Package 'ELMER'

Description

Calculate the distance between probe and gene TSS

Usage

addDistNearestTSS(data, NearGenes, genome, met.platform, cores = 1)

Arguments

Examples

## Not run: 
 data <- ELMER:::getdata("elmer.data.example")
 NearbyGenes <- GetNearGenes(
  data = data, 
  probes = c("cg15924102", "cg24741609"),  
  numFlankingGenes = 20
 )
 NearbyGenes <- addDistNearestTSS(data = data, NearGenes = NearbyGenes)

## End(Not run)

Description

Adds mutation information to MAE

Usage

addMutCol(
  data,
  disease,
  genes,
  mutant_variant_classification = c("Frame_Shift_Del", "Frame_Shift_Ins",
    "Missense_Mutation", "Nonsense_Mutation", "Splice_Site", "In_Frame_Del",
    "In_Frame_Ins", "Translation_Start_Site", "Nonstop_Mutation")
)

Arguments

Examples

## Not run: 
 data <- ELMER:::getdata("elmer.data.example") # Get data from ELMER.data
 data <- ELMER:::addMutCol(data, "LUSC","TP53")

## End(Not run)

Description

Idea For a given region R linked to X genes G merge R with nearest TSS for G (multiple) this will increse nb of lines i.e R1 - G1 - TSS1 - DIST1 R1 - G1 - TSS2 - DIST2 To vectorize the code: make a granges from left and onde from right and find distance collapse the results keeping min distance for equals values

Usage

calcDistNearestTSS(links, TRange, tssAnnot)

Arguments

Author(s)

Tiago C. Silva

Examples

## Not run: 
 data <- ELMER:::getdata("elmer.data.example")
  NearbyGenes <- GetNearGenes(
    data = data, 
    probes = c("cg15924102", "cg24741609"),  
    numFlankingGenes = 20
 )
 
  NearbyGenes <- ELMER:::calcDistNearestTSS(
       links = NearbyGenes,
       tssAnnot =  getTSS(genome = "hg38"),
       TRange = rowRanges(getMet(data))
  )

## End(Not run)

Description

Calculates fisher exact test

Usage

calculateEnrichement(foreground, background)

Arguments

Examples

foreground <- Matrix::Matrix(sample(0:1,size = 100,replace = TRUE), 
                             nrow = 10, ncol = 10,sparse = TRUE)
rownames(foreground) <- paste0("region",1:10)
colnames(foreground) <- paste0("motif",1:10)
background <- Matrix::Matrix(sample(0:1,size = 100,replace = TRUE), 
                             nrow = 10, ncol = 10,sparse = TRUE)
rownames(background) <- paste0("region",1:10)
colnames(background) <- paste0("motif",1:10)
calculateEnrichement(foreground,background)

Description

Create a bigwig for IGV visualization of DNA methylation data (Array)

Usage

createBigWigDNAmetArray(
  data = NULL,
  genome = "hg38",
  met.platform = "450K",
  track.names = NULL,
  dir = "IGV_tracks"
)

Arguments

Author(s)

Tiago Chedraoui Silva (tiagochst at gmail.com)

Examples

## Not run:             
 data <- assay(getMet(ELMER:::getdata("elmer.data.example")))
 createBigWigDNAmetArray(data = data, met.platform = "450K", genome = "hg38")
 
## End(Not run)

Description

Create a junction track for IGV visualization of interection

Usage

createIGVtrack(
  pairs,
  met.platform = "450K",
  genome = "hg38",
  filename = "ELMER_interactions.bed",
  color.track = "black",
  track.name = "junctions",
  gene.symbol = NULL,
  all.tss = TRUE
)

Arguments

Author(s)

Tiago Chedraoui Silva (tiagochst at gmail.com)

Examples

## Not run:             
data <- ELMER:::getdata("elmer.data.example")
nearGenes <-GetNearGenes(TRange=getMet(data)[c("cg00329272","cg10097755"),],
                         geneAnnot=getExp(data))
Hypo.pair <- get.pair(data=data,
                       nearGenes=nearGenes,
                       permu.size=5,
                       group.col = "definition",
                       group1 = "Primary solid Tumor", 
                       group2 = "Solid Tissue Normal",
                       raw.pvalue = 0.2,
                       Pe = 0.2,
                       dir.out="./",
                       label= "hypo")
 createIGVtrack(Hypo.pair,met.platform = "450K", genome = "hg38")
 
## End(Not run)

Description

This function will receive a gene expression and DNA methylation data objects and create a Multi Assay Experiment.

Usage

createMAE(
  exp,
  met,
  colData,
  sampleMap,
  linearize.exp = FALSE,
  filter.probes = NULL,
  met.na.cut = 0.2,
  filter.genes = NULL,
  met.platform = "450K",
  genome = NULL,
  save = TRUE,
  save.filename,
  TCGA = FALSE
)

Arguments

Value

A MultiAssayExperiment object

Examples

# NON TCGA example: matrices has different column names
gene.exp <- S4Vectors::DataFrame(
   sample1.exp = c("ENSG00000141510"=2.3,"ENSG00000171862"=5.4),
   sample2.exp = c("ENSG00000141510"=1.6,"ENSG00000171862"=2.3)
)
                  
dna.met <- S4Vectors::DataFrame(
   sample1.met = c("cg14324200"=0.5,"cg23867494"=0.1),
   sample2.met =  c("cg14324200"=0.3,"cg23867494"=0.9)
)
sample.info <- S4Vectors::DataFrame(
  primary =  c("sample1","sample2"), 
  sample.type = c("Normal", "Tumor")
 )
 
sampleMap <- S4Vectors::DataFrame(
    assay = c("Gene expression","DNA methylation","Gene expression","DNA methylation"),
    primary = c("sample1","sample1","sample2","sample2"), 
    colname = c("sample1.exp","sample1.met","sample2.exp","sample2.met")
)

mae <- createMAE(
    exp = gene.exp, 
    met = dna.met, 
    sampleMap = sampleMap, 
    met.platform ="450K",
    colData = sample.info, 
    genome = "hg38"
) 
                 
# You can also use sample Mapping and Sample information tables from a tsv file
# You can use the createTSVTemplates function to create the tsv files
readr::write_tsv(as.data.frame(sampleMap), path = "sampleMap.tsv")
readr::write_tsv(as.data.frame(sample.info), path = "sample.info.tsv")

mae <- createMAE(
   exp = gene.exp, 
   met = dna.met, 
   sampleMap = "sampleMap.tsv", 
   met.platform ="450K",
   colData = "sample.info.tsv", 
   genome = "hg38"
) 
                 
# NON TCGA example: matrices has same column names
gene.exp <- S4Vectors::DataFrame(sample1 = c("ENSG00000141510"=2.3,"ENSG00000171862"=5.4),
                  sample2 = c("ENSG00000141510"=1.6,"ENSG00000171862"=2.3))
dna.met <- S4Vectors::DataFrame(sample1 = c("cg14324200"=0.5,"cg23867494"=0.1),
                       sample2=  c("cg14324200"=0.3,"cg23867494"=0.9))
sample.info <- S4Vectors::DataFrame(primary =  c("sample1","sample2"), 
                                    sample.type = c("Normal", "Tumor"))
sampleMap <- S4Vectors::DataFrame(
     assay = c("Gene expression","DNA methylation","Gene expression","DNA methylation"),
    primary = c("sample1","sample1","sample2","sample2"), 
    colname = c("sample1","sample1","sample2","sample2")
)
mae <- createMAE(
  exp = gene.exp, 
  met = dna.met, 
  sampleMap = sampleMap, 
  met.platform ="450K",
  colData = sample.info, 
  genome = "hg38"
) 

