Team:Heidelberg/Interviews

Expert
Interviews


Prof. Dr. Kevin Esvelt

Director of the Sculpting Evolution group

Assistant Professor, MIT Media Lab

Massachusetts Institute of Technology

Speaking to Kevin Esvelt, the inventor of PACE, was a real pleasure. We had the great opportunity to host Kevin Esvelt for a lecture at our university and when he visited our iGEM lab he provided extremely helpful advice not only on how to set up PACE technically, but also on how to design corresponding experiments safely and responsibly and how to communicate our project outside the scientific community.
Kevin Esvelt made us aware of possible challenges we would meet when running PACE, for instance, he mentioned that phage-washouts in the lagoon are a problem he faced several times when setting up PACE in his own lab for the first time. Therefore, we developed a real-time PCR readout to test, whether phages are still present in the lagoon and if not, to re-inoculate to keep the evolution going.

Antonio D´Isanto

PhD student in Astroinformatics

HITS – Heidelberg Institute for Theoretical Studies

Antonio D´Isanto works at the Heidelberg Institute for Theoretical Studies and applies Deep Learning to determine probabilistic photometric redshifts in astrological images of stars. Antonio emphasized the advantages of convolutional neural networks for automated feature extraction, and suggested an application on protein sequences. Furthermore, Antonio gave us an introduction on the general working principle of convolutional neural networks, and served as a key contact for technical discussions throughout the whole project.

Dr. Dorothea Kaufmann

Study Program Manager for Molecular Biotechnology

Institute of Pharmacy and Molecular Biotechnology

University of Heidelberg

As Human Practice is one of our core projects we invested a lot of time in, we thought it is reasonable to get some advice from Dr. Dorothea Kaufmann who herself campaigns for open science and actively encourages discussions about controversial topics of gene technology. Dr. Kaufmann appreciated, supported and expanded our Human Practice concept in many respects. For instance, she encouraged us to actively invite and face the broad public with synthetic biology issues. Dr. Kaufmann enriched our work with her motto: "science affects everybody and therefore, everybody should be able to affect science". Following up this idea, we invited everybody for an open lecture with Kevin Esvelt, presented our project at TEDx, visited a school, and hosted a high school class for some DNA experiments in our iGEM lab. We received diverse encouragements during all these events, which struck new paths for our project. Furthermore, Dr. Kaufmann highlighted the importance of our PACE simplification PREDCEL. According to her statement, that methods should be cheap enough to be used by every scientist, who wants to contribute to the questions in synthetic biology, for which it is is essential to invest time in simplifying processes. Good science should be transparent, method descriptions should be easily accessible, and experiments should be reproducible. With our standardized evolutionary toolbox, the simplified setup PREDCEL and our inter-lab study for mutagenesis plasmid activity evaluation, we would like to contribute to open science for everybody.

Thomas Wollmann

PhD Student and Co-Founder of PW solutions

Department Bioinformatics & Functional Genomics

Heidelberg University

Thomas Wollmann is a PhD student at the Department for Bioinformatics & Functional Genomics and Co-Founder of PW solutions. He works on the applicability of Deep Learning methods for large scale segmentation and detection tasks in biomedical image data. He recommended us to work with the Tensorflow framework and gave us substantial advice how to handle it. In addition, he had significant contribution to the implementation of our neural network architecture. We integrated his input in our DeeProtein software tool.

Prof. Dr. Michael Wink

Academic Dean of the Institute for Pharmacy and Molecular Biotechnology

Heidelberg University

Prof. Michael Wink is one of the most experienced professionals we talked to. It was a great pleasure for us to be profiting from his store of knowledge in the field of evolution. Our concerns that directed evolution approaches like PREDCEL and PACE might be too ambitious and too complex for a student project were alleviated by Prof. Wink. He gave us intellectual input and helped us with recommendations about existing literature by which we were able to translate our ideas to the laboratory “bench”. As he pointed out that silicon besides carbon-based life is a possibility on other planets, we feel encouraged to pursue our research on enzymes that form carbon-silicon bonds.

Prof. Dr. Dr. Edith Huland

Founder & Managing Director of Immunservice GmbH

Medical Professional

Prof. Huland developed an inhalative interleukin-2 immunotherapy and is an international expert on the development and implementation of therapeutic and diagnostic innovations. Her impressive career comprises the responsibility as principal investigator and co-investor for various international clinical trials as well as scientific advisor for international government, university and industrial advisory boards. Giving her immense expertise, we discussed our directed evolution approach with Prof. Huland. She was impressed about the implications that the PACE and PREDCEL technology could offer in therapeutic inventions and research. In fact, she immediately came up with many creative and innovative ideas about proteins that could be evolved to benefit therapeutic applications.

