Structured output prediction of anti-cancer drug activity

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


We present a structured output prediction approach for classifying potential anti-cancer drugs. Our QSAR model takes as input a description of a molecule and predicts the activity against a set of cancer cell lines in one shot. Statistical dependencies between the cell lines are encoded by a Markov network that has cell lines as nodes and edges represent similarity according to an auxiliary dataset. Molecules are represented via kernels based on molecular graphs. Margin-based learning is applied to separate correct multilabels from incorrect ones. The performance of the multilabel classification method is shown in our experiments with NCI-Cancer data containing the cancer inhibition potential of drug-like molecules against 59 cancer cell lines. In the experiments, our method outperforms the state-of-the-art SVM method.
Original languageEnglish
Title of host publicationPattern Recognition in Bioinformatics
Subtitle of host publication5th IAPR International Conference, PRIB 2010, Nijmegen, The Netherlands, September 22-24, 2010. Proceedings
EditorsTjeerd M. H. Dijkstra, Evgeni Tsivtsivadze, Elena Marchiori, Tom Heskes
PublisherSpringer Verlag
ISBN (Electronic)978-3-642-16001-1
ISBN (Print)978-3-642-16000-4
Publication statusPublished - 2010
MoE publication typeA4 Article in a conference publication
EventIAPR International Conference on Pattern Recognition in Bioinformatics - Nijmegen, Netherlands
Duration: 22 Sep 201024 Sep 2010
Conference number: 5

Publication series

Name Lecture Notes in Computer Science
ISSN (Print)0302-9743


ConferenceIAPR International Conference on Pattern Recognition in Bioinformatics
Abbreviated titlePRIB

Fingerprint Dive into the research topics of 'Structured output prediction of anti-cancer drug activity'. Together they form a unique fingerprint.

Cite this