## Not run: 
   # TCGA example using TCGAbiolinks
   # Testing creating MultyAssayExperiment object
   # Load library
   library(TCGAbiolinks)
   library(SummarizedExperiment)
   
   samples <- c(
      "TCGA-BA-4074", "TCGA-BA-4075", "TCGA-BA-4077", "TCGA-BA-5149",
       "TCGA-UF-A7JK", "TCGA-UF-A7JS", "TCGA-UF-A7JT", "TCGA-UF-A7JV"
   )
   
   #1) Get gene expression matrix
   query.exp <- GDCquery(
       project = "TCGA-HNSC", 
       data.category = "Transcriptome Profiling", 
       data.type = "Gene Expression Quantification", 
       workflow.type = "STAR - Counts",
       barcode = samples
   )
   
   GDCdownload(query.exp)
   exp.hg38 <- GDCprepare(query = query.exp)
   
   # DNA Methylation
   query.met <- GDCquery(
      project = "TCGA-HNSC",
      data.category = "DNA Methylation", 
      data.type = "Methylation Beta Value",
      barcode = samples,
      platform = "Illumina Human Methylation 450"
   )
   
   GDCdownload(query.met)
   met <- GDCprepare(query = query.met)
   
   distal.enhancer <- get.feature.probe(
      genome = "hg19",
      met.platform = "450k"
   )
   
   
   mae.hg38 <- createMAE(
     exp = exp.hg38, met = met, 
     TCGA = TRUE, genome = "hg38",  
     filter.probes = distal.enhancer
   )
   values(getExp(mae.hg38))
   
   # Consisering it is TCGA and not SE
   mae.hg19.test <- createMAE(
     exp = assay(exp.hg19), met =  assay(met), 
     TCGA = TRUE, genome = "hg19",  
     filter.probes = distal.enhancer
   )
   
   mae.hg38 <- createMAE(
     exp = assay(exp.hg38), 
     met = assay(met), 
     TCGA = TRUE, 
     genome = "hg38",  
     filter.probes = distal.enhancer
   )
   values(getExp(mae.hg38))
   
   # Consisering it is not TCGA and SE
   # DNA methylation and gene expression Objects should have same sample names in columns
   not.tcga.exp <- exp.hg19 
   colnames(not.tcga.exp) <- substr(colnames(not.tcga.exp),1,15)
   not.tcga.met <- met 
   colnames(not.tcga.met) <- substr(colnames(not.tcga.met),1,15)
   
   phenotype.data <- data.frame(
      row.names = colnames(not.tcga.exp),
      primary = colnames(not.tcga.exp),
      samples = colnames(not.tcga.exp), 
      group = c(rep("group1",4),rep("group2",4))
   )
   
   distal.enhancer <- get.feature.probe(genome = "hg19",met.platform = "450k")
   mae.hg19 <- createMAE(
      exp = not.tcga.exp, 
      met =  not.tcga.met, 
      TCGA = FALSE, 
      filter.probes = distal.enhancer,
      genome = "hg19", 
      colData = phenotype.data
   )

## End(Not run)
createMAE

Description

This will output a list each TF motif and TFs that binding the motis. Multiple TFs may recognize a same motif such as TF family. The association between each motif famil and transcription factor was created using the (HOCOMOCO)[https://hocomoco11.autosome.org/human/mono?full=true] which TF structural families was created according to TFClass [@wingender2014tfclass] This data is stored as a list whose elements are motifs and contents for each element are TFs which recognize the same motif that is the name of the element. This data is used in function get.TFs in ELMER to identify the real regulator TF whose motif is enriched in a given set of probes and expression associate with average DNA methylation of these motif sites.

Usage

createMotifRelevantTfs(classification = "family")

Arguments

Value

A list of TFs and its family members

Description

This function will create a text file with the date of the last run, which aanalysis were performed, the values of the arguments so the user can keep track

Usage

createSummaryDocument(
  analysis = "all",
  argument.values = "defaults",
  genome = NULL,
  mae.path = NULL,
  mode = NULL,
  direction = NULL,
  group.col = NULL,
  group1 = NULL,
  group2 = NULL,
  results.path = NULL
)

Arguments

Description

This function will receive the DNA methylation and gene expression matrix and will create some examples of table for the argument colData and sampleMap used in ceeateMae function.

Usage

createTSVTemplates(met, exp)

Arguments

Examples

gene.exp <- S4Vectors::DataFrame(sample1.exp = c("ENSG00000141510"=2.3,"ENSG00000171862"=5.4),
                  sample2.exp = c("ENSG00000141510"=1.6,"ENSG00000171862"=2.3))
dna.met <- S4Vectors::DataFrame(sample1.met = c("cg14324200"=0.5,"cg23867494"=0.1),
                       sample2.met =  c("cg14324200"=0.3,"cg23867494"=0.9))
createTSVTemplates(met = dna.met, exp = gene.exp)

Description

ELMER is designed to use DNA methylation and gene expression from a large number of samples to infere regulatory element landscape and transcription factor network in primary tissue.

Description

To find for each probe the know motif we will use HOMER software (http://homer.salk.edu/homer/). Homer and genome should be installed before this function is executed Step: 1 - get DNA methylation probes annotation with the regions 2 - Make a bed file from it 3 - Execute section: Finding Instance of Specific Motifs from http://homer.salk.edu/homer/ngs/peakMotifs.html to the HOCOMOCO TF motifs Also, As HOMER is using more RAM than the available we will split the files in to 100k probes. Obs: for each probe we create a winddow of 500 bp (-size 500) around it. This might lead to false positives, but will not have false negatives. The false posives will be removed latter with some statistical tests.

Usage

findMotifRegion(
  regions,
  output.filename = "mapped_motifs_regions.txt",
  region.size = NULL,
  genome = "hg38",
  nstep = 10000,
  cores = 1
)

Arguments

Examples

## Not run: 
 # use the center of the region and +-250bp around it
 gr0 <- GRanges(Rle(c("chr2", "chr2", "chr1", "chr3"), 
                    c(1, 3, 2, 4)
                    ), 
               IRanges(1:10, width=10:1)
               )
 names(gr0) <- paste0("ID",c(1:10))
 findMotifRegion(regions = gr0, region.size = 500, genome = "hg38", cores = 1)
 
 # use the region size itself
 gr1 <- GRanges(Rle(c("chr2", "chr2", "chr1", "chr3"), c(1, 3, 2, 4)), 
                IRanges(1:10, width=sample(200:1000,10)))
 names(gr1) <- paste0("ID",c(1:10))
 findMotifRegion(regions = gr0, genome = "hg38", cores = 1)

## End(Not run)

Description

get.diff.meth applys one-way t-test to identify the CpG sites that are significantly hypo/hyper-methyalated using proportional samples (defined by minSubgroupFrac option) from group 1 and group 2. The P values will be adjusted by Benjamini-Hochberg method. Option pvalue and sig.dif will be the criteria (cutoff) for selecting significant differentially methylated CpG sites. If save is TURE, two getMethdiff.XX.csv files will be generated (see detail).