Prof. Dr. Philipp Stoellger

Theological Department

Heidelberg University

Should humanity harness the process of evolution for their own benefit? In fact, humanity is doing so since thousands of years by breeding animals and cultivating plants for their own demands. However, transferring a directed evolution approach to the laboratory could raise concerns in the general public.
We wanted to discuss our directed evolution approach with experts in the field of interdisciplinary ethics and philosophy of religion. Hence, we talked to Prof. Dr. Philipp Stoellger from the theological department of Heidelberg University. Prof. Stoellger pointed out that he, as a theologist, has no general concerns about intervening in evolution and harnessing its potential for human benefits as long as the experiments are kept under strict laboratory conditions. Nature is a natural field of cultural practices. However, ethical questions arise, if powerful and efficient techniques like PREDCEL and PACE are misused for military, economically or potentially harmful purposes or if modified organisms are released into the natural ecosystem. It is the duty of each and every scientist to reflect about the consequences of their research. Could this new discovery or invention be misused in any way? Which safety measures and precautions can be taken to prevent misuse?
To face ethical concerns about directed evolution approaches, Prof. Stoellger recommended to firmly integrate professionals from other fields into such substantial projects and proposals, like hermeneutics, ethics, theologies or even artists and politicians. This is reasonable because only an interdisciplinary group of professionals can make decisions with consequences affecting all humanity. His message to the natural sciences is: professionalize your unprofessional limits of research, and make use of professional competences where you are not competent. That message may affect your experimental constellations, your methods and your theory of science up to the members of research groups. Prof. Stoellger was impressed about the concept of iGEM and how this competition in Synthetic Biology encourages young scientists to reflect on safety, environment and the involvement and contribution of the general public to the design and elaboration of the scientific project. Following this concept, young scientists would grow to become responsible researchers in the future.

PD Dr. Ullrich Köthe

Visual Learning Lab

Heidelberg Collaboratory for Image Processing and

When we considered to use Artificial Intelligence (AI) for our software project for the first time, we wanted to ensure that we do not endanger anybody with this powerful, but controversial technology. Interviewing high school students, we learned about their fears regarding the use of AI. Dr. Ullrich Köthe elucidated all potential risks, and we discussed dystopian predictions where AI dominates the entire earth or universe and turns all available matter and energy into “Computronium”. To answer the question whether these dystopian predictions might come true one day, he referred to the onset of gene technology: People pointed out severe risks thirty years ago, but science has found effective counter-measures to keep these risks under control. The current discussion about AI is in many ways similar. Informing us about leading experts who anticipate computers to achieve general intelligence or even super-intelligence, he especially referred to Ray Kurzweil. Kurzweil predicts that “by 2029, computers will have human-level intelligence” and "set the date 2045 for singularity — which is when humans will multiply our effective intelligence a billion fold, by merging with the intelligence we have created” - Dr. Köthe considers this estimation to be quite optimistic. He points out that we cannot predict AI-related threats and dangers far into the future, because we have no idea how computers at this level of intelligence will work. Current dystopian ideas are mainly extrapolations of human weaknesses, which are unlikely to apply to intelligent machines in the same way. Existing (weak) AI systems usually make mistakes that are markedly different from typical human errors in the same situation. Therefore, it is necessary to advance our understanding of AI goal control and safety in close association with our improving understanding of AI itself. For the time being, AI will remain unable to cause severe trouble, unless actively misused by a human. All in all, Dr. Köthe quietened down our concerns and general fears of dystopian AI and made us feel more relaxed using AI for our software approach.

Dr. Michael Ernst

Director of the Agricultural Department

University of Hohenheim

To figure out potential agricultural applications for PACE and PREDCEL we talked to Dr. Ernst from the University of Hohenheim. He indicated that the ability of evolving proteins with specific, novel properties RN35, would undoubtably be of high interest for the agricultural industry. However, he expressed his concerns when it comes to acceptance utilizing gene technology for agricultural production by customers, organizations for customer protection as well as policy makers. Dr. Ernst made us aware of the fact, that a lack of acceptance of gene technology in agriculture in Europe is in fact a very fundamental problem present since several decades. In parts, this problem is, as Dr. Ernst says, due to a lack of information required for the stakeholders in the community to fairly weight potential benefits and risks. Dr. Ernst thus recommended an open communication of research in genetic engineering as well as close collaboration with Ethics Boards and organizations for customer protection. The PACE and PREDCEL approaches developed by our team during this iGEM summer allows evolving and tuning protein-protein interactions. Thereby, proteins can be created, that highly specifically bind to and affect harmful species and reprieve related harmless species. Hereby, biodiversity can be maintained and is even supported compared to conventional approaches. On top, we got the chance to discuss our carbon-silicon approach with Dr. Ernst. He was fascinated by our ideas and claimed, that applications of carbon-silicons have a high demand for usage as fungicides or insecticides by farmers and in the horticultural business.

