MIDAS supports the
loading of several file formats. For a complete description refer to the manual. Here's a
brief overview of supported file types.
This format is a tab
delimited text file format which allows a header for identification of columns
and comments are preceded by the pound (#) sign. The first nine columns are a
unique identifier, slide row, slide column, meta row,
meta column, sub row, sub column, intensity 1, intensity 2. The unique
identifier is used to match up corresponding annotation which is supplied from
annotation files which contain spot annotation for each element on the slide.
The annotation files, given a .ann
extension are also tab delimited and have entries corresponding to the unique
identifiers in the mev file. One or more annotation
files may be loaded to provide several types of annotation.
The Stanford format
file encapsulates several hybridizations in one file. The first several columns
are used for annotation and following the annotation the intensity data
corresponds to one column of log2ratio(cy5/cy3)
for each hybridization. The format includes a header to identify the labels for
the annotation as well as the labels for the individual hybridizations.
This is similar to the mev format but differs in that it does not permit a header
row or comment rows and there is no unique ID column. Each tav
file contains annotation following the columns for intensity and other spot
specific information.
The loading of tav files depends on having selected a
'preference' file which is used to identify the type and number of annotation
fields in the tav files.
Affymetrix arrays use a single dye system. The file
description is detailed on the Affymetrix web site
but includes a Affy probe
identifier, an intensity value, a detection flag to indicate whether or not a
spot signal is present or absent, and a spot description.
Loading requires
several basic steps that vary slightly depending on the file format chosen.
1.) Select a file format.
2.) Select a directory using the file navigation tree in the left window until
the selected files are reached.
3.) Select the data file or files.
For mev
format you will select several data files in the available files list and use
the add button to move them into the selected file list box. The same process
is used to select one or more annotation files. Once these files are selected
hit the OK button to load the files.
For
Stanford format you will select a path to the data file and once a file is
selected a table will display the first few rows of the file. Select the first
log ratio of the first, left most, hybridization in the table. This will
identify the upper left limit of the log ratio data. Columns to the left
contain annotation while columns above contain header information. Once the
proper cell has been selected the annotation field names should appear in a
window near the bottom of the form. Select OK to load.
For tav
Format there are several means to load the files. The most common is to use a
preference file to indicate annotation fields in the file. See the manual for a
full description of the tav format and the use of
preference files. Navigate through the tree to select the tav
files then use the file browser button to select an associated preference file.
There are four modes to
loading Affymetrix files.
1.) The absolute selection simply loads the single intensity value into one
channel and sets the other channel to 1.0. This will require the user to scale
the expression image, set ratio scale, to increase the dynamic range of the
image to compensate for the absolute intensities.
2.) The mean and median modes are related. For each spot the intensity is
averaged or the median is taken across the hybridizations (files). This average
value is used as a reference to scale the intensities corresponding the that spot. Each spot gets assigned one
intensity to the spots true intensity and the other intensity is set to
the mean or median intensity. The calculated log2(r) is taken where
r = (spot intensity)/(ave.
intensity)
3.) The reference loading mode allows the user to select a file to represent a
reference against which the other files are compared.
log2(spot intensity/ref. intensity)
Please refer to the manual for more details and information regarding file
formats and loading data.