Question

how to use idat file for analyses in bioconductor

0

Entering edit mode

kritikamish99 ▴ 10

@kritikamish99-9648

Last seen 6.3 years ago

India

Hello all,

I want to know how to use idat file for analyses in bioconductor.

I read certain packages manual (limma and lumi). There it is said for analyses i need .txt file which is generated from genome studio.

for my sample i run the genome studio too. How i will get text files for analyses using lumi or limma

thanks

microarray illumina idat • 11k views

ADD COMMENT • link updated 8.8 years ago by Gordon Smyth 51k • written 8.8 years ago by kritikamish99 ▴ 10

Gordon Smyth · Answer 1 · 2016-02-16

2

Entering edit mode

Gordon Smyth 51k

@gordon-smyth

Last seen 8 hours ago

WEHI, Melbourne, Australia

The function read.idat() is now provided in the limma package to read idat files. So you no longer need to export a text file from Genome Studio.

The function read.idat() is a new addition to the limma package and is not yet described in the limma User's Guide, which describes only how to the read exported text files from Genome Studio. The new function reads Illumina's encrypted binary IDAT files directly using the capabilities of the illuminio package.

To read and normalize the data:

x <- read.idat(idatfiles, bgxfile)
y <- neqc(data)

where 'idatfiles' and 'bgxfile' are the names of the IDAT and BGX files. Then 'y' can be input directly to lmFit(). You can follow the case study in Section 17.3 of the limma User's Guide, except that read.idat() replaces read.ilmn().

ADD COMMENT • link 8.8 years ago Gordon Smyth 51k

0

Entering edit mode

ok thanks ......

ADD REPLY • link 8.8 years ago kritikamish99 ▴ 10

0

Entering edit mode

what i understood after reading manual of lumi for illumina microarray data the file should be txt for example summary probe profile.txt and summary control probe profile.txt i m asking where i will get summary probe profile.txt and summary control probe profile.txt files

i am asking for probe.txt file i searched on net some where it said that it can be obtain from genome studio or bead studio . My question is how get this file from genome studio.

In limma it said that :- Illumina images are scanned by Bead-Scan software, and Illumina's BeadStudio or Genome Studio software can be used to export probe summary profile. The probe summary profile are tab-delimited files containing the intensity data. (section 4.6 Reading Illumina BeadChip Data)

ADD REPLY • link updated 8.8 years ago by Gordon Smyth 51k • written 8.8 years ago by kritikamish99 ▴ 10

1

Entering edit mode

You presumably have IDAT files that were produced by your BeadArray scanner. These contain the summarised intensities for all the probe types on the array. Unfortunately, these files are encrypted and for a long time we were not able to read the files directly. The only way to access the data was to use Illumina's BeadStudio software to read the files, then get BeadStudio to output to a text file, which could then be read using R. I believe this is the model lumi still uses. If you want to use lumi, you will need to get access to your copy of GenomeStudio. It will have been supplied with instructions, but you can find them on Illumina's website here, or ask Illumina's support team for advice on how to use their software.

Alternatively, you can use limma & illuminaio. The encryption problem was solved a few years ago, and illuminaio can read IDAT files from expression arrays. However illuminaio just provides an interface to read single files. The function read.idat() in limma uses it to read a whole bunch and stick them in a sensible structure for further analysis. If you can't or don't want to use BeadStudio, then you can ignore the probe summary text files and use limma.

ADD REPLY • link updated 8.8 years ago by Gordon Smyth 51k • written 8.8 years ago by Mike Smith ★ 6.6k

0

Entering edit mode

thanks mike

ADD REPLY • link 8.8 years ago kritikamish99 ▴ 10

0

Entering edit mode

But this is giving me error:-

> idatfiles = dir(path="/dummy_data/Image Data/9967669861/" , pattern="idat")
> bgxfile = dir(path = "/dummy_data/Image Data/Manifest/" ,pattern = ".bgx")

> bgxfile
[1] "humanht-12_v4_0_r2_15002873_b.bgx"

>data = read.idat(idatfiles, bgxfile)

error :-

Reading manifest file humanht-12_v4_0_r2_15002873_b.bgx ...
Error in open.connection(file, "rt") : cannot open the connection
In addition: Warning message:
In open.connection(file, "rt") :
cannot open compressed file 'humanht-12_v4_0_r2_15002873_b.bgx', probable reason 'No such file or directory'

i cross check my manifest is unzipped. and it is in same path which i gave for bgxfile vector

ADD REPLY • link 8.8 years ago kritikamish99 ▴ 10

0

Entering edit mode

I suspect this is because by default dir() gives you the names of the files it finds, but not the complete path to them. If your working directory isn't where the files are, it won't be able to find them. You can do one of two things.

Change the working directory to the same place as the files e.g.
```
setwd("/dummy_data/Image Data/Manifest/")
```

Get dir() to give you the full path e.g.

dir(path = "/dummy_data/Image Data/Manifest/", pattern = ".bgx", full.names = TRUE)

It looks like your idat files are in a separate folder to the bgx file, so it's probably easiest to use the second of these.

ADD REPLY • link 8.8 years ago Mike Smith ★ 6.6k

1

Entering edit mode

yes they are in separate folder. i tried the second command it was not working same error so i put idat and manifest file in one folder so it worked

thanks mike for helping

ADD REPLY • link 8.8 years ago kritikamish99 ▴ 10