LMU professors Thomas Carell (Department of Chemistry), Hinrich Schütze (Center for Information and Speech Processing), Ludger Klein (Faculty of Medicine) and Jonathan Harrington (Institute for Phonetics and Speech Processing) have all won Advanced Investigator Grants in the ERC’s latest funding round. Indeed, this is the second such award for Jonathan Harrington.
ERC Advanced Investigator Grants are each worth up to 2.5 million (in exceptional cases as much as 3.5 million) euros over 5 years, and are intended for established researchers in all disciplines who wish to undertake highly ambitious and innovative projects that promise to break new ground in their respective fields.
Proposals submitted by researchers based at LMU have been notably successful in the ERC’s highly competitive funding programs since the agency was set up in 2007. In fact according to the latest “Annual Report on the ERC’s Activities and Achievements”, LMU has received more ERC grants than any other university in Germany.
The newly funded projects:
Professor Thomas Carell is Professor of Organic Chemistry at LMU, and his research focuses on the biochemistry of the so-called epigenetic code – a set of chemical modifications of the canonical nucleosides that make up the universal genetic code, induced by environmental factors and by endogenous alterations in the cell’s differentiation state.
In his new ERC project, “The Chemical Basis of RNA Epigenetics”, Carell will explore how and why organisms chemically modify the nucleoside subunits of the nucleic acids DNA and RNA, focusing in particular on messenger RNAs (mRNAs). Such modifications act as a supplementary layer of information, which helps to determine which of the genes encoded in the cell’s DNA genome are active, or susceptible to activation, at any given time. The epigenetic code thus enables the organism to regulate the activity of its genes in a highly flexible manner. This second level of information involves the site-specific chemical modification of the nucleoside subunits that make up both DNA and RNA molecules. About 150 nucleoside variants are now known but, according to Carell, there may well be many more. His goal is to identify as yet uncharacterized modifications and determine their biological functions. The new project promises not only to yield new insights into the structures and functions of modified nucleosides, but could also shed new light on the origin of the genetic code itself. According to the so-called RNA world hypothesis, the era of biological evolution was preceded by a phase of chemical evolution, in which RNA and atypical modified bases are thought to have played an important role. Last year, in a paper that appeared in the journal Science, Carell and his colleagues described a novel reaction mechanism that could have given rise to purines (one of the two types of nucleosides used in the universal genetic code) under prebiotic conditions on the early Earth.
Thomas Carell was born in Herford in 1966. He studied at the University of Münster and obtained his PhD at the Max Planck Institute for Medical Research in Heidelberg. He was a Postdoctoral Fellow in Professor Julius Rebek’s group at the Massachusetts Institute of Technology in Cambridge from 1993 until 1995, before moving to the ETH in Zürich to set up his own research group. He was appointed to a professorship at the Marburg University in 2000, and took up his present position at LMU in 2004.
Professor Jonathan Harrington is Professor of Phonetics and Digital Speech Processing and Director of the Institute of Phonetics and Speech Processing at LMU. He has now become one of the very first researchers anywhere in the world who have won two Advanced Grants from the ERC. His research interests include experimental phonetics and phonology, the development of language databases and shifts in intonation.
In his second ERC project, Harrington will study how accents evolve through interactions between speakers, and how accents themselves contribute to language diversification. In essence, Harrington wants to understand how changes to individual accents affect the accent of the language community. The goal is to develop a computational model that can predict how random interactions between individuals can give rise to group-specific accents. In order to construct and constrain the model, Harrington and his colleagues will analyse how children's accents develop when they are in schools in relatively isolated communities with a homogeneous accent compared with those in an urban setting and with a mixture of accents. In addition, they will look at what happens to accents in groups that spend months at a time in areas remote from their normal surroundings – such as researchers who spend the winter in Antarctica. The resulting agent-based computer model will elucidate how dialects and accents evolve through interactions between individual members of language communities, and how spoken accents when they become isolated from each other can give rise to languages with markedly different sound patterns. The project will also trace the long-term impact of increased migration on language change and diversification.
