Personal tools

Download mGene

Original Code for mGene.init, mGene.mulit and mGene.seq

The source code of the versions mGene.init, mGene.multi and mGene.seq that have been used for generating the results presented in our paper [1] will be provided here. The intention for releasing this source code - together with initial training data, the trained model, and the test regions - is to enable the reproducibility of our results. These mGene versions will be provided under an academic license and will only be supported to reproduce the published results.

The release will follow in a few days.

Standalone Tools for Training and Prediction with mGene

Additionally, we provide the source code of a mature version of mGene.init licensed under GPL version 3. This will be a further developed version that uses a simpler and thus more general state model than the original version (i.e. no trans-splicing and no operon prediction). It is based on C/C++ and Octave and will be fully supported. The same version is also available with our webservice mGene.web [2]. Using this system one may obtain a trained gene finding system within 24h: for instance, using the nGASP training regions, training signal and content sensors and corresponding predictions each take 2-5h and can run in parallel; the training of the integration step takes about 6h. See [2] for more details.



[1]Schweikert et al. mGene: Accurate Computational Gene Finding with Application to Nematode Genomes. Submitted to Genome Research 2009.
[2](1, 2) Gabriele Schweikert, Jonas Behr, Alexander Zien, Georg Zeller, Cheng Soon Ong, Sören Sonnenburg, and Gunnar Rätsch (2009). mGene.web: a web service for accurate computational gene finding. Nucleic Acids Research, Web server issue.
Document Actions