makeVenn: High-level function for making Venn diagrams and outputting...

In affycoretools: Functions useful for those doing repetitive analyses with Affymetrix GeneChips

Description

This function is designed to output CSV and HTML tables based on an analysis using the limma or edgeR packages, with output generated using the ReportingTools package. Please note that a DGEGLM object from edgeR is simply converted to an MArrayLM object from limma and then used in the default MArrayLM method, so all arguments for the MArrayLM object pertain to the DGEGLM method as well.

Usage

makeVenn(object, ...)

## S4 method for signature 'MArrayLM'
makeVenn(
  object,
  contrast,
  design,
  groups = NULL,
  collist = NULL,
  p.value = 0.05,
  lfc = 0,
  method = "both",
  adj.meth = "BH",
  titleadd = NULL,
  fileadd = NULL,
  baseUrl = ".",
  reportDirectory = "./venns",
  affy = TRUE,
  probecol = "PROBEID",
  ...
)

## S4 method for signature 'DGEGLM'
makeVenn(
  object,
  contrast,
  design,
  comp.method = c("glmLRT", "glmQLFTest", "glmTreat"),
  lfc = 0,
  ...
)

Arguments

object	An MArrayLM or DGEGLM object.
...	Used to pass other arguments to lower level functions.
contrast	A contrasts matrix, produced either by hand, or by a call to makeContrasts
design	A design matrix.
groups	This argument is used when creating a legend for the resulting HTML pages. If NULL, the groups will be generated using the column names of the design matrix. In general it is best to leave this NULL.
collist	A list containing numeric vectors indicating which columns of the fit, contrast and design matrix to use. If NULL, all columns will be used.
p.value	A p-value to filter the results by.
lfc	A log fold change to filter the results by.
method	One of "same", "both", "up", "down", "sameup", or "samedown". See details for more information.
adj.meth	Method to use for adjusting p-values. Default is 'BH', which corresponds to 'fdr'. Ideally one would set this value to be the same as was used for decideTests.
titleadd	Additional text to add to the title of the HTML tables. Default is NULL, in which case the title of the table will be the same as the filename.
fileadd	Additional text to add to the name of the HTML and CSV tables. Default is NULL.
baseUrl	A character string giving the location of the page in terms of HTML locations. Defaults to "."
reportDirectory	A character string giving the location that the results will be written. Defaults to "./venns"
affy	Boolean. Are these Affymetrix data, and should hyperlinks to the affy website be generated in the HTML tables?
probecol	This argument is used in concert with the preceding argument. If these are Affymetrix data , then specify the column header in the MArrayLM object that contains the Affymetrix IDs. Defaults to "PROBEID", which is the expected result if the data are annotated using a BioC annotation package.
comp.method	Character. For DGEGLM objects, the DGEGLM object must first be processed using one of glmLRT, glmQLFTest, or glmTreat. Choose glmLRT if you fit a model using glmFit, glmQLFTest if you fit a model using glmQLFit, or glmTreat if you fit either of those models, but want to incorporate the log fold change into the comparison.

Details

The purpose of this function is to output HTML and text tables with lists of genes that fulfill the criteria of a call to decideTests as well as the direction of differential expression. This is a high-level function that calls vennSelect2 internally, and is intended to be used with vennPage to create a set of Venn diagrams (on an HTML page) that have clickable links in each cell of the diagram. The links will then pass the end user to individual HTML pages that contain the genes that are represented by the counts in a given cell of the Venn diagram.

In general, the only thing that is needed to create a set of Venn diagrams is a list of numeric vectors that indicate the columns of the contrast matrix that are to be used for a given diagram. See the example below for a better explanation.

Some important things to note: First, the names of the HTML and text tables are extracted from the colnames of the TestResults object, which come from the contrasts matrix, so it is important to use something descriptive. Second, the method argument is analogous to the include argument from vennCounts or vennDiagram. Choosing "both" will select genes that are differentially expressed in one or more comparisons, regardless of direction. Choosing "up" or "down" will select genes that are only differentially expressed in one direction. Choosing "same" will select genes that are differentially expressed in the same direction. Choosing "sameup" or "samedown" will select genes that are differentially expressed in the same direction as well as 'up' or 'down'.

Note that this is different than sequentially choosing "up" and then "down". For instance, a gene that is upregulated in one comparison and downregulated in another comparison will be listed in the intersection of those two comparisons if "both" is chosen, it will be listed in only one comparison for both the "up" and "down" methods, and it will be listed in the union (e.g., not selected) if "same" is chosen.

Unlike vennSelect, this function automatically creates both HTML and CSV output files.

Also please note that this function relys on annotation information contained in the "genes" slot of the "fit" object. If there are no annotation data, then just statistics will be output in the resulting HTML tables.

Value

A list containing the output from calling vennSelect2 on the columns specified by the collist argument. This is intended as input to vennPage, which will use those data to create the HTML page with Venn diagrams with clickable links.

Methods (by class)

• MArrayLM: Make a Venn diagram using an MArrayLM object.

• DGEGLM: Make a Venn diagram using a DGEGLM object.

Author(s)

James W. MacDonald [e-mail]

Examples

  ## Not run: 
    mat <- matrix(rnorm(1e6), ncol = 20)
    design <- model.matrix(~factor(1:4, each=5))
    colnames(design) <- LETTERS[1:4]
    contrast <- matrix(c(1,-1,0,0,1,0,-1,0,1,0,0,-1,0,1,-1,0,0,1,0,-1),
    ncol = 5)
    colnames(contrast) <- paste(LETTERS[c(1,1,1,2,2)],
    LETTERS[c(2,3,4,3,4)], sep = " vs ")
    fit <- lmFit(mat, design)
    fit2 <- contrasts.fit(fit, contrast)
    fit2 <- eBayes(fit2)
    ## two Venn diagrams - a 3-way Venn with the first three contrasts
    ## and a 2-way Venn with the last two contrasts
    collist <- list(1:3,4:5)
    venn <- makeVenn(fit2, contrast, design, collist = collist)
    vennPage(venn, "index.html", "Venn diagrams")
    
## End(Not run)

affycoretools documentation built on Nov. 8, 2020, 6 p.m.