Aims of the Project
Injury of the adult central nervous system of mammals results in lasting deficits including permanent motor and sensory impairments due to lack of profound neuronal regeneration. In particular, patients suffering from spinal cord injury (SCI) remain paralyzed for the rest of their lives and often suffer from additional complications. Preclinical research in the field of central nervous system trauma is advancing at a fast pace and yields over 8,000 new publications per year growing at an exponential
rate, yielding a total amount of approximately 160,000 PubMed-listed papers today. However, translational neuroscience faces a strong disproportion between the immense preclinical knowledge available on the one hand, and the lack of successful clinical trials in spinal cord injury therapy on the other hand.
By automatically analyzing scientific literature on a large scale by state-of-the-art information extraction techniques, we can assess the level-of-evidence and robustness of therapies such that a translation of an empirically reliable treatment for spinal cord injury into a clinical trial can be proposed with a high chance of success. In particular, we hypothesize that the prospective success of a potential therapy can be to some extent predicted by determining its likelihood of being successfully transferred between different organisms, and eventually to humans. This insight can be exploited to empirically derive a scoring function that predicts the likelihood of successful transferability of a therapeutic approach across organisms. If this hypothesis turns out to be true, this scoring function can be used to assess therapies with respect to their prospective success in order to identify promising SCI therapies.
As the central goal of this project, we aim at estimating the prospective success of therapeutic approaches to spinal cord injury treatment and pave the way for the design of a clinical trial. To reach this goal, all available knowledge about preclinical experiments scattered in scientific publications has to be analyzed.
We will develop an information extraction workflow for automatically structuring this knowledge and storing it in a database covering all published preclinical experiments on SCI treatment. Being made available to clinical and preclinical researchers, this database will support the selection of the most promising therapeutic setting for clinical trials or animal experiments, help to compare their
results to previous experiments with similar settings or specifically altered parameters. Thus, it will enable meta-analyses and systematic comparisons on the basis of all scientific data available in the scientific literature in order to identify aspects of preclinical studies which increase the probability for therapies to be successfully translated across species, with the final goal to translate them into humans.