**ClassificationApp** performs a binary classification (such as splice sites vs decoys) of sequences that can be scored by two iPMMs. Alternatively, it allows to classify sequences based on an iPMM as foreground model and a homogeneous Markov chain of given order that is learned on a given negative data set set.

If the content of the "Input data" file starts with '>', it is interpreted as FastA file. Otherwise it is interpreted as plain text, where every line contains a single sequence. 
The input expects upper- and lower case letters of the standard DNA alphabet {A,C,G,T}. If other symbols from the IUPAC code (such as N) are encountered, they are replaced by a random sample from the distribution of {A,C,G,T} in the data set.
All sequences in the data set need to be of width *W* so that they can be scored by the given models.

As "Foreground model" any successfully learned iPMM of width *W* can be used and it does not matter which of the learned tools has produced it.
 
The "Background" determines the model that is to be used for the background class. 
It is a selection parameter, which means that different parameters may need to be specified depending on the selection made.
If the selection is *Uniform PWM*, no additional parameters need to be set. The background model is then a simple PWM model with all position-specific nucleotide probabilities equalling 1/4. 
If the selection is *Previously learned*, a background "Model" can be an iPMM (in XML representation) of possibly different order than "Foreground model", but it must share the same width *W*.
If the selection is *Generating from data*, a homogeneous Markov chain of user-specified "Order" is learned from user-specified background "Data". This model in then used as background model during the classification.

Alternative to a given background model, the user may also specify "Background data" on which a homogeneous Markov model of order "Background order" is learned.
This model in then used as background model during the classification. 
Both "Background data" and "Background order" are accessed only if "Background model" is not specified (otherwise ignored), but in that case they need to be given.

If no "Name" is specified, it is set by default to "Classification".
The tool returns a list with pairs of sequence IDs and class assignment.