Dr. Lorenz Adlung

PhD in Systems Biology

Weizmann Institute of Science, Israel

Dr. Lorenz Adlung is an expert in the field of Systems Biology, as he is working on mathematical models of biological systems. He strongly encouraged us to extend our directed evolution approach in silico and to combine our work in the wet lab with the knowledge we gain from our software GAIA. The work in the laboratory greatly benefits from the digital revolution that we are experiencing in the 21th century. In his future vision, an intense interaction between living and modeled systems heralds a new revolution in biological science. As a former iGEM participant, he is aiming to bring cutting-edge science to the general public. Therefore, he joins open forums and panel discussions on a regular basis and explains scientific discoveries on several platforms in social media. His immense outreach inspired our team to bring science to public and to integrate the public opinion into our research aim.

Birgit Degner

Judge

Ordinary Jurisdiction North Rhine-Westphalia, Germany

We contacted the experienced judges Birgit Degner and Dr. Jens Degner to get their legal opinion on our research project. They were not able to give us a legal assessment, because our project is far too complex and based on too many different fields of law to find a quick answer. Instead, they explained to us their personal opinion.
From their point of view, scientific research should be as unrestricted as possible. This also implies that a crucial role is assigned to the scientist: Every scientist has to prove if his or her research is compatible with their own moral values and has to make sure that all necessary safety measures and precautions are taken. Scientists are connected to and responsible for their research, but in the end the consumers can and should use their market power to give a feedback if the findings are compatible with general ethical values. In addition, ethical review committees can lead discussions, clarify research topics, and help to determine moral values.
They concluded that science should not be limited unnecessarily, because the ultimate goal of research is the improvement of the general quality of life for every human.

Dr. Ingo Janausch

Representative Biological Security and Gene Technology

Department for Biological Safety

Heidelberg University

Directed evolution experiments lead to altered DNA sequences and proteins could evolve towards toxic and hazardous sequences. Before evolving sequences in our PACE and PREDCEL experiments, we wanted to confirm that no hazardous products can evolve. We had the idea to establish a SafetyNet, which screens the input sequence for potential hazardous outcomes. Therefore, each sequence that was used in our experiments was first compared to a predefined list with dangerous sequences using our SafetyNet. We discussed with Dr. Janausch, which protein classes should be included in the list of hazardous sequences and he recommended us to test the relation to oncogenes too. Therefore, we edited our SafetyNet to integrate oncogenes as potentially harmful sequences. We discussed the different safety aspects and Dr. Janausch concluded, that PACE and PREDCEL experiments should be save when tested in our SafetyNet first and discussed with our supervisors before starting the experiment. Indeed, he recommended that we integrate our SafetyNet to our toolbox guide, since every researcher that wants to evolve proteins using PACE or PREDCEL should reflect on the outcome of their research. With the input of Dr. Janausch we could improve our Safety Net and hope, that other scientists apply the safety recommendations we developed with Dr. Janausch for their PACE or PREDCEL experiments too.

Dr. Jens Degner

Judge

Ordinary Jurisdiction North Rhine-Westphalia, Germany

We contacted the experienced judges Birgit Degner and Dr. Jens Degner to get their legal opinion on our research project. They were not able to give us a legal assessment, because our project is far too complex and based on too many different fields of law to find a quick answer, but they gave us a personal opinion.
From their point of view, science should be as unrestricted as possible, but by doing this, the scientist is assigned a crucial role: Each scientist needs to think if his research is compatible with their own moral values and make sure that all necessary safety measures and precautions are taken. The scientist is connected to and responsible for the research done, but in the end the consumers can and should use their market power to see if the findings are compatible with general ethical values. Ethical commissions can lead discussions, clarify research topics and help finding moral values.
They concluded that science should not be limited unnecessarily, because the ultimate goal is the improvement of the general quality of life for every human.

