Perspective on “Der Spiegel: Germany’s most expensive experiment”

The European X-ray Free-Electron Laser, a joint enterprise involving Germany, Russia and nine other European partner nations costing more than 1.2B euro was the lead article in this week’s edition of der Spiegel, a widely-read weekly news periodical published in Hamburg. A face that is very familiar to Melbourne physicists featured prominently in the article: Dr Adrian Mancuso, a former PhD student of Professor Keith Nugent, is now Group Leader and Leading Scientist at the European XFEL.

Adrian was interviewed by der Spiegel on his personal perspectives regarding the development of the Facility and its program to use X-ray pulses to determine the structures of isolated biomolecules. Keith is also deeply embedded within European XFEL as a member of its International Scientific Advisory Committee. He is coordinating the Centre’s efforts across the Physics nodes towards contributing to the goals of obtaining images of single molecules using XFEL sources and, ultimately, real-time movies of molecules undergoing structural and chemical changes with unprecedented spatial and temporal resolution. Another Melbourne alumnus, Professor Henry Chapman, is a Centre Principal Investigator and Director of the Centre for Free-Electron Laser science at Hamburg that is developing many of the techniques described by der Spiegel. The Physics Group of the Centre has also embedded a researcher at the European XFEL, Dr Marjan Hadian, in collaboration with our partners in ANSTO. She will be involved with analysing the data from the very first experiments that take place at the facility, sometime after September of this year.

The article in der Spiegel likens the entire process of building the European XFEL and using it to obtain images of molecules to the fall of a sequence of dominos; each tile must be in its proper place in order for the triggering event of the first tile to produce the cascade of events that lead to the formation of an image in the final tile.  Once one has gained access to the facility, that “first tile” may be found in the village of Bahrenfeld, where electrons are started on their 3.4 km journey to the experimental hall in Schenefeld in the neighbouring region of Schleswig-Holstein.  From the immunologists’ perspective, a separate cascade of biochemical processes has allowed them to produce, purify and deliver a suitable sample, which is the target tile sitting in Schenefeld. But this is not the “final tile”. After the data are collected the image may well be produced anywhere in the world and that final tile may well fall in a far distant laboratory, including one in the Centre for Advanced Molecular Imaging in Australia, where the analysis of XFEL data is performed.  As part of the experimental program, researchers in the Centre Physics program at La Trobe University are developing methods that will allow movies to be made of samples which interact with the intense X-ray pulses, in effect, synchronizing two separate domino cascades to meet perfectly at a single tile. The theoretical physics program of the Centre at the University of Melbourne has the task of analysing data from the distribution of targets that will be imaged, each producing its own separate cascade of information. The domino analogy in der Spiegel accurately captures the character of the program and the many sequences of events that have to be combined to form the complete cascade, from the production and acceleration of electrons, the generation of X-ray pulses, the preparation of samples and their delivery into the X-ray beam, the interactions between targets and pulses which finally ends with the collection and analysis of data to produce images.

Is all this effort really worth it? Of course, der Spiegel also asks whether the price tag, which is largely being met by German tax-payers really justifies the huge investment and whether the project will succeed in its goals. As a counterbalance to the optimistic view of our friend and colleague Adrian Mancuso, der Spiegel also sought the opinion of Professor Dr Holger Stark, Director of the Max-Planck Institute for Biophysical Chemistry in Göttingen. Stark determines the structures of biomolecules to near-atomic resolution using cryo-electron microscopy, a trend which we can also find reflected in the research activities of members of our own Centre.  In assessing the potential of XFEL sources for biomolecular imaging, Stark identified two main shortcomings, one of which was physical and the other practical, both of which are of significance to our Centre.

In measuring scattering using XFEL sources one measures intensity but, as Stark correctly points out, the phase information is lost, just as it is in conventional crystallography. What Stark apparently is not aware of, however is that this scientific lemon was turned into lemonade long ago with the development of coherent diffractive imaging, which not only allows the phase to be determined uniquely from scattering measurements, but also the image to be determined from a completely random starting guess, completely eliminating any structural bias in its determination. This is a stronger statement than can be made for cryo electron microscopy, for which the choice of an “unbiased” starting guess remains a critical issue. Members of the Centre physics program have made seminal contributions to the development of coherent diffractive imaging over the past decade, including CIs Nugent, Quiney and Abbey and Senior Research Associate Timur Gureyev.

Stark’s practical criticism of the XFEL program is that he is already able to determine molecular structures using laboratory-based equipment costing around 10M Euro, so that the European XFEL is likely to cater mainly for “unusual” problems that cannot be addressed in any other way. It is already the case, however, that LCLS, the XFEL source at Stanford in the US, has opened new frontiers in biophysical and material science by making accessible the time-dependence of chemical processes and structural changes. By its design, cryo-EM is not a convenient technique for following processes that occur on chemical timescales, so rather than seeing these two techniques as direct competitors, they should instead be regarded as complementary tools in structure and dynamics

And what of the future? Precisely the same diversity of opinion about the future of molecular imaging is currently circulating within the Centre as is expressed in the article in der Spiegel. It ranges from the infectious enthusiasm of Adrian Mancuso for molecular movies in native aqueous environments to a desire to add to Stark’s snapshot album of static structures of molecules entombed in vitreous ice. As the old Chinese proverb says, however, “when men speak of the future, the gods laugh” and we would be hostages to fortune to predict how things will evolve. Perhaps it is simply best that we thank our good fortune for being in a well-funded Centre in which we may each play a major role in deciding that future. We’ll all have a good laugh when it arrives.