A recent paper used TCGA data to show that cancer patients with higher expression/amplified copy number of MTBP had reduced survival. Thus, their data revealed that "MTBP significantly contributes to breast cancer and is a potential novel therapeutic target in the treatment of TNBC". To evaluate MTBP expression, the paper researched mRNA data, copy number variation data and patient survival data from TCGA (The Cancer Genome Atlas) to validate the results. Omicsoft users can easily use OncoLand, and the TCGA Land to verify and explore these results in minutes. We archived a nice OncoLand case study in our wiki:
- 1.1 Visualize MTBP expression in tumor vs. normal breast cancer tissue
- 1.2 Categorize breast tumor samples by MTBP expression quartiles
- 1.3 Categorize breast tumor samples by MTBP copy number
- 1.4 Create a SampleSet of Triple-Negative Breast Cancer samples
- 1.5 Filter and Group per-sample gene expression by a SampleSet
- 1.6 Subgroup sample expression data by multiple variables
A typical case starts with visualizing the data. After selecting TCGA2015 Land, users can search for MTBP gene in the search box. Go to Gene FPKM View to visualize gene expressions. Next, filter Tumor Type to breast cancer (BRCA) and select Sample Type for grouping. Thus, you will generate a plot that looks like this:
The user can subset by using MTBP expression level to test whether tumors with high expression correlate with other variables, such as survival. You can achieve this by generating a MTBP Gene FPKM custom query and labeling samples according to the query. Only tumor samples express high level of MTBP.
Next, we can correlate MTBP expression with survival. As we can see, survival of high MTBP expressing samples is significantly worse than low expressing samples:
If you want to explore how amplification or deletion of a gene correlates with expression, mutation, or clinical metadata, you can perform a "custom query" on Copy Number Variation as well (please refer to our OncoLand case study wiki for details). Our results are consistent with the results from the paper, for example, MTBP was significantly elevated in breast cancer samples (Figure 1A) and those patients with elevated MTBP expression exhibited reduced survival (Figure 1B):
Breast cancers that are estrogen receptor-negative, progesterone receptor-negative and HER2-negative can not be targeted by many common treatments for breast cancer, for instance, common HER2 inhibitors like trastuzumab, however other treatment plans are still available.
We can use TCGA clinical meta data to identify triple negative samples. By filtering ER, PR and HER2 status from Sample | Clinical Data | Procedure |Genetic Testing filter options, you can select those triple negative data and create a Sampleset using Group Sample Set From Selection:
You can quickly partition a gene's expression by each permutation of multiple clinical variables i.e. triple-negative breast cancer samples in this case. And as we can see, triple-negative samples indeed have higher MTBP expression than any other samples:
To learn how to perform the analyses, please visit our wikipage MTBP expression and copy number correlates with poor patient survival