Setup: Location of annotation data files

CDF files

A chip definition file (CDF) contains information on which probes belong to what probeset, the (x,y) location of each probe, which the middle nucleotides in the target and the probe are (from which PM/MM status is inferred), and so on. Note that there might be multiple different CDFs for the same chip type.

Aroma.affymetrix searches for CDF files in the annotationData/ directory of the current working directory. Place the CDF for chip type <chipType> in a directory of format:

  annotationData/chipTypes/<chip type>/

All other CDF files with filename format <chip type>,<tags>.cdf should also go into this directory. For instance, so called monocell CDFs named <chip type>,monocell.cdf, should be placed in this directory (Footnote: Monocell CDF files are created automatically if missing and will/should by default be written to the correct directory.). For example,

annotationData/chipTypes/Mapping50K_Hind240/Mapping50K_Hind240.cdf
annotationData/chipTypes/Mapping50K_Hind240/Mapping50K_Hind240,monocell.cdf

Moreover, the name of a CDF file should always be identical to the name of the chip type followed by filename extension "cdf" (or "CDF"), e.g. the CDF for chip type "Mapping250K_Nsp" the filename should be "Mapping250K_Nsp.cdf". Indeed, since the CDF itself does not contain any information about the chip type, the chip type is inferred from the filename.

Example

cdf <- AffymetrixCdfFile$byChipType("Mapping50K_Hind240")
print(cdf)

AffymetrixCdfFile:
Path: annotationData/chipTypes/Mapping50K_Hind240
Filename: Mapping50K_Hind240.CDF
Filesize: 53.43MB
File format: v4 (binary; XDA)
Chip type: Mapping50K_Hind240
Dimension: 1600x1600
Number of cells: 2560000
Number of units: 57299
Cells per unit: 44.68
Number of QC units: 9
RAM: 0.00MB

Affymetrix NetAffx CSV files

Affymetrix NetAffx CSV files are comma-separated and tabular ASCII files containing data exported from the NetAffx data base. There is no well-defined filename convention of these. Place them under the corresponding chip type directory, i.e.

annotationData/chipTypes/<chipType>/

For further separation of files, they, like any other annotation file, may be place un in subdirectories of the above, e.g.

  annotationData/chipTypes/<chipType>/NetAffx/

For example:

annotationData/chipTypes/Mapping50K_Hind240/NetAffx/Mapping50K_Hind240_annot.csv

Example

csv <- AffymetrixNetAffxCsvFile$byChipType("Mapping50K_Hind240", pattern="_annot[.]csv$")

## AffymetrixNetAffxCsvFile:
## Name: Mapping50K_Hind240_annot
## Tags:
## Pathname:
## annotationData/chipTypes/Mapping50K_Hind240/NetAffx/Mapping50K_Hind240_annot.csv
## File size: 59.67MB
## RAM: 0.00MB
## Columns [26]: 'Probe Set ID', 'Affy SNP ID', 'dbSNP RS ID',
## 'Chromosome', 'Genome Version', 'DB SNP Version', 'Physical Position',
## 'Strand', 'ChrX pseudo-autosomal region', 'Cytoband', 'Flank', 'Allele
## A', 'Allele B', 'Associated Gene', 'Genetic Map', 'Microsatellite',
## 'Fragment Length Start Stop', 'Freq Asian', 'FreqAfAm', 'Freq Cauc',
## 'Het Asian', 'Het AfAm', 'Het Cauc', 'Num chrm Asian', 'Num chrm AfAm',
## 'Num chrm Cauc'

Affymetrix probe-tab files

So called Affymetrix probe-tab files are annotation files containing information about probe sequences etc. They are available from the Affymetrix web site. These are needed in order to do, for instance, GCRMA background correction. Unfortunately, there is not formal format for the names of these files, but they are typically starting with something that looks like the chip type followed by the suffix '_probe_tab' without a filename extension. As for all annotation files, place them under the corresponding chip type directory, e.g.

annotationData/chipTypes/Mapping50K_Hind240/Mapping50K_Hind_probe_tab

DChip annotation files

Aroma.affymetrix recognizes several of the dChip chip-type specific annotation file formats, e.g. SNP information files and genome information files. These are available for several chip types from http://www.dchip.org/. Make sure to put these under the corresponding directory

  annotationData/chipTypes/<chip type>/

The dChip files do not have to be renamed. In order to organize annotation files further, it is possible to put the files in subdirectories of the above, e.g.

annotationData/chipTypes/Mapping50K_Hind240/dChip/50k hind genome info AfAm june 05 hg17.xls
annotationData/chipTypes/Mapping50K_Hind240/dChip/Mapping100K_Hind snp info.txt

Example

si <- DChipSnpInformation$byChipType("Mapping50K_Hind240")
print(si)

## DChipSnpInformation:  
## Name: Mapping100K_Hind snp info  
## Tags:  
## Pathname:
## annotationData/chipTypes/Mapping50K_Hind240/dChip/Mapping100K_Hind snp
## info.txt  
## File size: 6.76MB  
## RAM: 0.00MB  
## Chip type: Mapping50K_Hind240

si <- DChipGenomeInformation$byChipType("Mapping50K_Hind240")
print(si)

## DChipGenomeInformation:
## Name: 50k hind genome info AfAm june 05 hg17
## Tags:
## Pathname: annotationData/chipTypes/Mapping50K_Hind240/dChip/50k hind
## genome info AfAm june 05 hg17.xls
## File size: 2.47MB
## RAM: 0.00MB
## Chip type: Mapping50K_Hind240