLevelsByCumsumprop(myVals, 0.7) ## A and B are left, C and D are merged
mergeInfreqLevelsByCumsumprop(myVals, 0.5) ## A and B are left, C and D are merged
mergeInfreqLevelsByCumsumprop(myVals, 0.4) ## A is left
mergeInfreqLevelsByCumsumprop(myVals, 0.3) ## A is left

Description

Testing whether multiple objects are identical

Usage

midentical(
  ...,
  num.eq = TRUE,
  single.NA = TRUE,
  attrib.as.set = TRUE,
  ignore.bytecode = TRUE,
  ignore.environment = FALSE,
  ignore.srcref = TRUE,
  extptr.as.ref = FALSE
)

Arguments

Details

midentical extends identical to test multiple objects instead of only two.

Value

A logical value, TRUE if all objects are identical

Author(s)

Jitao David Zhang <[email protected]>

See Also

identical

Examples

set1 <- "HSV"
set2 <- set3 <- set4 <- c("HSV", "FCB")

midentical(set1, set2)
midentical(list(set1, set2))

midentical(set2, set3, set4)
midentical(list(set2, set3, set4))

## other options passed to identical
midentical(0, -0, +0, num.eq=FALSE)
midentical(0, -0, +0, num.eq=TRUE)

Description

Multiple matching between two vectors. Different from R-native match function, where only one match is returned even if there are multiple matches, mmatch returns all of them.

Usage

mmatch(x, table, nomatch = NA_integer_)

Arguments

Details

Multiple matches can be useful in many cases, and there is no native R function for this purpose. User can write their own functions combining lapplying with match or %in%, our experience however shows that such non-vectorized function can be extremely slow, especially when the x or table vector gets longer.

mmatch delegates the multiple-matching task to a C-level function, which is optimized for speed. Internal benchmarking shows improvement of hundred fold, namely using mmatching costs about 1/100 of the time used by R-implementation.

Value

A list of the same length as the input x vector. Each list item contains the matching indices in ascending order (similar to match).

Author(s)

Jitao David Zhang <[email protected]>, C-code was adapted from the program written by Roland Schmucki.

See Also

match

Examples

vec1 <- c("HSV", "BVB", "FCB", "HSV", "BRE", "HSV", NA, "BVB")
vec2 <- c("FCB", "FCN", "FCB", "HSV", "BVB", "HSV", "FCK", NA, "BRE", "BRE")

mmatch(vec1, vec2)

## compare to match
match(vec1, vec2)

Description

Set operation functions in the base package, union, intersect and setdiff, can only be applied to binary manipulations involving two sets. Following functions, munion, mintersect and msetdiff, extend their basic versions to deal with multiple sets.

Usage

munion(...)

Arguments

Details

These functions apply set manipulations (union, intersect, or difference) in a sequential manner: the first two sets are considered first, then the third, the fourth and so on, till all sets have been visited.

Value

A vector of set operation results. Can be an empty vector if no results were returned.

Author(s)

Jitao David Zhang <[email protected]>

See Also

union, intersect and setdiff.

Examples

set1 <- c("HSV", "FCB", "BVB", "FCN", "HAN")
set2 <- c("HSV", "FCB", "BVB", "HAN")
set3 <- c("HSV", "BVB", "FSV")

munion(set1, set2, set3)
mintersect(set1, set2, set3)
msetdiff(set1, set2, set3)

## sets can be given in a list as well
munion(list(set1, set2, set3))
mintersect(list(set1, set2, set3))
msetdiff(list(set1, set2, set3))

Description

Replace NA in a vector with FALSE

Usage

na.false(x)

Arguments

Value

Logical vector or matrix with NAs replaced by FALSE

Author(s)

Jitao David Zhang <[email protected]>

See Also

myX <- c("HSV", "FCK", "FCN", NA, "BVB") res <- myX == "HSV" na.false(res)

Description

Calculate pairwise distances between each pair of items in a list

Usage

naivePairwiseDist(list, fun = jaccardIndex)

Arguments

Value

A symmetric matrix of dimension mxm, where m is the length of the list

This function is inefficient compared with matrix-based methods. It is exported just for education and for verifying results of matrix-based methods.