Jonathan Harrington was born in England in 1958, and studied at the University of Cambridge (Downing College), where he also obtained his doctorate in Linguistics. He went on to take up positions at the University of Edinburgh and at Macquarie University in Sydney. In 2002, he was appointed to the Chair of Phonetics and Digital Speech Processing at the University of Kiel, before taking up his present position at LMU in 2006. Harrington received his first ERC Advanced Investigator Grant in 2011.
Professor Ludger Klein is at the Institute of Immunology (Faculty of Medicine) at LMU. His research focuses on the workings of the adaptive immune system, which is responsible for fighting infection by recognizing and eliminating invasive pathogens. This in turn requires the capacity to distinguish invaders (non-self) from the body’s own cells (self), which is based on the phenomenon of self-tolerance. Failure of this self-tolerance mechanism results in autoimmune diseases such as multiple sclerosis. Klein is interested in how immunological tolerance is maintained among the body’s repertoire of so-called T cells, which originate and mature in the thymus. Every day this lymphoid organ produces millions of T cells, each of which carries a specific type of antigen receptor that recognizes surface structures on other cells. Since these receptors are generated by a tightly controlled but essentially random genetic recombination mechanism, many of them are capable of binding to ‘self’ structures and triggering the destruction of the cells that carry them.
In his ERC project, Ludger Klein intends to study how these potentially harmful (autoreactive) T-cells are disarmed. The immune system actually uses two quite distinct mechanisms to accomplish this task. One is based on the elimination of such cell clones by the activation of programmed cell death (clonal deletion). The other works by reprogramming them to take on the role of so-called regulatory T-cells (clonal diversification). Klein’s primary goal is to identify the factors that determine the fate of autoreactive T-cells. This will involve the application of new methods for the selection of rare antigen-specific T-cells and characterization of the antigen receptors on individual T cells before and after their maturation in the thymus. By comparing the repertoires that survive in the presence and absence of specific self-antigens, one can infer which autoreactive T-cells undergo elimination or diversification and identify features that influence their respective fates.
Ludger Klein (b. 1965) became Professor of Cellular Immunology at LMU in 2007. He studied Biology at the University of Cologne and wrote his Diplom thesis at the Max Planck Institute for Plant Breeding Research in the same city. Klein also obtained his PhD from Cologne University in 1997 for research carried out at the German Center for Cancer Research in Heidelberg. Following a stint (1997-1999) as a postdoc in Bruno Kyewski’s group in Heidelberg, he joined Harald von Boehmer’s lab at the Dana Farber Cancer Institute in Boston for 3 years (1999-2002), before moving to the Institute of Molecular Pathology in Vienna as head of his own research group devoted to T-cell tolerance.
Hinrich Schütze was named Professor of Computational Linguistics and Director of the Center for Information and Speech Processing in the Faculty of Languages and Literatures at LMU in 2013. His primary research interest is Statistical Natural Language Processing.
In his ERC project Schütze plans to develop the technology of Natural Language Processing (NLP) with a view to improving communication between humans and machines. Computer programs that make use of NLP are not only capable of deciphering, analyzing and translating texts and spoken language they can also generate entirely new texts. Nevertheless, computers still lag far behind humans in their ability to decode natural language. This is particularly true in the case of languages that are morphologically more complex than English, of texts on social media platforms which employ dialect or non-standard vocabulary and slang, or are peppered with more-or-less deliberate errors, and of videos and sound recordings with high levels of background noise.
The major aim of Schütze’s project is to develop a new approach for the automatic analysis of natural language. While the conventional approach to NLP uses fixed rules to break up text into elements such as syllables, words and phrases, Schütze’s strategy will make use of methods for the recognition of the semantic content of character combinations based on their context. This strategy exploits recent advances in deep learning, and promises to significantly facilitate human-machine communication.
Hinrich Schütze was born in 1964 in Celle in Lower Saxony. He studied at the Technical University in Braunschweig and at Stuttgart University, and obtained his PhD at Stanford University in California. He then spent 5 years at the nearby Xerox Palo Alto Research Center 5 further years in Silicon Valley, where he worked on search engines and methods of text mining. In 2004 he became Professor of Computational Linguistics in Stuttgart, before moving to his present position at LMU in 2013.