ERC FUNDED PROJECTS

The proposal is to look for, and investigate, new ergodic theoretic types of behavior for dynamical systems which act on non compact spaces. These could include transience and non-trivial ways of escape to infinity, critical phenomena similar to phase transitions, and new types of measure rigidity. There are potential applications to smooth ergodic theory (non-uniform hyperbolicity), algebraic ergodic theory (actions on homogeneous spaces), and probability theory (weakly dependent stochastic processes).

539 479 €

Start date: 2009-10-01, End date: 2014-09-30

The aim of this project is to try to understand the complex molecular mechanisms involved in DNA editing, repair and recombination during immunoglobulin class switch recombination (CSR) and somatic hypermutation (SHM). We have developed a series of PCR-based assays to study in vivo generated CSR junctions and the pattern of mutations introduced in the immunoglobulin variable region genes in human B cells, allowing us to characterize CSR and SHM in patients with immunodeficiency due to defect(s) in DNA repair/recombination. Novel in vitro CSR assays, based on GFP expression, allowing quantitative measurement of substrate recombination, are also being developed. In addition, we have initiated an evolutionary analysis of the function and structure of activation-induced deaminase, an essential molecule involved both in CSR and SHM, aiming to identify CSR specific-cofactor(s). Combining these approaches, we will be able to define the DNA repair pathways involved in CSR and SHM. The suggested project requires access to patients with various defects in the DNA repair pathways. Many of these diseases are exceedingly rare. However, through worldwide collaboration, we have obtained samples from a majority of the diagnosed patients. We are also refining the existing screening methods and developing novel methods, that will allow identification of additional patients both with recognized and new diseases caused by mutations in DNA repair pathways. Finally, we hope to be able to address the question whether illegitimate CSR events are associated with predisposition to lymphomagenesis in patients with immunodeficiency/DNA repair defect(s), by analyzing the CSR induced chromosomal breaks and translocations in these patients. A large-scale sequencing project is also planned to characterize the CSRnome in B-cell lymphoma samples.

1 888 166 €

Start date: 2009-12-01, End date: 2014-11-30

Data mining provides large benefits to the commercial, government and homeland security sectors, but the aggregation and storage of huge amounts of data about citizens inevitably leads to erosion of privacy. To achieve the benefits that data mining has to offer, while at the same time enhancing privacy, we need technological solutions that simultaneously enable data mining while preserving privacy. The current state of the art has focused on providing privacy-preserving solutions for very specific problems, and has thus taken a local perspective. Although this is an important first step in the development of privacy-preserving solutions, it is time for a global perspective on the problem that aims for providing full integrated solutions. Our goal in this research is to study privacy and develop comprehensive solutions for enhancing it in the digital era. Our proposed research project includes foundational research on privacy, an infrastructure level for achieving anonymity over the Internet, key cryptographic tools for constructing privacy-preserving protocols, and development of large-scale applications that are built on top of all of the above. The novelty of our research is in our focus on fundamental issues towards comprehensive solutions that are aimed for large-scale data sources. The project s outcome will allow migration from local solutions for specific problems that are suited for small to medium scale data sources to comprehensive privacy-preserving database and data mining solutions for large scale data warehouses. Achieving this great challenge carries immense scientific, technological and societal rewards.

1 921 316 €

Start date: 2009-10-01, End date: 2014-09-30

The project described in this proposal studies formal proofs and their interaction with computation. The study of propositional proofs is connected to a spectrum of problems in our field, starting with the meta-mathematical quest to explain our failure to understand computation and make progress on the basic questions haunting our field (such as P vs. NP), and ending with the industry-driven quest for better algorithms for solving instances of the satisfiability problem. In a seemingly different direction, the recent introduction of magical probabilistically checkable proofs (PCPs) has opened new horizons in computer science, ranging from a deeper understanding of approximation algorithms and their limits to the construction of super-efficient protocols for the verification of proofs and computations. We suggest to study proofs and computation with three main objectives. First, to construct better SAT solvers via a better understanding of propositional proof systems. Second, to expand the range of applications of PCPs and transform them from the purely theoretical objects that they currently are to practical and accessible formats for use in all settings where proofs are encountered. Third, to expand our theoretical understanding of the intrinsic limits of proofs, with an eye towards explaining why we are unable to make significant progress on central questions in computational complexity. We believe this project can bridge across different regions of computer science such as SAT solving and proof complexity, theory and practice, propositional proofs and probabilistically checkable ones. And its expected impact will start on the theoretical mathematical level that forms the foundation of computer science and percolate to more practical areas of our field.