Examples

myList <- list(first=LETTERS[3:5], second=LETTERS[1:3], third=LETTERS[1:5], fourth=LETTERS[6:10])
naivePairwiseDist(myList, fun=jaccardIndex)
## despite of the name, any function that returns a number can work
naivePairwiseDist(myList, fun=jaccardDistance)

Description

Build a factor using the order of input character strings

Usage

ofactor(x, ...)

Arguments

Value

Factor with levels in the same order of the input strings.

Author(s)

Jitao David Zhang <[email protected]>

See Also

factor

Examples

testStrings <- c("A", "C", "B", "B", "C")
(testFac <- factor(testStrings))
(testOfac <- ofactor(testStrings))

stopifnot(identical(levels(testOfac), c("A", "C", "B")))

Description

Reorder the groups by their group size

Usage

orderCutgroup(groups, decreasing = TRUE)

Arguments

Description

Overlap coefficient, also known as Szymkiewicz-Simpson coefficient

Usage

overlapCoefficient(x, y, checkUniqueNonNA = FALSE)

overlapDistance(x, y, checkUniqueNonNA = FALSE)

Arguments

Value

The overlap coefficient

See Also

jaccardIndex

overlapCofficient calculates the overlap coefficient, and overlapDistance is defined by 1-overlapCoefficient.

Examples

myX <- 1:6
myY <- 4:9
overlapCoefficient(myX, myY)

myY2 <- 4:10
overlapCoefficient(myX, myY2)
## compare the result with Jaccard Index
jaccardIndex(myX, myY2)

## overlapDistance
overlapDistance(myX, myY2)

Description

Overwrite a directory

Usage

overwriteDir(dir, action = c("ask", "overwrite", "append", "no"))

Arguments

Value

If action is set to overwrite, the directory will be deleted recursively if it exists, a new directory with the same name will be created, and the function returns TRUE. If append is set, the function creates the directory if necessary and returns TRUE. If no is set, the function does nothing and returns.

If action is set to ask, user will be prompted for actions.

If overwrite is set, the directory will be removed and written anew.

If append is set, in contrast to overwrite, the directory and the files in it are not removed if they exists. In this case, files with the same name will be overwritten. Otherwise, new directories or files will be simply created. On the other hand, if the directory does not exist, it will be created.

If no is set, no action will be taken. The funciton returns FALSE.

Examples

## Helper to create a test directory with files
createTestDir <- function() {
  testdir <- file.path(tempdir(), "overwriteDir_test")
  dir.create(testdir, showWarnings = FALSE)
  writeLines("First file", file.path(testdir, "file1.txt"))
  writeLines("Second file", file.path(testdir, "file2.txt"))
  return(testdir)
}
addFileToDir <- function(testdir) {
  writeLines("New file", tempfile(tmpdir=testdir))
}

## overwrite: delete the directory and create it anew
testdir <- createTestDir()
length(dir(testdir)) ## two files should be there
overwriteDir(testdir, action="overwrite")
addFileToDir(testdir)
length(dir(testdir)) ## now there should be only one file

## append: keep existing files, add new ones
overwriteDir(testdir, action="append")
addFileToDir(testdir)
length(dir(testdir)) ## now two files

## no: no action, returns FALSE
noRes <- overwriteDir(testdir, action="no")
noRes

## cleanup
unlink(testdir, recursive = TRUE)

## Not run: 
  ## ask: prompts user for action (interactive only)
  testdir <- createTestDir()
  overwriteDir(testdir, action="ask")

## End(Not run)

Description

The function maps p values between 0 and 1 to continuous scores ranging on R by the following equation: $abs(log10(p))*sign$

Usage

pAbsLog10Score(p, sign = 1, replaceZero = TRUE)

Arguments

Value

A numeric vector of transformed p-values using signed -log10 transformation.

