Dr. Grabe, you work in the field of medical systems biology, or MSB, which is still quite a new discipline that aims to understand biological organisms in their entirety. How did this field of research emerge and what makes it so special?
From a historical point of view, after all the initial enthusiasm about the sequencing of the human genome, a sense of disappointment set in when it was discovered that individual diseases generally can’t actually be linked to individual genes. Instead, it became increasingly apparent that the decisive factor was the interplay between genes and their products, such as RNA and proteins. For example, gene A and B interact with gene C, which also interacts with gene D. To get a grasp of the way these structures function, recourse was taken to networks, as a formal method – in the same way they’ve been used for a long time to analyze complex systems. The idea that an individual gene could correlate with a disease stems from a highly simplistic view of the world, driven by the reduction of elements to a technical function – which, from today’s standpoint of course, has turned out to be wrong. Looking back, the pioneers of computing like Turing, McCulloch and Pitts were already referring to complex biological systems as networks as early as the 1940s. So complex biological systems can only be captured by using complex theoretical models, and that is what systems biology does. This applies in particular to medical systems biology which sees a causal link between human conditions and misdirected networks – something which can be captured through complex measurement. So systems biology is highly complex. It spans a broad spectrum of methods rooted in bioinformatics, mathematics, high-throughput technology, cell biology, but also – if not especially – medical expertise.
At your Steinbeis Transfer Center for Medical Systems Biology (MSB), you apply systems biology methods to cancer research and look at the quantitative analysis of biomarkers. What are your aims with this?
Our aim is to help our partners make better use of the potential offered by systems biology in order to conduct research into new diagnostic methods and treatments. We also want to transfer our recent results from research into practice. Systems biology is closely, although not exclusively, based on high-throughput data from sources like sequencing. By evaluating this data it becomes possible to pinpoint new areas that tumors may attack. This is particularly useful as we can ascertain this individually by patient. So it’s possible that there are completely different optimum scenarios for different patients with a certain type of tumor. These have to be identified and treatments have to be developed for each individual patient.
Another area is the development of computer models for certain types of tissues like the skin, for which we’ve already published initial models. One highly interesting issue is the connection between vitro models from the lab and the quantitative high-throughput analysis of biomarkers. There are now some pretty good tissue models on the market, for things like the skin. These can be used as research tools, which can be put to excellent use to test substances systematically and with high throughputs. The changes these substances bring about can then be ascertained indirectly by observing changes in morphological appearance, for example using histological sections. Working systematically, high-throughput evaluations of histological sections and the subsequent processing of images provide us with detailed insights.
What do you think will happen in the future in the field of systems biology? What challenges will this pose for researchers in the years to come?
After the initial phase of mild euphoria, systems biology is going through a period of consolidation in which it has to demonstrate its practical potential. The concept is understood – and certainly, further development will be crucial. From where we stand at the moment, “humongous” volumes of data are being generated and these are sure to grow in the future. So we’re currently only considering this as an initial step in the direction we’ll actually be going eventually. My prediction is that the universities will not be in a position to put the organizational and technical infrastructures in place to deal systematically with this data in the long term – because of the lack of research funding. Industrial pharmaceutical research is taking place, but, in essence, this is only on paper. So smaller companies will bridge the gap and piece together the necessary technology. This will probably happen mainly in the United States and it’ll be an exciting development.
Another area your Steinbeis Enterprise focuses on is digital pathology. What does that involve?
Digital pathology is about the transformation of pathology – from what it’s been in the 150 years since Rudolf Virchow into a digital discipline. The advent of high-throughput scanners has made it possible to fully automate the process of examining objects on glass slides. Sectional images of patient tissue samples can be used to produce pictures for pathological analysis. Digital imaging is already standard clinical practice in radiology but that’s not the case in pathology, which is currently undergoing a sea change. Unlike radiology, in pathology, computer-aided image processing systems are increasingly being used to acquire complex morphological attributes from patient samples which can then help the diagnostic process. In some areas, these will increasingly replace the manual analysis carried out by pathologists. For example, in my research team, we’ve developed the first system for recognizing cervical cancer fully automatically. This is done by using smears, p16 markers and digital pathology. But this isn’t just important for diagnostic reasons. It is also very valuable for research and clinical trials of new substances. Digital pathology is also particularly useful in the field of systems biology research to systematically obtain quantitative tissue data. Confirming the impact of substances – effectively, objectively and quantitatively – is not just about the costs incurred by the pharmaceutical company, it’s more crucial in terms of the time it takes to make new treatments quickly available to patients, especially with potentially deadly conditions like cancer.
Associate professor Dr.-Ing. Niels Grabe, based at the Steinbeis Transfer Center for Medical Systems Biology (MSB) at the University of Heidelberg, conducts quantitative assessment of biomarkers in the field of cancer research, carries out systems biology modeling of pathological processes in tissues, and is involved in digital pathology.
Associate professor Dr.-Ing. Niels Grabe
Steinbeis Transfer Center for Medical Systems Biology (Heidelberg)
su1745@stw.de