Prof. Dr. Erika Musterfrau

Head of Department Musterpatterns

Pattern Scientist

Dies ist ein Typoblindtext. An ihm kann man sehen, ob alle Buchstaben da sind und wie sie aussehen. Manchmal benutzt man Worte wie Hamburgefonts, Rafgenduks oder Handgloves, um Schriften zu testen. Manchmal Sätze, die alle Buchstaben des Alphabets enthalten - man nennt diese Sätze »Pangrams«. Sehr bekannt ist dieser: The quick brown fox jumps over the lazy old dog. Oft werden in Typoblindtexte auch fremdsprachige Satzteile eingebaut (AVAIL® and Wefox™ are testing aussi la Kerning), um die Wirkung in anderen Sprachen zu testen. In Lateinisch sieht zum Beispiel fast jede Schrift gut aus. Quod erat demonstrandum. Seit 1975 fehlen in den meisten Testtexten die Zahlen, weswegen nach TypoGb. 204 § ab dem Jahr 2034 Zahlen in 86 der Texte zur Pflicht werden. Nichteinhaltung wird mit bis zu 245 € oder 368 bestraft. Genauso wichtig in sind mittlerweile auch Âçcèñtë, die in neueren Schriften aber fast immer enthalten sind. Ein wichtiges aber schwierig zu integrierendes Feld sind OpenType-Funktionalitäten. Je nach Software und Voreinstellungen können eingebaute Kapitälchen, Kerning oder Ligaturen (sehr pfiffig) nicht richtig dargestellt werden. Dies ist ein Typoblindtext. An ihm kann man sehen, ob alle Buchstaben da sind und wie sie aussehen. Manchmal benutzt man Worte wie Hamburgefonts, Rafgenduks oder Handgloves, um Schriften zu testen. Manchmal Sätze, die alle Buchstaben des Alphabets enthalten - man nennt diese Sätze »Pangrams«. Sehr bekannt ist dieser: The quick brown fox jumps over the lazy old dog. Oft werden in Typoblindtexte auch fremdsprachige Satzteile eingebaut (AVAIL® and Wefox™ are testing aussi la Kerning), um die Wirkung in anderen Sprachen zu testen. In Lateinisch sieht zum Beispiel fast jede Schrift gut aus. Quod erat demonstrandum. Seit 1975 fehlen in den meisten Testtexten die Zahlen, weswegen nach TypoGb. 204 § ab dem Jahr 2034 Zahlen in 86 der Texte zur Pflicht werden. Nichteinhaltung wird mit bis zu 245 € oder 368 bestraft. Genauso wichtig in sind mittlerweile auch Âçcèñtë, die in neueren Schriften aber fast immer enthalten sind. Ein wichtiges aber schwierig zu integrierendes Feld sind OpenType-Funktionalitäten. Je nach Software und Voreinstellungen können eingebaute Kapitälchen, Kerning oder Ligaturen (sehr pfiffig) nicht richtig dargestellt werden. Dies ist ein Typoblindtext. An ihm kann man sehen, ob alle Buchstaben da sind und wie sie aussehen. Manchmal benutzt man Worte wie Hamburgefonts, Rafgenduks oder Handgloves, um Schriften zu testen. Manchmal Sätze, die alle Buchstaben des Alphabets enthalten - man nennt diese Sätze »Pangrams«. Sehr bekannt ist dieser: The quick brown fox jumps over the lazy old dog. Oft werden in Typoblindtexte auch fremdsprachige Satzteile eingebaut (AVAIL® and Wefox™ are testing aussi la Kerning), um die Wirkung in anderen Sprachen zu testen. In Lateinisch sieht zum Beispiel fast jede Schrift gut aus. Quod erat demonstrandum. Seit 1975 fehlen in den meisten Testtexten die Zahlen, weswegen nach TypoGb. 204 § ab dem Jahr 2034 Zahlen in 86 der Texte zur Pflicht werden. Nichteinhaltung wird mit bis zu 245 € oder 368 $ bestraft. Genauso wichtig in sind mittlerweile auch Âçcèñtë, die in neueren Schriften aber fast immer enthalten sind. Ein wichtiges aber schwierig zu integrierendes Feld sind OpenType-Funktionalitäten. Je nach Software und Voreinstellungen können eingebaute Kapitälchen, Kerning oder Ligaturen (sehr pfiffig) nicht richtig dargestellt werden. Dies ist ein Typoblindtext. An ihm kann man sehen, ob alle Buchstaben da sind und wie sie aussehen. Manchmal benutzt man Worte wie Hamburgefonts, Rafgenduks.