I would like to come to the last part, some conclusions and prospectus of the project. First of all, I'd like to make clear what the project is not. The project is not an endocrine disruptor screening project. We are using endocrine disruption because endocrine disruption is a test case. It's an interesting test case, but it is all about sharpening our tools. It is not a test development project. It is about creating a point of reference similar to the human genome, which is not to test, something that is a data which we can refer to and similarly, we hope to create with the human toxome a point of reference, a repository of agreed pathways of toxicity. It is also not an academic publish-or-perish project. This project necessarily focused very much on quality control as you have seen. And this is counterproductive for the fast and dirty publications which are typically done by academia. And it is not a five-year project. So it is necessary to move on. The ground has been laid. We have a new partner for followup applications with the OECD, the Organization for Economic Cooperation and Development. They are entertaining the effect of pedia database and we are at the moment trying to find ways of feeding our human toxome knowledge-based plans and merge, by doing so, the adverse outcome pathway and the pathway of toxicity concept. We had visibility. I mentioned many of these activities before and these journals and articles. There are many more. This is just showing that there was international interest in running into this project. It is altogether a contribution to the big data approaches in toxicology. The Omics technologies in our case contributing to a new paradigm of generating and mining large data sets. We are trying to combine it more recently also with high-content imaging approaches and we had another workshop to develop these concepts. We've been dealing a lot with the questions of untargeted metabolomics, its pros and cons, and it has some pros. It is a very sensitive technology and once you have a machine, it is very inexpensive compared to omics technologies. It is close to phenotype, shows very little species differences most metabolites are the same in mouse, rat, and humans. We only measuring thousands of parameters, not tens and hundreds thousands, and we have the knowledge and biochemical pathways which are connecting metabolites But also we have problems, we have small effect strengths by the gene expression often is tenfold, hundred fold increased 20 percentmore of a metabolite is often already a dramatic change. Very often flux through the system is more important than the absolute change. The changes are fast. We have incomplete extractions and measurements so we don't have full metabolom. They only measure parts of the metabolom. We have tremendous differences still is the identification of metabolites especially for mass spectra and quality assurance has been put forward, but it is still incomplete and other important take home message is, it is dramatically difficult to make a big sense of big data, and if the data are trash, whatever analysis we're applying it will be trashed again. This holds true for both of a cell-culture systems and the animal models which we use as a reference data. If you use the wrong analysis tools, and I showed you some, we are actually multiplying the problem, and if you combine this then with data-rich noise for technologies like omics technologies this is getting even worse. There's big challenges, but it is the way forward, and it is something we're, at least there is some light as we hope at the end of the tunnel. The big challenge will be how to quality assure pathways, whether we call them adverse outcome pathways or pathways of toxicity. How do we validate our findings. These are things we are lacking still in concepts. The entire concept of pathway toxicity and 21st-century technology use are corresponding very closely. The pathways of toxicity are mainly mechanistic toxicology multi-omics, high-throughput screening, and some other type of technology of the 21st century. But there's also the same technologies or similar technologies both on both sides of the pathways of toxicity. The exposure metabolite makes the measurement of metabolites in the environment, in the blood of people exposed all the responsiveness, the biomarkers of response on the side of individual responses and populations, these are things which are at the moment coming and the name of human exposome has been coined for these two types of measurements, and that's a revolution of exposure sciences. I just think that there's actually quite a bit of correspondence between the systems especially when we're moving towards biomarker identification and making sense of these biomarkers because we can help to interpret the biomarker changes with the mechanistic understanding of toxicology. Quite interestingly, D Sarigiannis is one of the leading experts in exposome research. He shared with me this slide, and you can see here very nicely the correspondence of ideas between the two sides. They are suggesting to start with untargeted gene area and metabolism. They took up our concepts and terminology of a pathway of toxicity, and they then want to follow up this targeted type of measurements and understanding these concepts. The long term vision is actually to create systems biology understanding. A level where these pathways help us to make sense of what is happening in the entire organism system's toxicology. This is a short definition on what the systems biology, in short it says, to produce big data from high content measurements then mining literature systematically and modeling doing virtual type of experiments, that's exactly the workflow we have in mind moving from data generation to understanding this data as the databases and systematic literature, analysis, and then moving to systems toxicology-type of approach. This is certainly up high in the sky. This is nothing. We have reached yet, but I hope I've shown you in this lecture today that technologies of the 21st century, which allow us a different type of understanding of what is happening in a perturbed organism after exposure to a toxicant. We would like to close, Mr. Maynard Keynes, an economist who said very wisely, "The difficulty lies, not in the new ideas, but in escaping from the old ones." Hope you took some new ideas from this lecture, and I hope that you will be part of escaping from the old ideas. Thanks a lot.