See Also

pQnormScore, pScore, replaceZeroPvalue

Examples

testPvals <- c(0.001, 0.01, 0.05, 0.1, 0.5, 1)
pAbsLog10Score(testPvals)
testPvalSign <- rep(c(-1,1), 3)
pAbsLog10Score(testPvals, sign=testPvalSign)
testLog <- rep(c(TRUE, FALSE),3)
pAbsLog10Score(testPvals, testLog)

Description

Calculate pairwise Jaccard Indices between each pair of items in a list

Usage

pairwiseJaccardIndex(list)

pairwiseJaccardDistance(list)

Arguments

Value

A symmetric matrix of dimension mxm, where m is the length of the list

pairwiseJaccardDistance is defined as 1-pairwiseJaccardIndex.

Examples

myList <- list(first=LETTERS[3:5], second=LETTERS[1:3], third=LETTERS[1:5], fourth=LETTERS[6:10])
pairwiseJaccardIndex(myList)

poormanPJI <- function(list) {
  sapply(list, function(x) sapply(list, function(y) jaccardIndex(x,y)))
}
stopifnot(identical(pairwiseJaccardIndex(myList), poormanPJI(myList)))

Description

Calculate pairwise overlap coefficients between each pair of items in a list

Usage

pairwiseOverlapDistance(list)

pairwiseOverlapCoefficient(list)

Arguments

Value

A symmetric matrix of dimension mxm, where m is the length of the list

pairwiseOverlapDistance is defined the pairwise overlap distance.

Examples

myList <- list(first=LETTERS[3:5], second=LETTERS[1:3], third=LETTERS[1:5], fourth=LETTERS[6:10])
pairwiseOverlapCoefficient(myList)
pairwiseOverlapDistance(myList)

poormanPOC <- function(list) {
  sapply(list, function(x) sapply(list, function(y) overlapCoefficient(x,y)))
}
stopifnot(identical(pairwiseOverlapCoefficient(myList), poormanPOC(myList)))

Description

Print a decimal number in procent format

Usage

percentage(x, fmt = "1.1")

Arguments

Value

Character string

Examples

percentage(c(0,0.1,0.25,1))
percentage(c(0,0.1,0.25,1), fmt="1.4")
percentage(c(0,-0.1,0.25,-1), fmt="+1.1")

Description

Quantile function, also known as the inverse of cumulative distribution function of the normal distribution, is used to map p-values to continuous scores raging on $R$. The signs of the resulting scores are positive by default and are determined by the parameter sign.

Usage

pQnormScore(p, sign = 1, replaceZero = TRUE)

Arguments

Value

A numeric vector of transformed p-values using signed quantile normal transformation.

See Also

pAbsLog10Score, pScore, double

Examples

testPvals <- c(0.001, 0.01, 0.05, 0.1, 0.5, 1)
pQnormScore(testPvals)
testPvalSign <- rep(c(-1,1), 3)
pQnormScore(testPvals, sign=testPvalSign)
testLog <- rep(c(TRUE, FALSE),3)
pQnormScore(testPvals, testLog)

Description

Print BEDAinfo object

Usage

## S3 method for class 'BEDAinfo'
print(x, ...)

Arguments

Value

Invisible NULL, only side effect is used

Examples

print(bedaInfo())

Description

The function wraps other functions to map p values ranging on $(0,1]$ to continuous scores ranging on $R$ in a number of ways.

Usage

pScore(p, sign = 1, method = c("qnorm", "absLog10"), replaceZero = TRUE)

Arguments

Value

A numeric vector of transformed p-values using the specified method.

