Discrepancies in gene expression pattern data: How to interpret?
|
| Sample User Question
|  |
| KLK3 (Prostate Specific Antigen) is, as expected, highly expressed in normal and cancerous prostate tissue in the SAGE Anatomic Viewer. It is not expressed to a significant degree in any other tissue. However, in the UniGene Expression Profilr of KLK3 it appears to be present in normal and cancerous breast tissue as well. Similarly, KLK3 is expressed in the GEO Profile of HCC 1954 (Breast Cancer cell line) viewed earlier in this module. How might one resolve this seeming conflict in RNA expression data? |
| Step By Step Guide |
|
- A search of PubMed on "KLK3 AND Breast Cancer" results in a number of recent references showing PSA protein is also a marker for cancer is breast neoplasms in males and females (see PMID: 16575473). KLK3 protein is present at a low level in 7 out of 12 breast tissue samples in the Human Protein Atlas (http://www.hpr.se/).
- It is not clear that in any of these databases what breast tissue samples are being used and if these samples overlap between the different databases.
- Furthermore, there is a question of such issues as normalization accross he different platforms used
Bottom line--this is an example of how lack of standardization at a number of levels renders expression data from the various sources questionable and hard to inerpret...
|
|
