Helsinki Institute for Information Technology

About: Helsinki Institute for Information Technology is a facility organization based out in Espoo, Finland. It is known for research contribution in the topics: Population & Bayesian network. The organization has 630 authors who have published 1962 publications receiving 63426 citations.

Next step in electronic brainstorming: collaborative creativity with the web

Abstract: Brainstorming is an essential technique in creative group work. Research literature indicates the strengths of electronic brainstorming over face-to-face work. Despite this evidence, the old practice dominates. We believe that this is due to the inadequate integration of new tools to existing practices and the tendency to focus on idea production alone. This paper explores how to augment traditional, collocated Brainstorming and make electronic brainstorming feasible and accessible with web-based technology. We introduce an electronic brainstorming application prototype and justify its design principles. Our system aimed at facilitating conceptual design and we present design insights from a pilot study with the prototype used by 27 design students. The paper argues that by structuring the generative group process with a low-cost tool, users can sprint through a creative process, from problem definition to defining a solution.

Comprehensive data-driven analysis of the impact of chemoinformatic structure on the genome-wide biological response profiles of cancer cells to 1159 drugs

Abstract: Detailed and systematic understanding of the biological effects of millions of available compounds on living cells is a significant challenge. As most compounds impact multiple targets and pathways, traditional methods for analyzing structure-function relationships are not comprehensive enough. Therefore more advanced integrative models are needed for predicting biological effects elicited by specific chemical features. As a step towards creating such computational links we developed a data-driven chemical systems biology approach to comprehensively study the relationship of 76 structural 3D-descriptors (VolSurf, chemical space) of 1159 drugs with the gene expression responses (biological space) they elicited in three cancer cell lines. The analysis covering 11350 genes was based on data from the Connectivity Map. We decomposed these biological response profiles into components, each linked to a characteristic chemical descriptor profile. The integrated quantitative analysis of the chemical and biological spaces was more informative about protein-target based drug similarity than either dataset separately. We identified ten major components that link distinct VolSurf features across multiple compounds to specific biological activity types. For example, component 2 (hydrophobic properties) strongly links to DNA damage response, while component 3 (hydrogen bonding) connects to metabolic stress. Individual structural and biological features were often linked to one cell line only, such as leukemia cells (HL-60) specifically responding to cardiac glycosides. In summary, our approach identified specific chemical structures shared across multiple drugs causing distinct biological responses. The decoding of such systematic chemical-biological relationships is necessary to build better models of drug effects, including unidentified types of molecular properties with strong biological effects.

Multilabel classification through random graph ensembles

Abstract: We present new methods for multilabel classification, relying on ensemble learning on a collection of random output graphs imposed on the multilabel, and a kernel-based structured output learner as the base classifier. For ensemble learning, differences among the output graphs provide the required base classifier diversity and lead to improved performance in the increasing size of the ensemble. We study different methods of forming the ensemble prediction, including majority voting and two methods that perform inferences over the graph structures before or after combining the base models into the ensemble. We put forward a theoretical explanation of the behaviour of multilabel ensembles in terms of the diversity and coherence of microlabel predictions, generalizing previous work on single target ensembles. We compare our methods on a set of heterogeneous multilabel benchmark problems against the state-of-the-art machine learning approaches, including multilabel AdaBoost, convex multitask feature learning, as well as single target learning approaches represented by Bagging and SVM. In our experiments, the random graph ensembles are very competitive and robust, ranking first or second on most of the datasets. Overall, our results show that our proposed random graph ensembles are viable alternatives to flat multilabel and multitask learners.