======================================== Supporting Data for Coelho et al. (2013) ======================================== This is supporting data for the paper: Luis Pedro Coelho, Joshua D. Kangas, Armaghan Naik, Elvira Osuna-Highley, Estelle Glory-Afshar, Margaret Fuhrman, Ramanuja Simha, Peter B. Berget, Jonathan W. Jarvik, and Robert F. Murphy (2013) Local Features Provide Better Generalization of Subcellular Location Classifiers to New Proteins. Bioinformatics DOI: 10.1093/bioinformatics/btt392 http://dx.doi.org/10.1093/bioinformatics/btt392 This paper should be cited in publications using the data and consulted for details. About ----- These datasets consists of fluorescent microscope images of GFP-tagged proteins locating to different organelles. The organelles were annotated by visual inspection. For each organelle, multiple proteins locating to it were identified; and for each protein, multiple images were acquired. Images contain two channels: GFP tagged protein and a nuclear marker (Hoechst). Structure --------- Data consists of two subsets: 1) The widefield dataset 2) The confocal dataset In both cases cells were tagged using CD tagging as described by García Osuna et al. (2007). The first dataset was acquired on a widefield microscope (IC 100 from Vala Sciences), the second one on a confocal microscope (Zeiss). Both datasets are organised using a directory per class with files named according to the pattern (PROTEIN ID)-(IMAGE ID)-(CHANNEL ID).tiff. For example, 02-011-dna.tiff is the DNA channel of image 11 for protein 2. References ---------- * García Osuna E, Hua J, Bateman NW, Zhao T, Berget PB, Murphy RF. Large-scale automated analysis of location patterns in randomly tagged 3T3 cells. Ann Biomed Eng. 2007 Jun;35(6):1081-7. Epub 2007 Feb 7.