Question

Methylation profile through chromosome

0

Entering edit mode

Adam • 0

@adam-10025

Last seen 12 months ago

Poland

Hello,

I'm looking for some R package or even software to make methylation profile of chromosomes. I found two very nice packages for R:

ChromHeatMap (plots made by this are what I'm looking for) quantsmooth

Unfortunetaly my data isn't ExpressionSet or any object like this, so I don't know how to put it in. My data presentation(simple table in Excel with procentage methylation values):

TargetID    SYMBOL  CHR Position    cell_line_1 cell_line_2 cell_line_3
tag1    BRCA1   1   20771708    0,541441    0,4672899   0,3527385
tag2    BRCA1   1   20770604    0,3824306   0,59445525  0,408887625
tag3    GADPH   2   48719844    0,04331239  0,07833141  0,060533235
tag4    RASA1   3   100590962   0,0758754   0,12259825  0,11587337
tag5    RASA1   3   100589629   0,03625321  0,06974737  0,06018182
tag6    RASA1   3   100572000   0,05042755  0,090109245 0,10497023
tag7    ENG     5   98420934    0,3212603   0,16265565  0,137117475
tag8    ENG     5   98422056    0,1903218   0,19699885  0,491916625
tag9    CD5     4   114018313   0,1364216   0,127996    0,064798523
tag10   CD5     4   114018975   0,0766691   0,094443825 0,093477908

What am I looking for? Picture on 2nd page:

https://www.bioconductor.org/packages/devel/bioc/vignettes/ChromHeatMap/inst/doc/ChromHeatMap.pdf

R heatmap • 1.5k views

ADD COMMENT • link 8.2 years ago Adam • 0

1

Entering edit mode

you can convert the table to an expressionSet if this is all what needed

https://www.bioconductor.org/packages/devel/bioc/vignettes/Biobase/inst/doc/ExpressionSetIntroduction.pdf

ADD REPLY • link 8.2 years ago Assa Yeroslaviz ★ 1.5k

0

Entering edit mode

Yes indeed this is what I'm looking for! However, have you done this before? Of course I'll try to convert it myself but let this topic to be open for possible further questions. Thank you :)

ADD REPLY • link 8.2 years ago Adam • 0

0

Entering edit mode

Yes. I have done it myself before. it is quite simple and the vignette is clear to follow.

ADD REPLY • link 8.2 years ago Assa Yeroslaviz ★ 1.5k

score 0 · Answer 1 · 2016-08-23

Ok, with Assa Yeroslavizhelp I've done what I need, nevertheless I'd like to share my solution, maybe it'll be useful for somebody.

There are a few simple steps from simple '\t' separated file to nice visualization.

1. Install and load all stuff which will be needed. Data from my microarray was from the illumina so I installed suitable package.

source("http://bioconductor.org/biocLite.R")
biocLite("IlluminaHumanMethylation27k.db")
biocLite(c("Biobase"))

2. Load you file into R.

exprs <- as.matrix(read.table("your_path/your_data.txt", header=TRUE, sep="\t",row.names=1,as.is=TRUE))

3. Convert your file stored in variable into the ExpressionSet. ExpSet contains a few different types of data, experiment info etc, but expression or methylation profile will be enough to further analysis (in this case).

minimalSet <- ExpressionSet(assayData=exprs)

4. Put some annotation (chromosomes, features locations) into your data - it can provide some difficulties, you have to know from what type of microarray your data comes from.

chrdata<-makeChrStrandData(exprs(minimalSet), lib="IlluminaHumanMethylation27k.db")

5. Make a nice heatmap!

plotChrMap(chrdata, 1, strands="both")

Fell free to check other features of ChromHeatMap package.

All the best!

Adam