CEA: Researchers find breakthroughs in millions of rows of data


Antonio Cosma, Research Scientist at CEA, and his team now answer data questions in minutes with Tableau. Today, CEA can analyze big, multi-dimensional data sets. For example, one sample contains approximately 1 million cells—and the research team analyzes many samples at a time. With Tableau, they can see an overview of all of the data, or drill down into specific details. Testing hypotheses used to take up to 3 hours. Now, Antonio says, “If I have a question, I can address it right away.”


Tableau: What led you to look for a visual analytics solution? Antonio Cosma, Research Scientist: The CEA needed a database, so we began collecting data, but suddenly a problem arose. How could we visualize the data? We weren’t satisfied with the tools that were available. So I began searching for a solution to visualize and analyze my data. Tableau: How did you come to choose Tableau? Antonio: After just a few hours of using Tableau, I realized that this was the tool I needed. I'd say it was the tool that I'd been looking for to see my data multi-dimensionally and aggregate it rapidly. Tableau: Give an example of how Tableau helps you see and understand your data. Antonio: This flow cytometry data is unique because we are analyzing the cells of an organism one by one. When we have an individual, a sample, we analyze, for example, 1 million cells. This means an analysis of 1 million lines. If we visualize this data in an Excel file, for example, we have something like thirty or forty parameters for each line. This is an enormous quantity of multi-dimensional data.

After just a few hours of using Tableau, I realized that this was the tool I needed. I'd say it was the tool that I'd been looking for to see my data multi-dimensionally and aggregate it rapidly.

Antonio: Tableau allows the user to create filters and visualize the data in a simple way. Moreover, as I have already described, it is a sample. It is clear that we don’t have a single sample; we have many samples associated with many individuals and multiple groups of individuals, who, for example, are suffering from various diseases. Tableau: What has changed in your workflow since you started using Tableau? Antonio: In our industry, we often raise a question and then say, "OK, in order to resolve this issue, I need to recover all of my data as well as filter and process it." This is when we ask ourselves, "Is this question worth the trouble of two or three hours of work?" And sometimes we say, "OK, well, I’m going to put this question aside." With Tableau, in contrast, I don't have this problem anymore. If I have a question, I can address it right away. Even if the question is of lesser importance with only a slight chance of success, I can visualize it immediately. Tableau: How have your colleagues received your work in Tableau? Antonio: In general, our data analyses are rather impressive for our partners (colleagues). Many people approach us to carry out analyses with Tableau and our system of organizing multi-dimensional data. As a result, the CEA started out with my license and now we have more than twenty licenses. We are in the process of rolling out Tableau to all of our departments, which means more than a hundred users. And there are many laboratories working with us, collaborators with whom we intend to use Tableau, so the number of licenses is sure to grow.As a result, the CEA started out with my license and now we have more than twenty licenses. We are in the process of rolling out Tableau to all of our departments, which means more than a hundred users. And there are many laboratories working with us, collaborators with whom we intend to use Tableau, so the number of licenses is sure to grow.