Usage

get.diff.meth(
  data,
  diff.dir = "hypo",
  cores = 1,
  mode = "unsupervised",
  minSubgroupFrac = 0.2,
  pvalue = 0.01,
  group.col,
  min.samples = 5,
  group1,
  group2,
  test = t.test,
  sig.dif = 0.3,
  dir.out = "./",
  save = TRUE
)

Arguments

Details

save: When save is TRUE, function will generate two XX.csv files.The first one is named getMethdiff.hypo.probes.csv (or getMethdiff.hyper.probes.csv depends on diff.dir). The first file contains all statistic results for each probe. Based on this file, user can change different P value or sig.dir cutoff to select the significant results without redo the analysis. The second file is named getMethdiff.hypo.probes.significant.csv (or getMethdiff.hyper.probes.significant.csv depends on diff.dir). This file contains statistic results for the probes that pass the significant criteria (P value and sig.dir). When save is FALSE, a data frame R object will be generate which contains the same information with the second file.

Value

Statistics for all probes and significant hypo or hyper-methylated probes.

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

data <- ELMER:::getdata("elmer.data.example")
Hypo.probe <- get.diff.meth(data,
                            diff.dir="hypo",
                            group.col = "definition",
                            group1 = "Primary solid Tumor",
                            group2 = "Solid Tissue Normal",
                            sig.dif = 0.1) # get hypomethylated probes
Hyper.probe <- get.diff.meth(data,
                            diff.dir="hyper",
                            group.col = "definition",
                            sig.dif = 0.1) # get hypomethylated probes

Description

get.enriched.motif is a function make use of Probes.motif data from ELMER.data package to calculate the motif enrichment Odds Ratio and 95% confidence interval for a given set of probes using fisher test function, after performing the Fisher's exact test, the results for all transcription factors are corrected for multiple testing with the Benjamini-Hochberg procedure. If save is TURE, two output files will be saved: getMotif.XX.enriched.motifs.rda and getMotif.XX.motif.enrichment.csv (see detail).

Usage

get.enriched.motif(data, probes.motif, probes, min.motif.quality = "DS",
                   background.probes,  pvalue = 0.05, lower.OR = 1.1, min.incidence = 10,
                   dir.out = "./", label = NULL, save = TRUE, plot.title="")

Arguments

Details

background.probes: For enhancer study, it is better to use probes within distal enhancer probes as background.probes. For promoter study, it is better to use probes within promoter regions as background.probes. Because enhancer and promoter have different CG content and harbors different clusters of TFs motif.

save: if save is TRUE, two files will be save on the disk. The first file is getMotif.XX.motif.enrichment.csv (XX depends on option label). This file reports the Odds Ratio and 95% confidence interval for these Odds Ratios which pass the significant cutoff (lower.OR and min.incidence). The second file is getMotif.XX.enriched.motifs.rda (XX depends on option lable). This file contains a list R object with enriched motifs as name and probes containing the enriched motif as contents. This object will be used in get.TFs function. if save is FALSE, the function will return a R object which is the same with second file.

Value

A list contains enriched motifs with the probes regions harboring the motif.

A list (R object) with enriched motifs as name and probes containing the enriched motif as contents. And hypo.motif.enrichment.pdf plot will be generated.

Author(s)

Lijing Yao (creator: [email protected])

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

probes <- c("cg00329272","cg10097755","cg08928189", "cg17153775","cg21156590",
"cg19749688","cg12590404","cg24517858","cg00329272","cg09010107",
"cg15386853", "cg10097755", "cg09247779","cg09181054","cg19371916")
  data <- tryCatch(ELMER:::getdata("elmer.data.example"), error = function(e) {
  message(e)
  data(elmer.data.example, envir = environment())
  })
bg <- rownames(getMet(data))
data(Probes.motif.hg38.450K,package = "ELMER.data")
enriched.motif <- get.enriched.motif(
   probes.motif = Probes.motif.hg38.450K,
   probes = probes,
   background.probes = bg,
   pvalue = 1,
   min.incidence = 2,
   label = "hypo"
)
# If the MAE is set, the background and the probes.motif will
# be automatically set
enriched.motif <- get.enriched.motif(
    data = data,
    min.motif.quality = "DS",
    probes=probes,
    pvalue = 1,
    min.incidence=2,
    label="hypo"
)

Description

get.feature.probe is a function to select the probes falling into distal feature regions or promoter regions.

This function selects the probes on HM450K that either overlap distal biofeatures or TSS promoter.

Usage

get.feature.probe(
  feature = NULL,
  TSS,
  genome = "hg38",
  met.platform = "450K",
  TSS.range = list(upstream = 2000, downstream = 2000),
  promoter = FALSE,
  rm.chr = NULL
)

Arguments

Details

In order to get real distal probes, we use more comprehensive annotated TSS by both GENCODE and UCSC. However, to get probes within promoter regions need more accurate annotated TSS such as UCSC. Therefore, there are different settings for promoter and distal probe selection. But user can specify their own favorable TSS annotation. Then there won't be any difference between promoter and distal probe selection. @return A GRanges object contains the coordinate of probes which locate within promoter regions or distal feature regions such as union enhancer from REMC and FANTOM5. @usage get.feature.probe( feature, TSS, TSS.range = list(upstream = 2000, downstream = 2000), promoter = FALSE, rm.chr = NULL )

Value

A GRange object containing probes that satisfy selecting critiria.

Examples

# get distal enhancer probe
## Not run: 
Probe <- get.feature.probe()

## End(Not run)
# get promoter probes
## Not run: 
Probe <- get.feature.probe(promoter=FALSE)

## End(Not run)
# get distal enhancer probe remove chrX chrY
Probe2 <- get.feature.probe(rm.chr=c("chrX", "chrY"))

Description

get.pair is a function to predict enhancer-gene linkages using associations between DNA methylation at enhancer CpG sites and expression of 20 nearby genes of the CpG sites (see reference). Two files will be saved if save is true: getPair.XX.all.pairs.statistic.csv and getPair.XX.pairs.significant.csv (see detail).

Usage

get.pair(data,
         nearGenes,
         minSubgroupFrac = 0.4,
         permu.size = 10000,
         permu.dir = NULL,
         raw.pvalue = 0.001,
         Pe = 0.001,
         mode = "unsupervised",
         diff.dir = NULL,
         dir.out = "./",
         diffExp = FALSE,
         group.col,
         group1 = NULL,
         group2 = NULL,
         cores = 1,
         correlation = "negative",
         filter.probes = TRUE,
         filter.portion = 0.3,
         filter.percentage = 0.05,
         label = NULL,
         addDistNearestTSS = FALSE,
         save = TRUE)

Arguments

Value

Statistics for all pairs and significant pairs

Author(s)

Lijing Yao (creator: [email protected]) Tiago C Silva (maintainer: [email protected])

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

data <- ELMER:::getdata("elmer.data.example")
nearGenes <- GetNearGenes(TRange=getMet(data)[c("cg00329272","cg10097755"),],
                         geneAnnot=getExp(data))
Hypo.pair <- get.pair(data=data,
                       nearGenes=nearGenes,
                       permu.size=5,
                       group.col = "definition",
                       group1 = "Primary solid Tumor",
                       group2 = "Solid Tissue Normal",
                       raw.pvalue = 0.2,
                       Pe = 0.2,
                       dir.out="./",
                       label= "hypo")

Hypo.pair <- get.pair(data = data,
                      nearGenes = nearGenes,
                      permu.size = 5,
                      raw.pvalue = 0.2,
                      Pe = 0.2,
                      dir.out = "./",
                      diffExp = TRUE,
                      group.col = "definition",
                      group1 = "Primary solid Tumor",
                      group2 = "Solid Tissue Normal",
                      label = "hypo")

Description

get.permu is a function to use the same statistic model to calculate random enhancer-gene pairs. Based on the permutation value, empirical P value can be calculated for the real enhancer-gene pair (see reference).

