Using Medians to Generate Consensus Rankings for Biological Data
Faced with the deluge of data available in biological databases, it becomes increasingly difficult for scientists to obtain reasonable sets of answers to their biological queries. A critical example appears in medicine, where physicians frequently need to get information about genes associated with a given disease. When they pose such queries to Web portals (e.g., Entrez NCBI) they usually get huge amounts of answers which are not ranked, making them very difficult to be exploited. In the last years, while several ranking approaches have been proposed, none of them is considered as the most promising.
Instead of considering ranking methods as alternative approaches, we propose to generate a consensus ranking to highlight the common points of a set of rankings while minimizing their disagreements. Our work is based on the concept of median, originally defined on permutations: Given m permutations and a distance function, the median problem is to find a permutation that is the closest of the m given permutations. We have investigated the problem of computing a median of a set of m rankings considering different elements and ties, under a generalized Kendall-τ distance. This problem is known to be NP-hard. In this paper, we present a new heuristic for the problem and we demonstrate the benefit of our approach on real queries using four different ranking methods.
Unable to display preview. Download preview PDF.
- 2.Ailon, N., Charikar, M., Newman, N.: Aggregating inconsistent information: Ranking and clustering. In: Proceedings of the 37th STOC, pp. 684–693 (2005)Google Scholar
- 4.Brancotte, B., Biton, A., Bernard-Pierrot, I., Radvanyi, F., Reyal, F., Cohen-Boulakia, S.: Gene List significance at-a-glance with GeneValorization. To Appear in Bioinformatics (Application Note) (February 2011)Google Scholar
- 6.Blin, G., Crochemore, M., Hamel, S., Vialette, S.: Medians of an odd number of permutations. To Appear in Pure Mathematics and Applications (2011)Google Scholar
- 8.Dwork, C., Kumar, R., Naor, M., Sivakumar, D.: Rank Aggregation Methods for the Web. In: Proceedings of the 10th WWW, pp. 613–622 (2001)Google Scholar
- 9.Fagin, R., Kumar, R., Mahdian, M., Sivakumar, D., Vee, E.: Comparing and Aggregating Rankings with Ties. In: Proceedings of PODS 2004, pp. 47–55 (2004)Google Scholar
- 12.Kenyon-Mathieu, C., Schudy, W.: How to rank with few errors. In: Proceedings of the 39th STOC, pp. 95–103 (2007)Google Scholar
- 14.Laignel, N.: Ranking biological data taking into account user’s preferences, Master thesis report (co-supervised by S. Cohen-Boulakia, C. Froidevaux, and U. Leser), University of Paris-Sud XI (2010)Google Scholar
- 15.Page, L., Brin, S., Motwani, R., Winograd, T.: The PageRank Citation Ranking: Bringing Order to the Web. Stanford University, Stanford (1998)Google Scholar
- 17.Varadarajan, R., Hritidis, V., Raschid, L., Vidal, M., Ibanez, L., Rodriguez-Drumond, H.: Flexible and Efficient Querying and Ranking on Hyperlinked Data Sources. In: Proceedings of the 12th International Conference on Extending Database Technology: Advances in Database Technology, pp. 553–564 (2009)Google Scholar