R3CPET (version 1.4.2)

ChiapetExperimentData-class: 3CPET used raw data

Description

The ChiapetExperimentData class is a container for storing the set of raw data used by 3CPET to do the prediction.

Usage

ChiapetExperimentData(pet='', tfbs='', ppi=NULL,   ## loadPETs options IsBed=TRUE, petHasHeader=FALSE, dist=1000, ## loadTFBS options tfbsHasHeader=FALSE, ## loadPPI options ppiType=c("HPRD","Biogid"), filter=FALSE, term="GO:0005634", annot=NULL, RPKM= NULL, threshold=1 )

Arguments

pet

(optional) a ChIAP-PET interactions file path or a GRanges object. The GRanges object should have a column named PET_ID. details on the file format can be found on the loadPETs help page.

tfbs

(optional) a file path to a BED file containing transcription factors binding site or a GRanges object. The GRanges object should have a metadata column named TF.

ppi

(optional) an igraph object that contains a user defined protein-protein interaction network. if this parameter is not specified, the ppiType paramter will be used.

IsBed

(optional) considered only if the pet parameter is a file path. More info about this paramters can be found in the loadPETs help page.

petHasHeader

(optional) logical. Indicates if the ChIA-PET interactions file has a header or not.

dist

(optional) The size of the region to consider arround the center of the interacting regions.

tfbsHasHeader

(optional) logical. Indicates if the TFBS file has a header or not.

ppiType

(optional) considered only if the ppi paramter is not specified. This paramter tell the pakage to load one of the PPI (HPRD or Biogrid) shiped with the package.

filter

(optional) logical. whether of not to filter the PPI network. if the RPKM paramter is specified then the RPKM dataset incorporated with the package will be used. if you want to to your own way of filtering, you ca set filter = FALSE and pass an already processed PPI to the ppi paramterer.

term

(optional) the GO term used to filter the nodes of the PPI. this is different from the filter parameter. in the filter parameter the PPI nodes are filtered by their gene expression, while in the term parameter they are filtered by their genomic location. by default "GO:0005634" is used for filtering.

annot

(optional) the annotation dataset used for filtering by default the geneLocations is used. The user can pass a custom data.frame. For more details check the loadPPI help page.

RPKM

(optional) a data.frame object that contains the expression value of each genes. by default the RPKMS dataset will be used (expression value in K562 celline). For more information about the format of the data passed to this parameter please check the loadPPI

threshold

(optional) threshold value used to filter gene expression. default: 1.

Value

Constructs a ChiapetExperimentData object with the specified fields populated.

Slots

pet

: Object of class GRanges that stores the genomic coordinated of the interactions. it can be populated using the method loadPETs

tfbs

: Object of class GRanges that stores the TF binding site. it can be populated using the method loadTFBS. NOTE: the TFBS locations can be obtained from a ChIP-Seq experiment or a motif finding software. for more information on the format of the provided data check loadTFBS

ppi

Object of class "igraph" used as the background PPI for further analysis. it can be populated using the method loadPPI

.dt

Object of class "list" contains a collection of data.table serving as indexes used internally by the package (not expected to be manipulated by the user). it can be populated using the method createIndexes

Accessors

The following methods can be used to get the content of a ChiapetExperimentData object x :

pet(x), pet(x) <- value: Get ChIA-PET interactions encoded as a GRanges object in x. The returned GRanges objects contains an attribute PET_ID in which the left side have an id of the form PET#\d+\.1 and the right side interaction have an id of the form PET#\d+\.2. for more information check loadPETs

          seqnames             ranges strand |      PET_ID
                          | 
      [1]     chr1 [1240734, 1242734]      * |     PET#1.1
      [2]     chr1 [1242224, 1244224]      * |     PET#1.2
      [3]     chr1 [1282208, 1284208]      * |     PET#2.1
      [4]     chr1 [1283334, 1285334]      * |     PET#2.2
      [5]     chr1 [1370371, 1372371]      * |     PET#3.1
      [6]     chr1 [1371822, 1373822]      * |     PET#3.2       
  

tfbs(x), tfbs(x) <- value: Get the GRanges storing the transcription factor binding sites.

ppi(x), ppi(x) <- value: Returns an igraph object used as a background PPI. check the loadPPI for more information.

Details

The ChiapetExperimentData class stores the genomic coordinates of the ChIA-PET interactions, the binding sites of the different transcription factor (TFBS) and the background protein-protein interaction (PPI) network used to infer the final chromatin maintainer networks.

References

Li G, Fullwood MJ, Xu H et al.ChIA-PET tool for comprehensive chromatin interaction analysis with paired-end tag sequencing. Genome Biology 2010, 11(2):R22

Mohamed Nadhir D, Yang C et al 3CPET: Finding Co-factor Complexes in Chia-PET experiment using a Hierarchical Dirichlet Process, ....

See Also

loadPETs, loadTFBS , loadPPI

Examples

## for example Reading ChIA-PET interaction results generated from ChIA-PET tool
## it should be formatted as follow:

## -------------------------------------------------------------------------------------------
## chromleft startleft endleft chromright startright endright counts      pvalue      qvalue
##      chr1    872113  879175       chr1     933836   938416     12 1.84529e-30 6.90983e-28
##      chr1    874165  879175       chr1     933340   938306     10 1.23139e-25 3.58932e-23
##      chr1    889676  896594       chr1     933897   938982     13 4.91311e-36 2.33753e-33
##      chr1    898753  907581       chr1     931133   939571     19 0.00000e+00 0.00000e+00
##      chr1    910103  918775       chr1     930834   938627     15 2.20004e-43 1.32812e-40
##      chr1    919314  922154       chr1     934212   937864      6 3.70292e-21 7.88551e-19  
##---------------------------------------------------------------------------------------------

## The counts, pvalue and qvalue fields are not considered in our case 
## it is up to the user to filter the interactions.

## The TFBS should be a BED file that contain the chromosome, start, end and the TF name 

## Not run: 
# 
#   ## load the different datasets
#   petFile <- file.path(system.file("example",package="R3CPET"),"HepG2_interactions.txt")  
#   tfbsFile <- file.path(system.file("example",package="R3CPET"),"HepG2_TF.txt.gz")  
#   
#   x <- ChiapetExperimentData(pet = petFile, tfbs=  tfbsFile, IsBed = FALSE, ppiType="HPRD", filter= TRUE) 
#   ## build the diffrent indexes
#   x <- createIndexes(x)
#   x
#   
#   ## Pass objects instead of files.
#   pet <- sample(pet(x),size = 20,replace = TRUE )
#   tfbs <- sample(tfbs(x), size=300, replace=TRUE)
#   ppi <- ppi(x)
#   tst <- ChiapetExperimentData(pet = pet, tfbs= tfbs, ppi=ppi) 
#   tst <- createIndexes(tst)
#   tst
# ## End(Not run)