Usage

get.permu(data,
          geneID,
          methy = NULL,
          unmethy = NULL,
          percentage = 0.2,
          rm.probes = NULL,
          correlation = "negative",
          permu.size = 10000,
          permu.dir = NULL,
          cores = 1)

Arguments

Value

Permutations

Note

Permutation is the most time consuming step. It is recommended to use multiple cores for this step. Default permutation time is 1000 which may need 12 hrs by 4 cores. However 10,000 permutations is recommended to get high confidence results. But it may cost 2 days.

Author(s)

Lijing Yao (creator: [email protected]) Tiago C Silva (maintainer: [email protected])

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

data <- ELMER:::getdata("elmer.data.example")
permu <-get.permu(data = data,
                  geneID=rownames(getExp(data)),
                  rm.probes=c("cg00329272","cg10097755"),
                  permu.size=5)

Description

Calculate empirical Pvalue

Usage

Get.Pvalue.p(U.matrix, permu)

Arguments

Value

A data frame with empirical Pvalue.

Description

summarize MR TF as a binary table with 1 if TF was found in the analysis, 0 if not

Usage

get.tab(dir, classification, top = TRUE)

Arguments

Examples

## Not run: 
dir.create("out")
dir.create("out2")
data <- tryCatch(
  ELMER:::getdata("elmer.data.example"),
  error = function(e) {
    message(e)
     data(elmer.data.example, envir = environment())
  })
enriched.motif <- list("P53_HUMAN.H11MO.1.A"= c("cg00329272", "cg10097755", "cg08928189",
                                 "cg17153775", "cg21156590", "cg19749688", "cg12590404",
                                 "cg24517858", "cg00329272", "cg09010107", "cg15386853",
                                 "cg10097755", "cg09247779", "cg09181054"))
TF <- get.TFs(data,
              enriched.motif,
              group.col = "definition",
              group1 = "Primary solid Tumor",
              group2 = "Solid Tissue Normal",
              TFs = data.frame(
                     external_gene_name=c("TP53","TP63","TP73"),
                     ensembl_gene_id= c("ENSG00000141510",
                                        "ENSG00000073282",
                                        "ENSG00000078900"),
                     stringsAsFactors = FALSE),
                     dir.out = "out",
             label="hypo")
TF <- get.TFs(data,
              enriched.motif,
              group.col = "definition",
              group1 = "Primary solid Tumor",
              group2 = "Solid Tissue Normal",
              TFs = data.frame(
                     external_gene_name=c("TP53","TP63","TP73"),
                     ensembl_gene_id= c("ENSG00000141510",
                                        "ENSG00000073282",
                                        "ENSG00000078900"),
                     stringsAsFactors = FALSE),
                     dir.out = "out2",
             label="hypo")
 ta.family <- get.tab(dir = c("out","out2"),classification = "family")
 ta.subfamily <- get.tab(dir = c("out","out2"),classification = "subfamily")    
 unlink("out")         
 unlink("out2")         

## End(Not run)

Description

Code used to create matrix for MR TF heatmap

Usage

get.tabs(dir, classification = "family", top = TRUE)

Arguments

Examples

## Not run: 
elmer.results <- dirname(
dir(path = "analysis",
   pattern = "*.hypo.pairs.significant.csv",
   recursive = T,
   full.names = T,
   all.files = T))
tabs <- get.tabs(dir = elmer.results, classification = "subfamily")

## End(Not run)

Description

get.TFs is a function to identify regulatory TFs based on motif analysis and association analysis between the probes containing a particular motif and expression of all known TFs. If save is true, two files will be saved: getTF.XX.significant.TFs.with.motif.summary.csv and getTF.hypo.TFs.with.motif.pvalue.rda (see detail).

Usage

get.TFs(data,
          enriched.motif,
          TFs,
          group.col,
          group1,
          group2,
          mode = "unsupervised",
          correlation = "negative",
          diff.dir = NULL,
          motif.relevant.TFs,
          minSubgroupFrac = 0.4,
          dir.out = "./",
          label = NULL,
          save.plots = FALSE,
          cores = 1,
          topTFper = 0.05, 
          save = TRUE)

Arguments

Details

save: If save is ture, two files will be saved. The first file is getTF.XX.significant.TFs.with.motif.summary.csv (XX depends on option lable). This file contain the regulatory TF significantly associate with average DNA methylation at particular motif sites. The second file is getTF.hypo.TFs.with.motif.pvalue.rda (XX depends on option label). This file contains a matrix storing the statistic results for significant associations between TFs (row) and average DNA methylation at motifs (column). If save is false, a data frame which contains the same content with the first file will be reported.

Value

Potential responsible TFs will be reported in a dataframe with 4 columns:

• motif: the names of motif.

• top.potential.TF.family: the highest ranking upstream TFs which are known recognized the motif. First item in potential.TFs.family

• top.potential.TF.subfamily: the highest ranking upstream TFs which are known recognized the motif. First item in potential.TFs.subfamily

• potential.TFs.family: TFs which are within top 5% list and are known recognized the motif (considering family classification).

• potential.TFs.subfamily: TFs which are within top 5% list and are known recognized the motif (considering subfamily classification).

• top_5percent: all TFs which are within top 5% list.

Author(s)

Lijing Yao (creator: [email protected]) Tiago C Silva (maintainer: [email protected])

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

data <- tryCatch(
  ELMER:::getdata("elmer.data.example"),
  error = function(e) {
    message(e)
     data(elmer.data.example, envir = environment())
  })
enriched.motif <- list(
   "P53_HUMAN.H11MO.1.A"= c(
      "cg00329272", "cg10097755", "cg08928189",
      "cg17153775", "cg21156590", "cg19749688", "cg12590404",
      "cg24517858", "cg00329272", "cg09010107", "cg15386853",
      "cg10097755", "cg09247779", "cg09181054"
      )
)
TF <- get.TFs(
   data,
   enriched.motif,
   group.col = "definition",
   group1 = "Primary solid Tumor",
   group2 = "Solid Tissue Normal",
   TFs = data.frame(
         external_gene_name=c("TP53","TP63","TP73"),
         ensembl_gene_id= c(
           "ENSG00000141510",
           "ENSG00000073282",
           "ENSG00000078900"
           ),
           stringsAsFactors = FALSE
           ),
 label = "hypo"
)
# This case will use Uniprot dabase to get list of Trasncription factors
TF <- get.TFs(
  data,
  group.col = "definition",
  group1 = "Primary solid Tumor",
  group2 = "Solid Tissue Normal",
  enriched.motif,
  label = "hypo"
)

Description

get450K to download HM40K DNA methylation data for certain cancer types from TCGA website. @description get450K is a function to download latest version of HM450K DNA methylation for all samples of certain cancer types from GDC website.

Usage

get450K(disease, basedir="./Data",filter=0.2, genome = "hg38")

Arguments

Value

Download all DNA methylation from HM450K level 3 data for the specified disease.

Description

getClinic is a function to download latest version of clinic data for all samples of certain cancer types from TCGA website.

Usage

getClinic(disease, basedir = "./Data")

Arguments

Value

Download all clinic information for the specified disease.

Description

Get Gene expression object from MAE

Usage

getExp(data)

Arguments

Description

Get Gene expression object samples from MAE

Usage

getExpSamples(data)

Arguments

Description

getGeneID to report gene id from symbol

Usage

getGeneID(data, symbol)

Arguments

Value

The gene ID for these gene symbols

Examples

data <- ELMER:::getdata("elmer.data.example")
getGeneID(data, symbol="ZNF697")

Description

Get DNA methylation object from MAE

Usage

getMet(data)

Arguments

Description

Get DNA methylation object samples from MAE

Usage

getMetSamples(data)

Arguments

Description

GetNearGenes is a function to collect equal number of gene on each side of one locus. It can receite either multi Assay Experiment with both DNA methylation and gene Expression matrix and the names of probes to select nearby genes, or it can receive two granges objects TRange and geneAnnot.

