
This is the dinucleotide weight tensor (DWT) Toolbox described in our
manuscript: 

Automated Incorporation of Pairwise Dependency in Transcription Factor
Binding Site Prediction Using Dinucleotide Weight Tensors .

Saeed Omidi, Mihaela Zavolan, Mikhail Pachkov, Jeremie Breda, Severin
Berger, and Erik van Nimwegen

http://biorxiv.org/content/early/2016/09/28/078212

-----------------------------------------------------------------------

A description of the terms of usage of this software is contained in
the file LICENCE.txt.

-----------------------------------------------------------------------


A description of how to install the DWT-toolbox is provided in the
file INSTALLATION.txt.


-----------------------------------------------------------------------

All source code is in the subdirectory Source.

Once installed, the toolbox contains the following executables:
1. DWT_model (For training DWT motifs and predicting sites with DWT models).
2. diLogo.py (For creating dilogo images of DWT motifs).
3. init_dwt_from_pwm.pl (For creating an initial DWT motif from a PWM file).

------------------------------------------------------------------------


A description of the usage of all programs and their parameters is
provided in the file: USER_MANUAL.txt

------------------------------------------------------------------------

A set of example usages of the DWT-toolbox is provided in the
subdirectory Examples.  In particular, the file Examples/README.txt
gives examples of the basic usage of the DWT-toolbox.

------------------------------------------------------------------------

The directory PeakSequences contains FASTA files with the peak
sequences that were obtained using our CRUNCH tool (see:
http://dwt.unibas.ch) for a colletion of 121 ENCODE ChIP-seq datasets
(see http://crunch.unibas.ch/ENCODE_REPORTS for links to the raw data
of these experiments).

------------------------------------------------------------------------

The directory PWM_models contains initial position-specific weight
matrix motifs that were inferred (by CRUNCH) on these peak
sequences. These PWM models were used as starting motifs for the DWT
inference.

------------------------------------------------------------------------

The directory DWT_models contains all the DWT models that were
inferred from running the DWT_model training on the peak
sequences. These models can be used for predicting binding sites for
the corresponding TFs in other datasets.



