Between September 2010 and August 2011 Dr. U. Petrovič, a member of the project team, has been a visiting professor and researcher at the Donnelly Centre for Cellular and Biomolecular Research (CCBR) of the University of Toronto, Canada. Pioneering work in the field of functional genomics and high-throughput genetics has been performed at the CCBR, and U. Petrovič has been involved in the project where these approaches are being further developed.
B.05 Guest lecturer at an institute/university
COBISS.SI-ID: 2073200Mojca Mattiazzi (co-advisors: Uroš Petrovič and Igor Križaj) was a young researchers that participated and much contributed to the project. She has recently defended her Ph. D. Dissertation. Mojca Mattiazzi has been very active in publishing the results of her research in a peer-reviewed international journals.
D.09 Tutoring for postgraduate students
COBISS.SI-ID: 23893543Lan Umek has developed a new computational tools to infer subgroups of data items with strong relations between experimental description (input features) and experimental outcome (output features). Among other practical demonstrations of utility of these methods he showed that they can be successfully employed in knowledge discovery in chemogenomics.
D.09 Tutoring for postgraduate students
COBISS.SI-ID: 8313684Computational methods that are developed within the project are implemented in Orange, a data mining framework. During the project we have extended Orange with various server-based preprocessing and storage of public chemogenomics and molecular biology data. We have used Orange framework to develop various new data mining approaches that are available either through Python scripting or visual programming. Publication that we are citing here provides a comparison of Orange and similar open-source frameworks.
F.06 Development of a new product
COBISS.SI-ID: 6280532To minimize the amount of required substances to be tested in the genome-wide screening, a higher density yeast colony array was developed. We upgraded the custom-made robotic manipulator to perform pinning at 1536 colonies/plate (previously 384). Currently used methods measure yeast colony area, which however does not take into account differences in three- dimensional shapes of the colonies. We developed an approach whereby we obtain information on the 3D shape by transmission scanning and using a dedicated algorithm to determine the volume.
F.10 Improvements to an existing technological process or technology
COBISS.SI-ID: 2073200