1 743 676 €

Start date: 2009-12-01, End date: 2015-09-30

We aim to study transcriptomes with single-cell resolution, a long-standing goal in biology, to answer fundamental questions about gene regulation. The main objective concerns gene regulation during in vivo differentiation and in pluripotent cells by studying single-cells from murine preimplantation embryos, a model system with natural single-cell resolution, important biology and medical potential. This would also allow us to explore general regulatory principles of gene expression programs of individual cells. This research program will be accomplished by novel deep sequencing technology of mRNAs (mRNA-Seq) to obtain quantitative, unbiased and genome-wide gene and isoform expression measurements. We are therefore developing new experimental and computational methods for genome-wide analyses of transcriptomes at single-cell resolution. The biological significances of the proposed research are unique insights into early embryonic development. Deep sequencing of transcriptomes will also reveal post-transcriptional gene regulation important for pluripotent cells and identified pluripotency-specific gene and isoform expressions will be important for future stem cell based therapies. The inherit single-cell nature of the model system together with its important biology makes it a model systems exceptionally well suited for a systems biology approach aiming to characterize gene regulation at single-cell resolution. The novel methodology has tremendous potential to enable complete mRNA characterization of individual cells. The deep sequencing approach with state-of-the-art computational analyses is both more quantitative than previous methods and it will give readouts on alternative isoforms generated by alternative promoters, splicing and polyadenylation.

1 654 384 €

Start date: 2010-02-01, End date: 2015-01-31

I propose to develop sophisticated anti-tumor agents targeted particularly to the location of activity. My team has recently introduced a new family of Ti(IV) complexes that demonstrates higher activity than known compounds with substantially higher stability and defined hydrolytic behavior, properties that were found to be essential. I propose to study various derivatives and identify the parameters affecting activity, including steric and electronic effects, enantiomeric purity, ligand lability etc., and elucidation various mechanistic aspects of reactivity. More importantly, I propose to construct pH-sensitive transport units that will allow protection of the sensitive active species throughout their delivery and release only near the target location based on the variable pH conditions of different human tissues. In particular, unique spherical molecules held together by metal-ligand interactions will be prepared. The building blocks will consist of the planar ligands of C3-axis bound to three biocompatible Ti(IV) ions each with defined angles and geometry. The resulting spherical compounds will be utilized to encapsulate the active complexes and release them upon hydrolysis at the desired pH based on the pH-dependent hydrolysis pattern already established for related compounds. Preliminary calculations have confirmed the possibility of forming these compounds, which are particularly matching in their expected size to encapsulate our complexes. Larger spheres will also be prepared as cavities for larger molecules, which may be linked together for the delivery of multiple drugs. These compounds may find applications in various areas where a protected environment or delivery of sensitive compounds is required, such as in gene therapy, nano-technology, and catalysis.

1 400 000 €

Start date: 2009-10-01, End date: 2014-09-30

Probabilistically Checkable Proofs (PCPs) encapsulate the striking idea that verification of proofs becomes nearly trivial if one is willing to use randomness. The PCP theorem, proven in the early 90's, is a cornerstone of modern computational complexity theory. It completely revises our notion of a proof, leading to an amazingly robust behavior: A PCP proof is guaranteed to have an abundance of errors if attempting to prove a falsity. This stands in sharp contrast to our classical notion of a proof whose correctness can collapse due to one wrong step. An important drive in the development of PCP theory is the revolutionary effect it had on the field of approximation. Feige et. al. [JACM, 1996] discovered that the PCP theorem is *equivalent* to the inapproximability of several classical optimization problems. Thus, PCP theory has resulted in a leap in our understanding of approximability and opened the gate to a flood of results. To date, virtually all inapproximability results are based on the PCP theorem, and while there is an impressive body of work on hardness-of-approximation, much work still lies ahead. The central goal of this proposal is to obtain stronger PCPs than currently known, leading towards optimal inapproximability results and novel notions of robustness in computation and in proofs. This study will build upon (i) new directions opened up by my novel proof of the PCP theorem [JACM, 2007]; and on (ii) state-of-the-art PCP machinery involving techniques from algebra, functional and harmonic analysis, probability, combinatorics, and coding theory. The broader impact of this study spans a better understanding of limits for approximation algorithms saving time and resources for algorithm designers; and new understanding of robustness in a variety of mathematical contexts, arising from the many connections between PCPs and stability questions in combinatorics, functional analysis, metric embeddings, probability, and more.

1 639 584 €

Start date: 2009-09-01, End date: 2016-06-30