See Also

pAbsLog10Score, pQnormScore

Examples

testPvals <- c(0.001, 0.01, 0.05, 0.1, 0.5, 1)
pScore(testPvals, method="absLog10")
pScore(testPvals, method="qnorm")
testPvalSign <- rep(c(-1,1), 3)
pScore(testPvals, sign=testPvalSign, method="absLog10")
pScore(testPvals, sign=testPvalSign, method="qnorm")
testLog <- rep(c(TRUE, FALSE),3)
pScore(testPvals, testLog, method="absLog10")
pScore(testPvals, testLog, method="qnorm")

testPvals <- 10^seq(-5, 0, 0.05)
plot(pScore(testPvals, method="qnorm"),
     pScore(testPvals, method="absLog10"),
     xlab="pQnormScore", ylab="pAbsLog10Score"); abline(0,1, col="red", lty=2)

Description

This function is helpful to export tables where certain columns are desired to be placed to the most left of the data.frame

Usage

putColsFirst(data.frame, columns)

Arguments

Value

data.frame with re-arranged columns

Author(s)

Jitao David Zhang <[email protected]>

Examples

clubs <- data.frame(Points=c(21,23,28,24), Name=c("BVB", "FCB", "HSV",
"FCK"), games=c(12,11,11,12))
putColsFirst(clubs, c("Name"))
putColsFirst(clubs, c("Name", "games"))

Description

Decode password encypted with pwencode.

Usage

pwdecode(password)

Arguments

Details

See pwdecode function documentation in BIOS for implemetnation details.

Note that since R does not support strings embedding null values (\000), the password to be decoded has to be given with two slashes, e.g. ‘ \001\000\129\235’.

Value

Decoded character string, or empty string if decoding fails

Author(s)

Jitao David Zhang <[email protected]>. The C library code was written by Detlef Wolf.

Examples

mycode <- " \\001\\000\\141\\314\\033\\033\\033\\033\\033\\142\\303\\056\\166\\311\\037\\042"
pwdecode(mycode)

Description

Encode a password

Usage

pwencode(label = "VAR", key)

Arguments

Value

Character string, encoded password

Description

Quitely quit R with messages in non-interactive sessions

Usage

qqmsg(..., status = 0, save = FALSE, runLast = TRUE)

Arguments

Details

The function prints messages in any case, and quits R if the current session is non-interactive, e.g. in the command-line running Rscript mode

Value

Invisible NULL, only side effect is used.

Author(s)

Jitao David Zhang <[email protected]>

See Also

quit

Examples

## the example should not run because it will lead the R session to quit
## Not run: 
  qqmsg()
  qqmsg("die", status=0)
  qqmsg("Avada kedavra", status=-1)
  qqmsg("Crucio!", "\n", "Avada kedavra", status=-100)

## End(Not run)

Description

Quietly runs a system command: the output is internalized and returned as an invisible variable, and the standard error output is ignored.

Usage

qsystem(command)

Arguments

Details

The function runs the system command in a quiet mode. The function can be useful in CGI scripts, for instance

Value

(Invisibly) the internalized output of the command

Author(s)

Jitao David Zhang <[email protected]>

Examples

dateIntern <- system("date")

Description

Factor variables with numbers as levels are alphabetically ordered by default, which requires rearrangements for various purposes, e.g. modelling or visualizations. This function re-orders levels of numeric factor variables numerically.

Usage

refactorNum(x, decreasing = FALSE)

Arguments

Value

A factor variable, with sorted numeric values as levels

Author(s)

Jitao David Zhang <[email protected]>

Examples

(nums <- factor(c("2","4","24","1","2","125","1","2","125")))
(nums.new <- refactorNum(nums))

Description

Users can register arbitrary numbers of loggers with registerLog, and the functions take care of low-level details such as openning and closing the connections.

Usage

registerLog(..., append = FALSE)

Arguments

Details

Input parameters can be either character strings or connections (such as the objects returned by stdout() or pipe().

If a character string is registered as a logger, it is assumed as a file name (user must make sure that it is writable/appendable). In case the file exists, new logging messages will be appended; otherwise if the file does not exists, it will be created and the logging messages will be written to the file.

