BioStar is a site for asking, answering and discussing bioinformatics questions and issues. We are members of the community and find it very useful. Often questions and answers arise at BioStar that are germane to our readers (end users of genomics resources). Every Thursday we will be highlighting one of those items or discussions here in this thread. You can ask questions in this thread, or you can always join in at BioStar.
This week’s question highlights the cancer data issues, and since I did a cancer database as “Tip of the Week” that got some decent interest, I thought I’d keep with the theme. Maybe some researchers who are more familiar with the cancer data sets will have some insights.
I was looking for expression levels for two genes involved in serous ovarian cancer data from TCGA via CBio portal. Based on a Z-score threshold of 2.0, I found the following percentage of samples (cases) have expression levels affected (UP or Down)
Case Set: Tumors with mRNA data (Agilent microarray): All samples with mRNA expression data (489 samples) CCNE1 – 11% CDK12 – 7%
For same dataset in Oncomine (I am using the free version):
TCGA Ovarian (517 samples <- this number is higher than what is reported in TCGA ovarian cancer publication) expression data is provided as log-2 median intensity and the in Oncomine shows that higher expression level of CCNE1 and CDK12 expression level is correlated with different grades – for example Grade 3 tumor (Grade 3 (431 samples) have higher expression level of both genes.
I have also noticed that the dataset 517 samples were assigned as No Associated Paper 2011/03/24. I am wondering if the data is reffering to this paper on TCGA ovarian cancer dataset. http://www.nature.com/nature/journal/v474/n7353/full/nature10166.html
I am wondering why such a discrepancy or am I missing something here.
PS. I have posted this question on both Oncomine and cBio list, but did not receive any responses yet. I am wondering if anyone here with experience on one of the platform could provide insight to this.
The other issue that interested me was the support problem. This is a skilled super-user trying hard to do it right–contacting the support teams of the sites, and getting no response. I think that’s one of the most frustrating things about this arena. Some projects are well resourced for user support. Some are not. But if smart users can’t figure out what’s going on with your site’s data, your resource isn’t as useful as you think it is. I wish support was valued more. But if you know what’s up–go over and offer an answer.