Big data and the fight against cancer

Being able to understand the way tumors change is crucial in order to distinguish between benign and lethal ones.  A Swedish team have recently created an open-access database that aims to map the various genetic changes a cancer goes through.  The hope is that it will provide insight into patient survival, whilst also supporting the development of new treatments.

The Pathology Atlas is documented in a recently published paper, and aims to highlight the power of big data in changing how medical research is undertaken.

Personalized medicine

The project hopes to help the journey towards personalized treatment for cancer patients.  It’s based on the transcriptome of 17 different cancer types, with data taken from 8,000 patients.

The tool contains an Interactive Survival Scatter plot to highlight survival data, with the database containing over 400,000 such plots.  The database also contains some 5 million different pathology-based images.

“This study differs from earlier cancer investigations, since it is not focused on the mutations in cancers, but the downstream effects of such mutations across all protein-coding genes. We show, for the first time, the influence of the gene expression levels demonstrating the power of “big data” to change how medical research is performed. It also shows the advantage of open access policies in science in which researchers share data with each other to allow integration of huge amounts of data from different sources,” the researchers say.

Early results

The project has already achieved a number of successes.  For instance, a large fraction of genes were found to express themselves differently in cancers, with this having a significant impact on patient survival rates.

The researchers also discovered that the gene expression patterns of individual tumors themselves varied considerably.  This could have a big impact in the variations seen between the different forms of cancer.

What’s more, lower patient survival rates were associated with the up-regulation of genes involved in cell growth, with down-regulation of genes involved in the differentiation of the cells.  The researchers were able to generate personalized, genome-scale models for cancer patients to try and hone in on the key genes involved in the growth of their tumors.

As with many of these big data projects in healthcare, it is heavily dependent upon the kind of computing power I’ve written about previously, with the likes of Cray and Intel both developing supercomputers specifically for the analysis of genetic data.

“We are now in possession of incredibly powerful systems biology tools for medical research, allowing, for the first time, genome-wide analysis of individual patients with regards to the consequence of their expression profiles for clinical survival,” the researchers say.

The team hope to demonstrate the value of the tool on lung and colorectal two particular cancers.  A number of prognostic genes were identified in the Atlas, and were analyzed using immunohistochemistry to validate the gene expression patterns at the protein level.

“We are pleased to provide a stand-alone open-access resource for cancer researchers worldwide, which we hope will help accelerate their efforts to find the biomarkers needed to develop personalized cancer treatments,” the team conclude.