You are here: Home › People › Tobias Girschick, Dipl.-Bioinf.

Tobias Girschick, Dipl.-Bioinf.

Tobias Girschick, Dipl.-Bioinf.

Ph.D. student

 Room: MI 01.09.042
Technische Universität München
Institut für Informatik / I12
Boltzmannstr. 3
85748 Garching b. München, Germany


Fax number: +49 (89) 289-19414


Research Interests:

machine learning, data mining, bio-/cheminformatics, toxicology, small molecule representations, in silico drug discovery, algorithms for small molecules, graph algorithms, chemical and biological similarity of small molecules


Research:

I am currently finishing my PhD thesis in the Machine Learning and Data Mining in Bioinformatics group in a collaborative project with the global research-based biopharmaceutical company AstraZeneca. The project is about small molecule similarity in quantitative structure activity relationships (QSARs) and how to improve those for various toxicological endpoints on publically available and AstraZeneca in-house datasets. The working title of my PhD thesis is Small Molecule Similarity for In silico toxicology prediction.

I also worked two and a half years as an algorithm and REST webservice developer and researcher on the EU FP7 project OpenTox (Project-No.: HEALTH-F5-2008-200787). The overall objective of the OpenTox project was to develop a framework that provides a unified access to toxicity data, (Q)SAR models, procedures supporting validation and additional information that helps with the interpretation of (Q)SAR predictions. I was responsible for algorithm and web service development and implementation as well as involved in infrastructure and API design. Additionally, I was in the organizing team of the international conference:  Innovation in Predictive Toxicology: OpenTox InterAction Meeting 2011 held in Munich.

I am also a alumni of the CeDoSIA class of 2008. CeDoSIA is a facultiy graduate school for computer science in the TUM Graduate School, that offers additional soft skill training for PhD candidates. Before joining Professor Kramer's lab, I studied Bioinformatics in the joint diploma program of Technische Universität München and Ludwig-Maximilians-Universität München and wrote my diploma thesis in  collaboration with the biotech company 4SC AG in Martinsried. The topic of the thesis was Machine learning based prediction of compound behavior in metabolism relevant Cytochromes P450.


Activities:

 

(Co-) Reviewing:

  • IEEE International Conference on Data Mining (ICDM-2008, 2009)
  • ACM Symposium on Applied Computing (SAC-2009 Data Mining Track, 2010)
  • European Conference on Machine Learning/European Conference on Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD-2008, 2010, 2011)
  • International Conference on Discovery Science (DS-2008, DS-2011)
  • International Workshop on Mining and Learning with Graphs (MLG-2008, 2009)
  • SIAM International Conference on Data Mining (SDM-2009, 2010)
  • AAAI-10 Special Track on Artificial Intelligence and Bioinformatics (AIB-2010)
  • International Conference on Extending Database Technology (EDBT-2012)
  • BMC Bioinformatics

Publications:

Girschick, T, Rückert, U, and Kramer, S (accepted).
Adapted Transfer of Distance Measures for Quantitative Structure-Activity Relationships and Data-Driven Selection of Source Datasets
The Computer Journal.

Alphonse, E, Girschick, T, Buchwald, F, and Kramer, S (2011).
A Numerical Refinement Operator based on Multi-Instance Learning
In: Inductive Logic Programming - 20th International Conference, ILP 2010, ed. by Paolo Frasconi and Francesca A. Lisi, vol. 6489, pp. 14-21, Springer. LNCS/LNAI.

Buchwald, F, Girschick, T, Seeland, M, and Kramer, S (2011).
Using Local Models to Improve (Q)SAR Predictivity
Molecular Informatics, 30(2-3):205-218.

Buchwald, F, Girschick, T, Frank, E, and Kramer, S (2010).
Fast Conditional Density Estimation for Quantitative Structure-Activity Relationships
In: Proceedings of the 24th AAAI Conference on Artificial Intelligence, pp. 1268-1273, AAAI Press.

Girschick, T, Buchwald, F, Hardy, B, and Kramer, S (2010).
OpenTox: A Distributed REST Approach to Predictive Toxicology
In: Proceedings of the 3rd Workshop on Third-Generation Data Mining: Towards Service-Oriented Knowledge Discovery (SoKD’10) at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2010, ed. by Melanie Hilario, Nada Lavrač, Vid Podpečan, Joost N. Kok, pp. 61-62.

Hardy, B, Douglas, N, Helma, C, Rautenberg, M, Jeliazkova, N, Jeliazkov, V, Nikolova, I, Benigni, R, Tcheremenskaia, O, Kramer, S, Girschick, T, Buchwald, F, Wicker, J, Karwath, A, Gütlein, M, Maunz, A, Sarimveis, H, Melagraki, G, Afantitis, A, Sopasakis, P, Gallagher, D, Poroikov, V, Filimonov, D, Zakharov, A, Langunin, A, Gloriozova, T, Novikov, S, Skvortsova, N, Druzhilovsky, D, Chawla, S, Gosh, I, Ray, S, Patel, H, and Escher, S (2010).
Collaborative Development of Predictive Toxicology Applications
Journal of Cheminformatics, 2(7).

Rückert, U, Girschick, T, Buchwald, F, and Kramer, S (2010).
Adapted Transfer of Distance Measures for Quantitative Structure-Activity Relationships
In: Proceedings of the 13th International Conference on Discovery Science, ed. by B. Pfahringer, G. Holmes, A. Hoffman, vol. 6332, pp. 341-355, Springer. LNCS/LNAI.

Seeland, M, Girschick, T, Buchwald, F, and Kramer, S (2010).
Online Structural Graph Clustering using Frequent Subgraph Mining
In: Proceedings of the European Conference of Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ed. by J.L. Balcazar, F. Bonchi, A.Gionis, M.Sebag, vol. 3, pp. 213-228.


Selected Talks and Tutorials:

Talks:

  • Fast Conditional Density Estimation for Quantitative Structure-Activity Relationships, at the 24th AAAI Conference on Artificial Intelligence (2010), Atlanta, GA
  • Machine learning based prediction of compound behavior in CYP 3A4, CYP 2D6 and CYP 2C9, at the 4th Russian-Bavarian Conference on Biomedical Engineering (2008), Moscow


Posters:

  • Girschick, T, Buchwald, F and Kramer, S (2010), Combining Local and Global Chemical Distance Measures for Quantitative Structure-Activity Relationships, 18th European Symposium on Quantitative Structure-Activity Relationships, Rhodes, Greece
  • Seeland, M, Girschick, T, Buchwald, F and Kramer, S (2010), Clustering Structure Databases Using Frequent Substructure Mining, 18th European Symposium on Quantitative Structure-Activity Relationships, Rhodes, Greece

Teaching:

Supervised Theses:

  • S. Seemayer; "Significant Similarities of Small Molecules in 3D", 2011. (Master Thesis)
  • K. Ciosek; "Teilüberwachtes Lernen mit Untergraphen" (in German) [Semi-supervised learning with Subgraphs], 2010 (Master Thesis)
  • M. Stransky; "Conditional Density Estimation with Continuous Intervals", 2010. (Bachelor Thesis)
  • S. Seemayer; "Molecular Paths - a Novel Structural Descriptor for Small Molecule Similarity", 2009. (Bachelor Thesis)
News
Wissenschaftsmarkt Posters (Sep 17, 2012)
New Web Page (Feb 16, 2012)