Usage

GetNearGenes(
  data = NULL,
  probes = NULL,
  geneAnnot = NULL,
  TRange = NULL,
  numFlankingGenes = 20
)

Arguments

Value

A data frame of nearby genes and information: genes' IDs, genes' symbols, distance with target and side to which the gene locate to the target.

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

geneAnnot <- getTSS(genome = "hg38")
probe <- GenomicRanges::GRanges(seqnames = c("chr1","chr2"), 
range=IRanges::IRanges(start = c(16058489,236417627), end= c(16058489,236417627)), 
name= c("cg18108049","cg17125141"))
names(probe) <- c("cg18108049","cg17125141")
NearbyGenes <- GetNearGenes(numFlankingGenes = 20,geneAnnot=geneAnnot,TRange=probe)

Description

This function will receive a pair gene probes and will return a random object with the following pattern, if a probe is linked to R1 and L3 genes the random pairs will be a random probes (a distal probe not in the input pairs) also linked to its R1 and L3 gene.

Usage

getRandomPairs(pairs, genome = "hg38", met.platform = "450K", cores = 1)

Arguments

Value

A data frame with the random linkages

Examples

## Not run: 
 data <- ELMER:::getdata("elmer.data.example")
 nearGenes <- GetNearGenes(TRange=getMet(data)[c("cg00329272","cg10097755"),],
                            geneAnnot=getExp(data))
                            
 pair <- get.pair(data = data,
                  group.col = "definition", 
                  group1 = "Primary solid Tumor", 
                  group2 = "Solid Tissue Normal",
                  mode = "supervised",
                  diff.dir = "hypo",
                  nearGenes = nearGenes,
                  permu.size = 5,
                  raw.pvalue =  0.001,
                  Pe = 0.2,
                  dir.out="./",
                  permu.dir = "permu_test",
                  label = "hypo")

## End(Not run)
 pair <- data.frame(Probe = rep("cg00329272",3), 
                    GeneID = c("ENSG00000116213","ENSG00000130762","ENSG00000149527"),
                    Sides = c("R5","R2","L4"))                    
 getRandomPairs(pair)

Description

Auxiliary function for GetNearGenes This will get the closest genes (n=numFlankingGenes) for a target region (TRange) based on a genome of refenrece gene annotation (geneAnnot). If the transcript level annotation (tssAnnot) is provided the Distance will be updated to the distance to the nearest TSS.

Usage

getRegionNearGenes(
  TRange = NULL,
  numFlankingGenes = 20,
  geneAnnot = NULL,
  tssAnnot = NULL
)

Arguments

Value

A data frame of nearby genes and information: genes' IDs, genes' symbols,

Author(s)

Tiago C Silva (maintainer: [email protected])

Examples

geneAnnot <-  ELMER:::get.GRCh("hg38",as.granges = TRUE)
tssAnnot <-  getTSS(genome = "hg38")
probe <- GenomicRanges::GRanges(seqnames = c("chr1","chr2"), 
range=IRanges::IRanges(start = c(16058489,236417627), end= c(16058489,236417627)), 
name= c("cg18108049","cg17125141"))
names(probe) <- c("cg18108049","cg17125141")
NearbyGenes <- getRegionNearGenes(numFlankingGenes = 20,
                             geneAnnot = geneAnnot,
                             TRange = probe,
                             tssAnnot = tssAnnot)

Description

getRNAseq is a function to download RNAseq data for all samples of a certain cancer type from TCGA

Usage

getRNAseq(disease, basedir = "./Data", genome = "hg38")

Arguments

Value

Download all RNA seq level 3 data for the specified disease.

Description

getSymbol to report gene symbol from id

Usage

getSymbol(data, geneID)

Arguments

Value

The gene symbol for input genes.

Examples

data <- ELMER:::getdata("elmer.data.example")
getSymbol(data, geneID="ENSG00000143067")

Description

getTCGA is a function to download DNA methylation, RNA expression and clinic data for all samples of certain cancer type from TCGA website. And downloaded data will be transform to matrixes or data frame for further analysis.

Usage

getTCGA(disease, Meth=TRUE, RNA=TRUE, Clinic=TRUE, basedir="./Data", genome = "hg38")

Arguments

Value

Download DNA methylation (HM450K)/RNAseq(HiseqV2)/Clinic data for the specified disease from TCGA.

Examples

getTCGA(
   disease = "BRCA",
   Meth = FALSE, 
   RNA = FALSE, 
   Clinic = TRUE, 
   basedir = tempdir(), 
   genome = "hg38"
)

Description

This function gets the last version of human TF list from the UNiprot database

Usage

getTF()

Value

A data frame with the ensemble gene id.

Description

Saves a bed file with the unmethylated probes (+-250bp) regions that was infered to be bound by a given TF

Usage

getTFBindingSites(
  tf = NULL,
  results.dir = NULL,
  genome = "hg38",
  met.platform = "450K"
)

Arguments

Examples

## Not run: 
  getTFBindingSites("HNF1A",
                    results.dir = "analysis/unsupervised/group-Tumor_vs_Normal/hypo/")

## End(Not run)

Description

This function uses ELMER analysis results and summarizes the possible genes targets for each TF

Usage

getTFtargets(
  pairs,
  enriched.motif,
  TF.result,
  dmc.analysis,
  mae,
  save = TRUE,
  dir.out = "./",
  classification = "family",
  cores = 1,
  label = NULL
)

Arguments

Examples

pairs <- data.frame(Probe = c("cg26992600","cg26992800","cg26992900"),
                    Symbol = c("KEAP1","DSP","ATP86"))
enriched.motif <- list("FOXD3_HUMAN.H11MO.0.D"= c("cg26992800","cg26992900"))
TF.result <- data.frame(motif = c("FOXD3_HUMAN.H11MO.0.D"),
                        potential.TF.family = c("TP63;TP73"))
getTFtargets(pairs,enriched.motif,TF.result)

## Not run: 
getTFtargets("../LUAD_LUSC_analysis_hg38/hyper/getPair.hyper.pairs.significant.csv",
enriched.motif = "../LUAD_analysis_hg38/hyper/getMotif.hyper.enriched.motifs.rda",
TF.result = "../LUAD_analysis_hg38/hyper/getTF.hyper.significant.TFs.with.motif.summary.csv")

## End(Not run)

Description

getTSS to fetch GENCODE gene annotation (transcripts level) from Bioconductor package biomaRt If upstream and downstream are specified in TSS list, promoter regions of GENCODE gene will be generated.

Usage

getTSS(genome = "hg38", TSS = list(upstream = NULL, downstream = NULL))

Arguments

Value

GENCODE gene annotation if TSS is not specified. Coordinates of GENCODE gene promoter regions if TSS is specified.

