Identification of Peptides with Deviating Regulation Factors Using a Robust Clustering Scheme
A new approach to clustering likelihood curves is introduced which is based on the maximal density estimator algorithm. The clustered objects are the results of the analysis of mass spectrometry data and represent regulatory information of peptides, which belong to the same protein. The aim of the research is to reveal peptides within a protein sequence that show deviating regulation factors, caused either by the presence of noise in the measurements, the assignment of a peptide to a wrong protein or a modification of a peptide. The proposed approach allows arranging all the studied proteins into two groups: those, consisting of a single cluster of peptides and those with more than one cluster or with one or more outlier peptides with a regulation differing from the main cluster of peptides belonging to the protein.
KeywordsPhosphorylation Site Regulation Factor Single Cluster Curve Number Mass Spectrometry Data
Unable to display preview. Download preview PDF.
- 2.Nasraoui, O., Krishnapuram, R.: A robust estimator based on density and scale optimization, and its application to clustering. In: Proc. FUZZIEEE, New Orleans, pp. 1031–1035 (1996)Google Scholar
- 3.Nasraoui, O., Krishnapuram, R., Joshi, A.: Relational Clustering Based on a New Robust Estimator with Applications to Web Mining. In: Proc. NAFIPS 1999 (1999)Google Scholar
- 4.Hundertmark, C., Jänsch, L., Klawonn, F.: Fuzzy clustering of likelihood curves for finding interesting patterns in expression profiles. In: Mumford, C.L., Jain, L.C. (eds.) Computational Intelligence: Collaboration, Fusion & Emergence. ISRL, vol. 1, pp. 599–622. Springer, Heidelberg (2009)Google Scholar