Gene Finding

The group’s major contribution to Genomics has been the development of a novel and very accurate gene finding system that uses the latest advances in Machine Learning. Our system outperformed other gene finding systems in a competition on nematodes (on average over the evaluation criteria). We have recently extended this system to additionally take advantage of direct transcriptome measurements, which leads to an improved prediction performance.

People specializing in this area

Graduate Students

Jonas Behr

Our gene finding system mGene is very flexible in terms of the features incorporated in training and prediction. Beside utilization of very accurate SVM based genomic signal and content detectors we are able to include transcriptome measurement features to improve the gene finding accuracy.

More about Jonas Behr...

Alumnae and Alumni

Peter Niermann

I'm currently doing my diploma thesis about gene finding on the human genome. I'm fitting the mGene pipeline to the large size of the human genome trying to outperform prediction accuracy of other genefinders.

More about Peter Niermann...

Gabriele Schweikert, Ph.D.

I significantly contributed to the development of our discriminative gene finding system, mGene. It is the extension to a full ab-initio gene finder from our earlier method mSplicer, that correctly predicted the exon/intron structure when given the DNA sequence of a gene. We participated in the international competition nGASP with an initial version of mGene and achieved very good results. I continued to refine the model to further improve the prediction performance. Additionally I applied the system to four nematode genomes (C. elegans, C.remanei, C. briggsae, C. brenneri and C.japonica). These predictions are now part of the official wormbase annotation. I also contributed to the development of a Galaxy-based web interface, that allows gene prediction on DNA sequences, as well as the retraining of the model for other organisms.

More about Gabriele Schweikert, Ph.D....

Sören Sonnenburg, Ph.D.

I am working on large scale learning algorithms required to get the involved machine learning algoritms (signal and content predictors, label sequence learning) applicable to genomic data sizes that often have heavy computation time and memory requirements.

More about Sören Sonnenburg, Ph.D....

Georg Zeller, Ph.D.

For the initial version of mGene, with which we participated at nGASP, I contributed to the development of the polyA signal sensor and to sensors for segment content and reading frames.

More about Georg Zeller, Ph.D....

Alexander Zien, Ph.D.

My main contribution to the genefinder mGene is in the development of signal sensors.

More about Alexander Zien, Ph.D....

Publications

• Schweikert, G, Zien, A, Zeller, G, Behr, J, Dieterich, C, Ong, CS, Philips, P, De Bona, F, Hartmann, L, Bohlen, A, Krüger, N, Sonnenburg, S, and Rätsch, G (2009).
  mGene: Accurate SVM-Based Gene Finding with an Application to Nematode Genomes
  Genome Research, 19:2133-2143.
  Supplementary website http://mgene.org
  Printable file

• Schweikert, G, Behr, J, Zien, A, Zeller, G, Ong, CS, Sonnenburg, S, and Rätsch, G (2009).
  mGene.web: a web service for accurate computational gene finding
  Nucleic Acids Research, 37(Web Server issue):W312–W316.
  Supplementary website http://mgene.org/web
  Printable file

• Rätsch, G, Sonnenburg, S, Srinivasan, J, Witte, H, Müller, KR, Sommer, R, and Schölkopf, B (2007).
  Improving the C. elegans genome annotation using machine learning
  PLoS Computational Biology, 3(2):e20.
  Supplementary website http://compbiol.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pcbi.0030020
  Printable file

• Sonnenburg, S, Rätsch, G, and Schölkopf, B (2005).
  Large Scale Genomic Sequence SVM Classifiers
  In: Proceedings of the International Conference on Machine Learning, ICML.
  Supplementary website http://www.tuebingen.mpg.de/raetsch/projects/LargeScaleSVMs\_ICML2005