Title :
Determination of protein location diversity via analysis of immunohistochemical images from the Human Protein Atlas
Author :
Rao, Arvind ; Murphy, Robert F.
Author_Institution :
Lane Center for Comput. Biol., Carnegie Mellon Univ., Pittsburgh, PA, USA
fDate :
March 30 2011-April 2 2011
Abstract :
The Human Protein Atlas (HPA) is a repository of location patterns of about 11000 proteins within tissues and cell lines. In this work we summarize some of our current work on analyzing immunohistochemical images of proteins within 7 distinct tissues. Firstly, we present our efforts to analyze spatial point patterns of protein staining and determine protein subcellular location from image-derived proximity features. Next, we present a method to identify proteins that have differential location between normal and cancer tissue. These methods provide a principled way to characterize location diversity of the proteome across normal and cancer tissues.
Keywords :
biochemistry; biological tissues; biology computing; cancer; feature extraction; image classification; medical image processing; molecular biophysics; proteins; proteomics; cancer tissue; cell lines; human protein atlas; image classification; image-derived proximity features; immunohistochemical images; normal tissue; protein location diversity; protein staining; proteome diversity; Biomarkers; Cancer; DNA; Humans; Immune system; Pixel; Proteins; biomarkers; graph proximity; protein atlas; random forest;
Conference_Titel :
Biomedical Imaging: From Nano to Macro, 2011 IEEE International Symposium on
Conference_Location :
Chicago, IL
Print_ISBN :
978-1-4244-4127-3
Electronic_ISBN :
1945-7928
DOI :
10.1109/ISBI.2011.5872738