This study will look at genetic changes which occur in the development of male and female breast cancer and other cancer. It will use a new technology called DNA microarray hybridization that looks at a wide array of genes to identify disease-associated patterns in the human genome (complete set of human genes). Numerous studies have linked particular genes to a given disease, but there is very little information on patterns of gene expression (production of proteins from genetic coding) in the entire human genome.

Pinpointing genetic abnormalities in disease may help classify different forms of cancer and perhaps lead to new avenues of treatment or prevention. A primary goal of this study will be to create a database of gene expression for human cancers and other disorders that will provide the basis for finding genetic abnormalities in disease.

Tumors specimens used in this study will be taken from tissues biopsied from patients with breast, colon cancer, sarcomas or melanoma as part of their routine care. Patients in the study will be among those receiving care at the: Department of Oncology, University Hospital, University of Lund, Sweden (breast cancer); Department of Medicine, University of Michigan, Ann Arbor, Michigan (breast cancer); Surgery Branch, National Cancer Institute, Bethesda, Maryland (melanoma), Johns Hopkins Univ. (colon cancer), Memorial Sloan Kettering (sarcoma).

Patients in the study will have a family history taken and will complete a questionnaire. Some patients will be asked to have a blood test. Breast cancer patients will have a mammogram if one has not been done within the last year.

The purpose of our study is to make use of a novel technology that the Cancer Genetics Branch of the NHGRI has been a leader in developing. This technology for genome-wide expression analysis, DNA microarray hybridization, is the focus of our protocol. We will access tissue banks collected by our collaborators that contain excess tissues obtained during routine clinical care. Specimens will be processed for large-scale gene expression analysis and DNA copy number determination using DNA microarrays. The development and analysis of this gene expression and gene copy number database are the primary purpose of this study. Currently available and new bioinformatics tools will be applied to the data for the characterization of disease subsets (e.g., early vs. advanced stage cancer) as well as to mine the data for specific genes which are linked to given disease states.

• Clinical inclusion/exclusion criteria will be dependent upon the collaborating Institutions' requirements.