Author(s)

Lijing Yao (maintainer: [email protected])

Examples

# get GENCODE gene annotation (transcripts level)
## Not run: 
    getTSS <- getTSS()
    getTSS <- getTSS(genome.build = "hg38", TSS=list(upstream=1000, downstream=1000))

## End(Not run)

Description

This heatmap will sort samples by their gene expression and show the DNA methylation levels of the paired probes to that gene. If no pairs are given, nearest probes will be selected. To use this function you MAE object (input data) will need all probes and not only the distal ones. This plot can be used to evaluate promoter, and intro, exons regions and closer distal probes of a gene to verify if their DNA methylation level is affecting the gene expression

Usage

heatmapGene(
  data,
  group.col,
  group1,
  group2,
  pairs,
  GeneSymbol,
  scatter.plot = FALSE,
  correlation.method = "pearson",
  correlation.table = FALSE,
  annotation.col = NULL,
  met.metadata = NULL,
  exp.metadata = NULL,
  dir.out = ".",
  filter.by.probe.annotation = TRUE,
  numFlankingGenes = 10,
  width = 10,
  height = 10,
  scatter.plot.width = 10,
  scatter.plot.height = 10,
  filename = NULL
)

Arguments

Value

A heatmap

Author(s)

Tiago Chedraoui Silva (tiagochst at gmail.com)

Examples

## Not run: 
  data <- ELMER:::getdata("elmer.data.example")
  group.col <- "subtype_Expression.Subtype"
  group1 <- "classical"
  group2 <- "secretory"
  pairs <- data.frame(ID = c("cg15924102","cg19403323", "cg22396959"),
                      GeneID = c("ENSG00000196878", "ENSG00000009790", "ENSG00000009790" ),
                      Symbol = c("TRAF3IP3","LAMB3","LAMB3"),
                      Side = c("R1","L1","R3"),
                      Distance = c(6017,168499,0),
                      stringsAsFactors = FALSE)
 heatmapGene(data = data, 
             group.col = group.col,
             group1 = group1, 
             group2 = group2,
             pairs = pairs, 
             GeneSymbol = "LAMB3",
             height = 5,
             annotation.col = c("ethnicity","vital_status"),
             filename = "heatmap.pdf")
 \dontrun{            
     heatmapGene(data = data, 
                 group.col = group.col,
                 group1 = group1, 
                 group2 = group2,
                 GeneSymbol = "ACP6",
                 annotation.col = c("ethnicity","vital_status"),
                 filename = "heatmap_closer_probes.pdf")
 }

## End(Not run)

Description

Heatmp plot of pairs gene and probes anti-correlated

Usage

heatmapPairs(
  data,
  group.col,
  group1,
  group2,
  pairs,
  subset = FALSE,
  cluster.within.groups = TRUE,
  plot.distNearestTSS = FALSE,
  annotation.col = NULL,
  met.metadata = NULL,
  exp.metadata = NULL,
  width = 10,
  height = 7,
  filename = NULL
)

Arguments

Value

A heatmap

Author(s)

Tiago Chedraoui Silva (tiagochst at gmail.com)

Examples

## Not run: 
  data <- ELMER:::getdata("elmer.data.example")
  group.col <- "subtype_Expression.Subtype"
  group1 <- "classical"
  group2 <- "secretory"
  pairs <- data.frame(Probe = c("cg15924102","cg19403323", "cg22396959"),
                      GeneID = c("ENSG00000196878", "ENSG00000009790", "ENSG00000009790" ),
                      Symbol = c("TRAF3IP3","LAMB3","LAMB3"),
                      Distance = c(6017,168499,0),
                      Raw.p = c(0.001,0.00001,0.001),
                      Pe = c(0.001,0.00001,0.001))
 heatmapPairs(
   data = data, group.col = group.col,
              group1 = group1, group2 = group2,
              annotation.col = c("ethnicity","vital_status","age_at_diagnosis"),
              pairs, filename = "heatmap.pdf",
              height = 4, width = 11
  )
  
## End(Not run)

Description

lable linear regression formula

Usage

lm_eqn(df, Dep, Exp)

Arguments

Value

A linear regression formula

Description

scatter.plot is a function to plot various scatter plots between gene expression and DNA methylation. When byPair is specified, scatter plot for individual probe-gene pairs will be generated. When byProbe is specified, scatter plots for one probes with nearby 20 gene pairs will be generated. When byTF is specified, scatter plot for TF expression and average DNA methylation at certain motif sites will be generated.

Usage

metBoxPlot(
  data,
  group.col,
  group1,
  group2,
  probe,
  min.samples = 5,
  minSubgroupFrac = 0.2,
  diff.dir = "hypo",
  legend.col = NULL,
  title = NULL,
  filename = NULL,
  save = TRUE
)

Arguments

Value

Box plot

Author(s)

Tiago Chedraoui Silva (tiagochst at gmail.com)

Examples

## Not run: 
  data <- ELMER:::getdata("elmer.data.example")
  group.col <- "subtype_Expression.Subtype"
  group1 <- "classical"
  group2 <- "secretory"
  metBoxPlot(data,
             group.col = group.col,
             group1 = group1,
             group2 = group2, 
             probe ="cg17898069",
             minSubgroupFrac = 0.2, 
             diff.dir = "hypo")

## End(Not run)

Description

motif.enrichment.plot to plot bar plots showing motif enrichment ORs and 95% confidence interval for ORs. Option motif.enrichment can be a data frame generated by get.enriched.motif or a path of XX.csv saved by the same function.

Usage

motif.enrichment.plot(motif.enrichment, 
                      significant = NULL, 
                      dir.out ="./", 
                      save = TRUE,
                      label = NULL,
                      title = NULL, 
                      width = 10, 
                      height = NULL,
                      summary = FALSE)

Arguments

Details

motif.enrichment If input data.frame object, it should contain "motif", "OR", "lowerOR", "upperOR" columns. motif specifies name of motif; OR specifies Odds Ratio, lowerOR specifies lower boundary of OR (95 upperOR specifies upper boundary of OR(95

significant A list used to select subset of motif.enrichment by the cutoff of OR, lowerOR, upperOR. significant=list(OR=1). More than one cutoff can be specified such as significant = list(OR=1, lowerOR=1,upperOR=4)

Value

A figure shows the enrichment level for selected motifs.

Author(s)

Lijing Yao (creator: [email protected])

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1.

Examples

motif.enrichment <- data.frame(motif = c("TP53","NR3C1","E2F1","EBF1","RFX5","ZNF143", "CTCF"),
                               OR = c(19.33,4.83,1, 4.18, 3.67,3.03,2.49),
                               lowerOR = c(10,3,1.09,1.9,1.5,1.9, 0.82),
                               upperOR = c(23,5,3,7,6,5,5),
                               stringsAsFactors = FALSE)
motif.enrichment.plot(motif.enrichment = motif.enrichment,
                      significant = list(OR = 3),
                      label = "hypo", save = FALSE)
motif.enrichment.plot(motif.enrichment = motif.enrichment,
                      significant = list(OR = 3),
                      label = "hypo", 
                      title = "OR for paired probes hypomethylated in Mutant vs WT",
                      save = FALSE)
motif.enrichment <- data.frame(motif = c("TP53","NR3C1","E2F1","EBF1","RFX5","ZNF143", "CTCF"),
                               OR = c(19.33,4.83,1, 4.18, 3.67,3.03,2.49),
                               lowerOR = c(10,3,1.09,1.9,1.5,1.5, 0.82),
                               upperOR = c(23,5,3,7,6,5,5),
                               NumOfProbes = c(23,5,3,7,6,5,5),
                               PercentageOfProbes = c(0.23,0.05,0.03,0.07,0.06,0.05,0.05),
                               stringsAsFactors=FALSE)
motif.enrichment.plot(motif.enrichment = motif.enrichment,
                      significant = list(OR = 3),
                      label = "hypo", save = FALSE)
motif.enrichment.plot(motif.enrichment = motif.enrichment,
                      significant = list(OR = 3),
                      label = "hypo", 
                      summary = TRUE,
                      title = "OR for paired probes hypomethylated in Mutant vs WT",
                      save = TRUE)

Description

This function has some filters to the DNA methylation data in each it selects probes to avoid correlations due to non-cancer contamination and for additional stringency.

