This blog post was originally published on here.
Last year, the Technical University of Munich (TUM) was named a winner of the SAP Innovation Awards for achieving a breakthrough in protein analysis on cancer research. Leveraging big data analytics, the team was able to make a major advancement in protein analysis: A team headed by Prof. Dr. Bernhard Küster succeeded to create a comprehensive map of proteins within the human body. Analyzing this data will help to fight deadly diseases more effectively with targeted treatments.
80% of the human proteome can be accessed “online”
“If we do understand, particularly in the context of diseases, how things work inside a cell, inside an organ, inside a human as a whole, we might better understand how we can tackle diseases,” said Prof. Dr. Bernhard Küster, Head of the Chair of Proteomics and Bioanalytics at the Technical University of Munich. “This has been a quest for mankind for centuries.”
Such a process requires the continuous collection of a high volume of data and complex aggregations that cannot be handled by conventional databases. TUM needed to find a solution that continually populated its database and integrate it with statistics and analytics applications running on a server. The predictive analytics capabilities of SAP HANA was key to achieving this.
“Our technology platform to analyze proteins creates a lot of data,” said Küster. “Each of our mass spectrometers produces data at the rate of 100 megabytes an hour, and we have half a dozen of them that do this, day in, day out. So, we have to tackle this mountain of data in a way that we can make it comprehensible for humans.”
Personalizing medicine with machine learning
The treatment of complex diseases, such as cancer, does not follow a one-size-fits-all approach. The process works differently for each patient because the presence and abundance of proteins differ from one person to another. The Technical University of Munich believes that decoding the proteins in the human body helps the medical industry understand diseases better and develop personalized medication for patients.
TUM set out a new massive in-memory database based on SAP HANA to capture the data generated in the laboratory.
“The underlying idea is to quickly distill the essential information from the vast quantities of junk data,” said Küster. “Doing this in real-time enables the group to process more information per unit of time.”
Leveraging the machine learning capabilities, TUM reduced overhead, streamlined, and accelerated the data processing within a multi-tenant database. This helps TUM to:
- Generate advice for doctors regarding personalized treatments in a ‘Molecular Tumor Board’
- Provide a better understanding of the mechanisms behind how cancer drugs work in patients
- Produce hard evidence that can guide treatments rather than relying on guesswork
Empowering people across the network
To further boost research worldwide, the Technical University of Munich launched a free online database called ProteomicsDB, powered by SAP HANA. This database enables internal and external researchers to contribute to and benefit from several databases. This database also supports other research teams, e.g., in the area of plants and animals, in developing their work further.
“With our ProteomicsDB based on SAP HANA, we have the perfect platform to help medical scientists around the world to create more effective treatment options to save lives in the future,” said Küster.
Streamlined data and accelerated data processing help TUM to gain a deeper understanding of the relationship between proteins and the way they interact with drugs. This helps scientists to identify previously unknown correlations and possible drug sensitivity predictions.
By collecting and analyzing huge volumes of data with a powerful database such as SAP HANA running on IBM Power Systems, researchers are empowered to improve the health outcomes across the world and open up new doors to unexplored areas of science.
To learn more about the Technical University of Munich, check out their SAP Innovation Awards Presentation.