How to proceed with SWAN normalized 450k Methyl array data

Question

[a.james]

Hello All, I have a SWAN normalized 450K methylation array data. I would like to know how can I proceed from this point towards differentially methylated regions using available packages like Minifi, IlluminaHumanMethylation450kanno.ilmn12.hg19 , etc.

I know all these packages take raw data. But with Normalized data how can i go with analysis

[igor]

It's probably SWAN (not SWAM) normalized.

Not sure how to proceed, but many minfi functions will actually take a matrix. Check the source to confirm. You can probably plug in your normalized matrix at an appropriate stage and proceed normally from there.

[EagleEye]

Which file format do you have??

[a.james]

The format is txt format

[EagleEye]

Try these tools from Bioconductor, (minfi seems to be better option in your case)

A) minfi

Detailed documentation: https://www.bioconductor.org/help/course-materials/2015/BioC2015/methylation450k.html#identifying-dmrs-and-dmps

B) wateRmelon

https://www.bioconductor.org/packages/devel/bioc/html/wateRmelon.html

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

---

5 Suggested analysis workflow

5.1 Load and tidy data

You can use a variety of methods to load your data, either from GenomeStudio final report text files or from iDAT files. methylumi and IMA can read text files, we recommend methylumi because the exprmethy450 object only stores betas and not raw intensities. methylumi and minfi can both read iDAT files, and produce objects that can be used by our functions. Neither contains the full annotation that comes inside the final report text file. If you use the GenomeStudio file we recommend saving the unnormalized, uncorrected version of the data. We also recommend keeping the barcode names (SentrixID RnnCnn) as the column headers or in a separate dataframe.

library(methylumi)

melon <- methyLumiR('finalreport.txt')

---

C) ChAMP (I guess this will work only with .iDAT as input)

https://www.bioconductor.org/packages/release/bioc/html/ChAMP.html

https://www.bioconductor.org/packages/release/bioc/vignettes/ChAMP/inst/doc/ChAMP.pdf

[a.james]

Thank for all suggestions, could you please explain how can I proceed with Minifi like this. I dont see any where a description where I can proceed with SWAN normalized file . Also methyLumiR function from methylumi package is not accepting the input normalized.

[EagleEye]

Please paste few lines of your text file (SWAN normalized file).

[a.james]

Here is how it looks,

Sample ID   cg00050873  cg00212031  cg00213748  cg00214611
3999337068_R01C01_Red   0.499329    0.445607    0.260135    0.41489 
3999337068_R01C02_Red   0.481271    0.146888    0.261434    0.362547  
3999337068_R02C01_Red   0.893196    0.0490534   0.919875    0.0466149