Jump to content
Powered by

Bioinformaticians shed light into the data jungle

In recent years, the University of Tübingen has developed a tightly integrated service structure for the computer-based management of life sciences data. This structure is the point of contact for all those that produce or work with “omics” data – in Tübingen and beyond.

Prof. Dr. Oliver Kohlbacher has been director of the QBiC since 2012. He studied chemistry and computer science. © private

The terms “genomics” and “proteomics” for the large-scale study of genes and proteins were the first “omics” terms to be introduced. Several more such terms have subsequently been coined, including “transcriptomics” and “metabolomics”, and more will certainly emerge in the future. The common feature of “omics” technologies is that the study of molecules and/or cellular networks leads to huge amounts of data and needs substantial computing and storage capacities. However, quantity does not necessarily equal quality. How good the data are, i.e. the benefit they ultimately have, depends on a number of factors. Many advantages are already apparent from the early phase of projects. Excellent data management from the very start is crucial. More and more universities are therefore establishing bioinformatics centres to provide effective support to scientists from a broad range of life sciences institutes and study areas.  

The University of Tübingen could be said to have played a pioneering role in the field of bioinformatics. The first ever bioinformatics course in Germany was established at the University of Tübingen way back in 1998. According to Prof. Dr. Oliver Kohlbacher, Tübingen now offers a service structure that is still quite unique in Germany. “In Germany, I would say the centre is the only one of its kind and I only know of a few similar centres in the EU,” Kohlbacher says. Tübingen University realised very early on that the university’s skills needed to be pooled in a central bioinformatics unit in order to achieve the objective of getting everyone pulling together in the same direction and as effectively as possible. As the units are closely integrated, Kohlbacher is one of three ZBIT directors as well as being QBiC director and coordinator of the CiBi.

But first things first: the Tübingen Centre for Bioinformatics (ZBIT) is effectively an old lady, at least in terms of bioinformatics. It was established back in 2000 and is the central bioinformatics research and teaching unit at the University of Tübingen. Its service portfolio has grown enormously over the years. “High-throughput technologies meant that we had to expand our services considerably, as our life sciences colleagues were finding it increasingly difficult without outside assistance to cope with the large amount of accumulated data,” says Kohlbacher. Three years ago, Tübingen won a tender to provide core facilities. The goal was to make bioinformatics services available. DFG funding enabled Kohlbacher and his colleagues to set up the Quantitative Biology Centre (QBIC) as a core facility for “omics” data acquisition and bioinformatics. “Prior to the establishment of the QBiC at the University of Tübingen, there were around ten core facilities at different institutes, which all owned devices for generating predominantly gene sequencing and mass spectrometry “omics” data. As a result, there was a duplication of devices across different institutes. But more importantly, the researchers lacked the expertise required for handling the data as well as expertise in the field of bioinformatics,” says Kohlbacher. The establishment of QBiC has completely changed all this. 

Schematic showing the typical course of a bioinformatics project at the QBiC, starting from when the first contact is established (left) to the analysis of the data (right).
Typical course of a bioinformatics project at the QBiC, starting from when the first contact is established (left) to the analysis of the data (right). © QBiC

Bioinformatics threads are pooled at the QBiC

The QBiC team offers services right from the initial planning phase, before the first experiments start. Scientists are given support in experiment design, which enables them to create a conclusive base for data generation. The data are still generated at the individual institutes using their own individual equipment. The QBiC purchases equipment such as sequencers and mass spectrometers with funds from the German Excellence Initiative, but the devices are set up in the existing core facilities. Kohlbacher explains that data generation is no longer the most expensive budget item. “We have reached a point where the costs of data generation no longer determine overall project costs. Nowadays, the principal cost factor is bioinformatics services.” The QBiC can be commissioned to carry out analyses and data interpretation. That said, data output can be handled by the QBiC, but it can also be done by the individual institutes. “We have established a web-based interface which enables users to access their raw data as well as the evaluation data. This interface also enables them to navigate their data,” says Kohlbacher who attaches great importance to sustainability. This means that redundant back-ups to archive and protect data are set up. These archives can be accessed for at least 10 years. The QBiC’s services are not only available to University of Tübingen staff. Research institutions and companies can also use the centre’s full range of services, from initial advice to analyses and support for data evaluation and interpretation. “Some of our customers request the full package, others do the sequencing themselves and contact us for support for processing the raw data,” says Kohlbacher. Other services offered by the bioinformaticians from Tübingen include: development of analysis workflows specifically tailored to the research issues of their life sciences colleagues, and development of customised web-based data visualisation tools. Considering the huge number of different tasks bioinformaticians have to deal with, it is no wonder that the discipline finds it difficult to recruit all the specialists it requires. This is a general problem, as Kohlbacher is well aware. “There are simply not enough bioinformaticians on the market. And people usually underestimate the time it takes and the cost of solving biological issues using bioinformatics methods and tools. Depending on the method, it might take ten minutes to process ten sequences but several months of computing time to process a hundred sequences.” In order to raise researcher awareness of the difference, Kohlbacher would like to see greater integration of bioinformatics into standard curriculums and training programmes. “People who have come across bioinformatics in lectures and practical courses will find it much easier to assess projects and the potential demand for bioinformatics methods and tools.”

Data analysis tools across all "omics" levels

Metabolomics example: section of a metabolic pathway that was visualised using bioinformatics analysis tools. 3D data peaks are shown in the background. © Kohlbacher, University of Tübingen

The most recent addition to the Tübingen bioinformatics infrastructure is the Center for Integrative Bioinformatics (CiBi), which is jointly run by the QBiC and the ZBIT. The CiBi is one of six bioinformatics centres in Germany that together make up the “de.NBI”, the German Network for Bioinformatics Infrastructure, a BMBF infrastructure initiative aimed at expanding and improving the availability of hardware, data resources and bioinformatics tools in the life sciences, and securing them in the long term. During the first funding phase, now completed, the project partners developed the concept that they will implement from 2015 onwards. The Tübingen bioinformaticians will form a kind of network-within-a-network with colleagues from Konstanz and Berlin with the goal of making bioinformatics tools suitable for application in several "omics" technologies.  

Kohlbacher explains the attraction of such an achievement: “An individual “omics” technology only provides insights into one particular aspect of a cell, i.e. its genes, proteins, metabolome, etc. Many biological processes can only be understood by combining several “omics” technologies. Such complex analyses require the parallel reduction of data on several “omics” levels before we can look at the whole.” Such processes require researchers to work according to a structured scheme, a workflow. The researchers from Konstanz provide the technology for such workflows and Berlin is in charge of next-generation sequencing tools. Tübingen brings proteomics and metabolomics tools into the project. The goal is to make combined analyses reproducible. “The complexity of modern analyses makes it increasingly difficult to understand what has been done and why. In many cases, there is no documentation stating which tool has been used with which parameters. We want to change this by developing suitable documentation structures,” says Kohlbacher. The Tübingen team and the other de.NBI partners will be funded by the BMBF for a period of five years during which researchers will pool and adapt existing expertise and resources in this area in order to support the sustainable provision of bioinformatics services and infrastructure.

Further information:
University of Tübingen
Applied Bioinformatics, Centre for Bioinformatics
Prof. Dr. Oliver Kohlbacher
Sand 14
72076 Tübingen
Tel.: +49 (0)7071 29-70457
E-mail: oliver.kohlbacher(at)uni-tuebingen.de

Website address: https://www.gesundheitsindustrie-bw.de/en/article/news/bioinformaticians-shed-light-into-the-data-jungle