• Filter 1: We usually call locus unmethylated when the methylation value < 0.3 and methylated when the methylation value > 0.3. Therefore Meth_B is the percentage of methylation value > K. Basically, this step will make sure we have at least a percentage of beta values lesser than K and n percentage of beta values greater K. For example, if percentage is 5%, the number of samples 100 and K = 0.3, this filter will select probes that we have at least 5 (5% of 100%) samples have beta values > 0.3 and at least 5 samples have beta values < 0.3. This filter is importante as true promoters and enhancers usually have a pretty low value (of course purity can screw that up). we often see lots of PMD probes across the genome with intermediate values like 0.4. Choosing a value of 0.3 will certainly give some false negatives, but not compared to the number of false positives we thought we might get without this filter.

Usage

preAssociationProbeFiltering(data, K = 0.3, percentage = 0.05)

Arguments

Value

An object with the same class, but with the probes removed.

References

Yao, Lijing, et al. "Inferring regulatory element landscapes and transcription factor networks from cancer methylomes." Genome biology 16.1 (2015): 1. Method section (Linking enhancer probes with methylation changes to target genes with expression changes).

Examples

random.probe <- runif(100, 0, 1)
 bias_l.probe <- runif(100, 0, 0.3)
 bias_g.probe <- runif(100, 0.3, 1)
 met <- rbind(random.probe,bias_l.probe,bias_g.probe)
 met <- preAssociationProbeFiltering(data = met,  K = 0.3, percentage = 0.05)
 met <- rbind(random.probe,random.probe,random.probe)
 met <- preAssociationProbeFiltering(met,  K = 0.3, percentage = 0.05)
 data <- ELMER:::getdata("elmer.data.example") # Get data from ELMER.data
 data <- preAssociationProbeFiltering(data,  K = 0.3, percentage = 0.05)
 
 cg24741609 <- runif(100, 0, 1)
 cg17468663 <- runif(100, 0, 0.3)
 cg14036402 <- runif(100, 0.3, 1)
 met <- rbind(cg24741609,cg14036402,cg17468663)
 colnames(met) <- paste("sample",1:100)
 exp <- met
 rownames(exp) <- c("ENSG00000141510","ENSG00000171862","ENSG00000171863")
 sample.info <- S4Vectors::DataFrame(primary = paste("sample",1:100),
                                     sample.type = rep(c("Normal", "Tumor"),50))
 rownames(sample.info) <- colnames(exp)
 mae <- createMAE(exp = exp, met = met, colData = sample.info, genome = "hg38") 
 mae <- preAssociationProbeFiltering(mae,  K = 0.3, percentage = 0.05)

Description

promoterMeth is a function to calculate associations of gene expression with DNA methylation at promoter regions.

Usage

promoterMeth(data, sig.pvalue = 0.01, minSubgroupFrac = 0.4,
             upstream = 200,  downstream = 2000, save = TRUE, cores = 1)

Arguments

Details

promoterMeth

Value

A data frame contains genes whose expression significantly anti-correlated with promoter methylation.

Examples

## Not run: 
  data(elmer.data.example.promoter)
  Gene.promoter <- promoterMeth(mae.promoter)

## End(Not run)

Description

Build HTML report

Usage

render_report(
  title = "Report",
  mae.file,
  group.col,
  group1,
  group2,
  direction,
  dir.out,
  genome = "hg38",
  mode = "supervised",
  minSubgroupFrac = 0.2,
  minMetdiff = 0.3,
  metfdr = 0.01,
  permu = 10000,
  rawpval = 0.01,
  pe = 0.01,
  nprobes = 10,
  lower.OR = 1.1,
  out_file = file.path(getwd(), "report.html"),
  funcivar = FALSE
)

Arguments

Examples

## Not run: 
render_report(
 group.col = "TN",
 group1 = "Tumor",
 group2 = "Normal",
 dir.out = "~/paper_elmer/Result/BRCA/TN_Tumor_vs_Normal/hypo/",
 direction = "hypo",
 mae.file = "~/paper_elmer/Result/BRCA/BRCA_mae_hg38.rda"
)

## End(Not run)

Description

scatter

Usage

scatter(
  meth,
  exp,
  legend.title = "Legend",
  category = NULL,
  xlab = NULL,
  ylab = NULL,
  ylim = NULL,
  dots.size = 0.9,
  title = NULL,
  correlation = FALSE,
  correlation.text.size = 3,
  color.value = NULL,
  lm_line = FALSE
)

Arguments

Value

A ggplot figure object

Description

scatter.plot is a function to plot various scatter plots between gene expression and DNA methylation. When byPair is specified, scatter plot for individual probe-gene pairs will be generated. When byProbe is specified, scatter plots for one probes with nearby 20 gene pairs will be generated. When byTF is specified, scatter plot for TF expression and average DNA methylation at certain motif sites will be generated.

Usage

scatter.plot(data, 
             byPair = list(probe = c(), gene = c()),
             byProbe = list(probe = c(), numFlankingGenes = 20), 
             byTF = list(TF = c(), probe = c()),
             category = NULL,
             ylim = NULL,
             dots.size = 0.9,
             correlation = FALSE,
             width = 7,
             height = 6,
             dir.out = "./",
             save = TRUE, ...)

Arguments

Details

byPair The output will be scatter plot for individual pairs.

byProbe The output will be scatter plot for the probe and nearby genes.

byTF The output will be scatter plot for the TFs and the average DNA methylation at the probes set specified in byTF list.

Value

Scatter plots.

Author(s)

Lijing Yao (maintainer: [email protected])

Examples

data <- ELMER:::getdata("elmer.data.example")
scatter.plot(data,
            byProbe=list(probe=c("cg19403323"),numFlankingGenes=20), 
            category="definition", save=FALSE)
scatter.plot(data,byProbe=list(probe=c("cg19403323"),numFlankingGenes=20), 
            category="definition", save=TRUE) ## save to pdf
# b. generate one probe-gene pair
scatter.plot(data,byPair=list(probe=c("cg19403323"),gene=c("ENSG00000143322")),
             category="definition", save=FALSE,lm_line=TRUE)

Description

schematic.plot is a function to plot schematic plots showing the locations of genes and probes.

Usage

schematic.plot(data,
               group.col = NULL,
               group1 = NULL,
               group2 = NULL,
               pair,
               byProbe,
               byGeneID,
               byCoordinate=list(chr=c(), start=c(), end=c()),
               statehub.tracks,
               dir.out="./",
               save=TRUE,...)

Arguments

Details

byProbes: When a vector of probes' name are provided, function will produce schematic plots for each individual probes. The schematic plot contains probe, nearby 20 (or the number of gene user specified.) genes and the significantly linked gene to the probe.

byGene: When a vector of gene ID are provided, function will produce schematic plots for each individual genes. The schematic plot contains the gene and all the significantly linked probes.

byCoordinate: When a genomic coordinate is provided, function will produce a schematic plot for this coordinate. The schematic plot contains all genes and significantly linked probes in the range and the significant links.

