File format descriptions
Below you can find a short description of the file formats used for and produced by HALO. Note that <TAB> represents the tab character.
Overview
Input files
Data from microarray or RNA-seq experiments
The data from your microarray or RNA-seq experiments can be loaded into HALO as a single '.txt'-file of the
following format:
##############################################################################
#########Information that is no data has to be masked with a sharp############
##############################################################################
####First line after the commentary has to contain labels for data columns####
probeset_id<TAB>label1<TAB>label2<TAB>label3<TAB>....<TAB>labeln
After the header the values are given, tab separated like in the following example:
123456<TAB>140.014<TAB>30.12<TAB>213.5<TAB>...<TAB>54.12
There can be only one probeset id per line.
Multiple fasta files
Gene sequences have to be provided in a multiple fasta file, where one column in the fasta header describes the
gene name. For an example of such a file see below.
> gene name|attribute|description|another attribute
ATCGTCAGAGATTATTACAGATACATTGAGATGAGTACGATGATAATGACATG
The sequence should be provided without any newline characters. It does not matter which column of the header
contains the gene name, but it has to be the same column for every entry.
Attribute files
You can load additional attributes corresponding to your data. These files can contain one or more attributes,
but they have to contain one column with probeset ids for mapping the attributes with the data. An example
attribute file looks like this:
############################################################################
##########################Commentary can be given###########################
############################################################################
probeset_id<TAB>attribute1<TAB>attribute2<TAB>....<TAB>attributen
123456<TAB>gene1<TAB>A<TAB>...<TAB>1.23
The order of the columns is not important, but you have to define the label of the probeset_ids previously if
it is not the same as in the original data.
Output files
Filtered data
The filtered data is in the same format as the input data. There is a header produced and the columns are
put to the output file in the order probeset id, total RNA replicates, newly transcribed RNA replicates, pre-existing RNA replicates,
attributes.
Half-life/Ratio values
When you have calculated the half-lives the program offers you three possible ways of saving: You can either
save only the half-lives, only the ratios on which the half-lives are based, or both.
Half-lives and Ratios only:
spotid<TAB>half-life (method1)<TAB>half-life (method2)
123456<TAB>1.23<TAB>2.34
Both half-lives and Ratios:
Here you are presented with half-life values first and ratio values second, separated by a <TAB>.
spotid<TAB>half-life (method1)<TAB>half-life (method2)<TAB>method1 (ratios)<TAB>method2 (ratios)
12345<TAB>1.23<TAB>2.34<TAB>3.45<TAB>4.56
Half-lives with attributes
You can also save the results of a single half-life calculation, combined with one or more attributes. The file
format of these output files looks like this:
spotid<TAB>half-life<TAB>attr_1<TAB>...<TAB>attr_n
12345<TAB>1.23<TAB>gene_1<TAB>...<TAB>A
Probeset quality scores
You can calculate the probeset quality score for each probeset. The output is structured
spotid<TAB>value, e.g.:
123456<TAB>0.123
Normalization plotting file
HALO allows you not only to plot your normalization results, but also save the plotting information in a
file for plotting with a program of your choice. Normalization plot display two types of information:
the data points in a scatter plot and the linear regression line. Since these two types of values seldom
have overlapping x-values, the plotting file is NOT structured in the usual x-value<TAB>y1-value
<TAB>y2-value, but instead can be separated into two parts. In the first part you will find
the data for the scatter-plot, structured as x-value<TAB>y-value, and after that
the data values for the regression line are displayed, structured in the same way. You can easily distinguish
between the two parts of the files through a header line inserted before each part. The file will look
something like this:
#Data points:x<TAB>y
1.2155489028028954<TAB>1.0138011284375397
2.3153839627405824<TAB>0.9239562391183962
#Regression line:x<TAB>y
0.0<TAB>0.07405405303601431
0.0010<TAB>0.07483151974610898
0.0020<TAB>0.07560898645620365
0.0030<TAB>0.07638645316629832
0.0040<TAB>0.07716391987639298
0.0050<TAB>0.07794138658648765
0.0060<TAB>0.07871885329658232
You can now go on and parse the two parts into separate files or a format fitting your needs.