A special case is the parameter value "-": it will be interpreted as standard output.

if a connection is registered as a logger, it must be writable in order to write the logging messages.

Each parameter will be converted to a connection object, which will be closed (when applicable) automatically before R quits.

If the parameter is missing (or set to NA or NULL), no logging will take place.

Value

No value returned: its side effect is used.

Note

Currently, the loggers are stored in a variable in the namespace of ribiosUtils named RIBIOS_LOGGERS. This is only for internal use of the package and may change any time, therefore users are not advised to manipulate this variable directly.

To clear the registered loggers, use clearLog.To flush the registered loggers, use flushLog. Usually it is not necessary to use flushLog in R scripts, since by program exit the active R session will automatically flush and close the connections (in addition, frequent flushing may decrease the program's efficiency). However, if used in interactive sessions, sometimes flushLog is needed to force R write all log files to all connections that are registered.

Author(s)

Jitao David Zhang <[email protected]>

See Also

doLog writes messages iteratively to each connection registered by registerLog.

Examples

## the following code section is not run to prevent issues with pkgdown
## Not run: 
  logfile1 <- tempfile()
  logfile2 <- tempfile()
  logcon3 <- stdout()
  if(.Platform$OS.type == "unix") {
    registerLog("/dev/null")
  } else {
    registerLog(tempfile())
  }
  registerLog(logfile1)
  registerLog(logfile2)
  registerLog(logcon3)

  doLog("Start logging")
  doLog("Do something...")
  doLog("End logging")

  flushLog() ## usually not needed, see notes

  txt1 <- readLines(logfile1)
  txt2 <- readLines(logfile2)

  cat(txt1)
  cat(txt2)

  clearLog()

  registerLog(logfile1, logfile2, logcon3)

  doLog("Start logging - round 2")
  doLog("Do something again ...")
  doLog("End logging - for good")

  flushLog() ## usually not needed, see notes

  txt1 <- readLines(logfile1)
  txt2 <- readLines(logfile2)

  cat(txt1)
  cat(txt2)

  ## clean up files and objects to close unused connections
  closeLoggerConnections()

## End(Not run)

Description

This function wraps relevelsByNamedVec for named vector and relevelsByNotNamedVec for not named vectors

Usage

relevels(
  x,
  refs,
  missingLevels = c("pass", "warning", "error"),
  unrecognisedLevels = c("warning", "pass", "error")
)

Arguments

Value

A vector of factor

See Also

relevelsByNamedVec and relevelsByNotNamedVec

Examples

oldFactor <- factor(c("A", "B", "A", "C", "B"), levels=LETTERS[1:3])
refLevels <- c("B", "C", "A")
refDict <- c("A"="a", "B"="b", "C"="c")
newFactor <- relevels(oldFactor, refLevels)
stopifnot(identical(newFactor, factor(c("A", "B", "A", "C", "B"), levels=c("B", "C", "A"))))
newFactor2 <-  relevels(oldFactor, refDict)
stopifnot(identical(newFactor2, factor(c("a", "b", "a", "c", "b"), levels=c("a", "b", "c"))))

Description

If names contain character strings other than the levels in the old factor and warning is set to TRUE, a warning will be raised.

Usage

relevelsByNamedVec(
  x,
  refs,
  missingLevels = c("pass", "warning", "error"),
  unrecognisedLevels = c("warning", "pass", "error")
)

Arguments

Details

The levels of the factor are the names of the ref vector, and the order of the ref vector matters: it is the levels of the new factor.

Value

A vector of factor

Examples

oldFactor <- factor(c("A", "B", "A", "C", "B"), levels=LETTERS[1:3])
factorDict <- c("A"="a", "B"="b", "C"="c")
newFactor <- relevelsByNamedVec(oldFactor, factorDict)
stopifnot(identical(newFactor, factor(c("a", "b", "a", "c", "b"), levels=c("a", "b", "c"))))
## TODO: test warning and error