Examples

data <- ELMER:::getdata("elmer.data.example")
pair <- data.frame(Probe = c("cg19403323","cg19403323", "cg26403223"),
                   GeneID = c("ENSG00000196878", "ENSG00000009790", "ENSG00000009790" ),
                   Symbol = c("TRAF3IP3","LAMB3","LAMB3"),
                   Raw.p =c(0.001,0.00001,0.001),
                   Pe = c(0.001,0.00001,0.001))
schematic.plot(data,
               group.col = "definition",
               group1 = "Primary solid Tumor",
               group2 = "Solid Tissue Normal",
               pair = pair,
               byProbe = "cg19403323")
schematic.plot(data,
               group.col = "definition",
               group1 = "Primary solid Tumor",
               group2 = "Solid Tissue Normal",
               pair = pair,
               byGeneID = "ENSG00000009790")

schematic.plot(data,
               group.col = "definition",
               group1 = "Primary solid Tumor",
               group2 = "Solid Tissue Normal",
               pair = pair,
               byCoordinate = list(chr="chr1", start = 209000000, end = 209960000))
## Not run: 
   schematic.plot(data,
                  group.col = "definition",
                  group1 = "Primary solid Tumor",
                  group2 = "Solid Tissue Normal",
                  pair = pair,
                  byProbe = "cg19403323",
                  statehub.tracks = "hg38/ENCODE/mcf-7.16mark.segmentation.bed")

## End(Not run)

Description

Stat.diff.meth

Usage

Stat.diff.meth(
  meth,
  groups,
  group1,
  group2,
  test = t.test,
  min.samples = 5,
  percentage = 0.2,
  Top.m = NULL
)

Arguments

Value

Statistic test results to identify differentially methylated probes.

Description

U test (non parameter test) for permutation. This is one probe vs nearby gene which is good for computing each probes for nearby genes.

Usage

Stat.nonpara(
  Probe,
  NearGenes,
  Top = NULL,
  correlation = "negative",
  unmethy = NULL,
  methy = NULL,
  Meths = Meths,
  Exps = Exps
)

Arguments

Value

U test results

Description

Stat.nonpara.permu

Usage

Stat.nonpara.permu(
  Probe,
  Gene,
  Top = 0.2,
  correlation = "negative",
  unmethy = NULL,
  methy = NULL,
  Meths = Meths,
  Exps = Exps
)

Arguments

Value

U test results

Description

This function uses ELMER analysis results and summarizes the MR TF identified in each analysis

Usage

summarizeTF(files = NULL, path = NULL, classification = "family", top = FALSE)

Arguments

Description

ELMER analysis pipeline for TCGA data. This pipeline combine every steps of ELMER analyses: get.feature.probe, get.diff.meth, get.pair, get.permu, get.enriched.motif and get.TFs. Every steps' results are saved.

Usage

TCGA.pipe(
  disease,
  genome = "hg38",
  analysis = "all",
  wd = getwd(),
  cores = 1,
  mode = "unsupervised",
  Data = NULL,
  diff.dir = "hypo",
  genes = NULL,
  mutant_variant_classification = c("Frame_Shift_Del", "Frame_Shift_Ins",
    "Missense_Mutation", "Nonsense_Mutation", "Splice_Site", "In_Frame_Del",
    "In_Frame_Ins", "Translation_Start_Site", "Nonstop_Mutation"),
  group.col = "TN",
  group1 = "Tumor",
  group2 = "Normal",
  ...
)

Arguments

Value

Different analysis results.

Examples

data <- ELMER:::getdata("elmer.data.example")
  TCGA.pipe(
    disease = "LUSC",
    data = data,
    analysis = c("diffMeth","pair", "motif","TF.search"), 
    mode = "supervised",
    group.col = "definition",
    group1 = "Primary solid Tumor", 
    group2 = "Solid Tissue Normal",
    diff.dir = c("hypo"),
    dir.out = "pipe",
    sig.dif = 0.0001,
    pvalue = 1.0,
    min.incidence = 0,
    lower.OR = 0.0
  )
## Not run: 
  distal.probe <- TCGA.pipe(disease = "LUSC", analysis="distal.enhancer", wd="~/")
  TCGA.pipe(disease = "LUSC",analysis = "all", genome = "hg19", cores = 1, permu.size=300, Pe=0.01)
  projects <- TCGAbiolinks:::getGDCprojects()$project_id
  projects <- gsub("TCGA-","",projects[grepl('^TCGA',projects,perl=TRUE)])
  for(proj in projects) TCGA.pipe(disease = proj,analysis = "download")
  plyr::alply(sort(projects),1,function(proj) {
       tryCatch({
         print(proj);
         TCGA.pipe(disease = proj,analysis = c("createMAE"))})
       }, .progress = "text")
  plyr::alply(sort(projects),1,function(proj) {
    tryCatch({
      print(proj);
      TCGA.pipe(disease = proj,
                 analysis = c("diffMeth","pair", "motif","TF.search"))})
  }, .progress = "text")

  # Evaluation mutation
  TCGA.pipe(disease = "LUSC",analysis = "createMAE",gene = "NFE2L2")
  TCGA.pipe(
     disease = "LUSC",analysis = c("diffMeth","pair", "motif","TF.search"), 
     mode = "supervised",
     group.col = "NFE2L2", group1 = "Mutant", group2 = "WT",
     diff.dir = c("hypo"),
     dir.out = "LUSC_NFE2L2_MutvsWT"
   )

## End(Not run)

Description

TF.rank.plot is a function to plot the scores (-log10(P value)) which assess the correlation between TF expression and average DNA methylation at motif sites. The the motif relevant TF and top3 TFs will be labeled in a different color.

Usage

TF.rank.plot(
  motif.pvalue,
  motif,
  title = NULL,
  TF.label = NULL,
  dir.out = "./",
  save = TRUE,
  cores = 1
)

Arguments

Value

A plot shows the score (-log(P value)) of association between TF expression and DNA methylation at sites of a certain motif.

Author(s)

Lijing Yao (maintainer: [email protected])

Examples

library(ELMER)
data <- tryCatch(ELMER:::getdata("elmer.data.example"), error = function(e) {
  message(e)
  data(elmer.data.example, envir = environment())
})
enriched.motif <- list("P53_HUMAN.H11MO.0.A"= c("cg00329272", "cg10097755", "cg08928189",
                                 "cg17153775", "cg21156590", "cg19749688", "cg12590404",
                                 "cg24517858", "cg00329272", "cg09010107", "cg15386853",
                                 "cg10097755", "cg09247779", "cg09181054"))
TF <- get.TFs(data, 
              enriched.motif, 
              group.col = "definition",
              group1 = "Primary solid Tumor", 
              group2 = "Solid Tissue Normal",
              TFs = data.frame(
                     external_gene_name=c("TP53","TP63","TP73"),
                     ensembl_gene_id= c("ENSG00000141510",
                                        "ENSG00000073282",
                                        "ENSG00000078900"),
                     stringsAsFactors = FALSE),
             label="hypo")
TF.meth.cor <- get(load("getTF.hypo.TFs.with.motif.pvalue.rda"))            
TF.rank.plot(motif.pvalue=TF.meth.cor,  
            motif="P53_HUMAN.H11MO.0.A", 
            TF.label=createMotifRelevantTfs("subfamily")["P53_HUMAN.H11MO.0.A"],
            save=TRUE)
TF.rank.plot(motif.pvalue=TF.meth.cor,  
            motif="P53_HUMAN.H11MO.0.A", 
            save=TRUE)
# Same as above
TF.rank.plot(motif.pvalue=TF.meth.cor,  
            motif="P53_HUMAN.H11MO.0.A", 
            dir.out = "TFplots",
            TF.label=createMotifRelevantTfs("family")["P53_HUMAN.H11MO.0.A"],
            save=TRUE)

Description

This function will create a survival plot for the samples with higher, midium, low expression of a given transcription factor. By defau;t samples with higher expression are the top 30

Usage

TFsurvival.plot(data, TF, xlim = NULL, percentage = 0.3, save = TRUE)

Arguments