There is rarely any thought spent on the fact that not only there is no structural similarity between some amino acids but that also the genetic distance is enormous, which may be defined as number of necessary base changes in the codon. For the first time I noticed this in my thesis, when the design of a protein by an amino acid exchange, far from being likely on the basis of natural variations in the protein class, gave excellent results. Also in immunology it appears to be a common problem that amino acids cannot easily mutate from one to another, although such drastic changes could have beneficial effects for vaccines and diagnostics.

After postdoctoral years in structural groups Uppsala and Berlin he joined in 1995 the department of pharmaceutical pharmacology at the University Uppsala to work on membrane receptors and the establishment on peptide phage display technologies for selection on whole cells. From 1998 to 2003 he was CSO of a German company dedicated to novel methods for peptide phage display. Thereafter he was in management positions in related drug discovery biotech companies. In 2009 he became the head of a new group dedicated to ligand development at the Fraunhofer Institute for Cell Therapy and Immunology in Leipzig, Germany. Here he has been turning his attention again to basic developments in peptide phage display. The main interests of the group are techniques to map immune responses in whole blood and serum as well as the discovery of peptide ligands to cell surfaces and the methods required for their application.

Η ΑΜΑΕ είναι γρήγορη και μπορεί να εντοπίσει σχεδόν όλα τα πιθανά αντιγόνα, ανεξάρτητα από τη συγκέντρωσή τους, την ώρα της έκφρασης και της ανοσογονικότητας. Σε αυτό το σεμινάριο, θα συζητηθεί η κύρια ιδέα της ΑΜΑΕ, η βιοπληροφορική πρόκληση και η δυνατότητα εφαρμογής μεθοδολογίας μηχανικής μάθησης με σκοπό την κατανόηση της γενομικής και πρωτεομικής υπογραφής υποψήφιων αντιγόνων για την ανάπτυξη εμβολίων.

Ως φοιτητής στο Πανεπιστήμιο του Cambridge ήταν επίσης Cambridge European Trust Fellow και μέλος της Πανεπιστημιακής Ομάδας Κανό. Πριν από αυτό, ήταν μέλος του εργαστηρίου Βιοφυσικής και Βιοπληροφορικής του Πανεπιστημίου Αθηνών. Πριν από την εγγραφή του στο Πανεπιστήμιο Αθηνών σπούδαζε Μηχατρονική στο Εργαστήριο Εφαρμοσμένης Μηχανικής στο Πολυτεχνείο Κρήτης. Σήμερα εργάζεται ως Ανώτερος Μηχανικός Λογισμικού στην ιδιωτική εταιρεία, Qualco SA, με έδρα την Αθήνα, συνεχίζοντας παράλληλα τη συνεργασία του με ερευνητικές ομάδες στο The Wellcome Trust Sanger Institute. Είναι κριτής για τα ακόλουθα περιοδικά: Bioinformatics, Nucleic Acids Research, Molecular Biology and Evolution. Είναι ο συγγραφέας δύο αυτόνομων λογισμικών βιοπληροφορικής: GeneViTo (http://bioinformatics.biol.uoa.gr/GENEVITO) και alien_hunter (http://www.sanger.ac.uk/resources/software/alien_hunter/). Έγινε γνωστός στην επιστημονική κοινότητα για την ανάπτυξη της καινοτόμου θεωρίας των "Ενσωματωμένων Μεταβλητού Μεγέθους Μοτίβων" (Interpolated Variable Order Motifs), ιδιαιτέρως χρήσιμη για την νουκλεοτιδική (πολυδιάστατη) ανάλυση αλληλουχιών DNA. Ερευνητικά ενδιαφέροντα: Μηχανική μάθηση (Hidden Markov Models, Relevance Vector Machines), ροή γενετικού υλικού και εξέλιξη μικροβιακών γονιδιωμάτων.

We confirmed as strongly associated with hypoxia in BC, miRNAs known to be regulated by hypoxia in cell lines, such as miR-210, -122a, -452* and the -23/24/27 clusters. Furthermore, we identified 11 new miRNA associated with hypoxia; some of which were validated by cell line experiments. A comprehensive data-mining was performed to discover miRNAs strongly and independently prognostic for distant-relapse after correction for pathway gene signatures and clinical covariates. We identified 4 in ER-positive and 9 in ER-negative BCs; of these, miR-210, -27b and -150 were strongly prognostic also in triple-negative BCs. Pathways associated with prognostic miRNAs included proliferation, immune response, invasion and hypoxia; confirming in several cases reports from functional studies.
Finally, in some cases these results could be confirmed at the level of gene expression of the precursor and/or predicted targets. For example, in the case of miR-27b, associated with invasion and member of hypoxia-related miR-23/24/27 clusters, the mean expression of targets predicted by 6 published databases was prognostic in univariate and multivariate analyses of distant-relapse in 3 independent large BC series.
In conclusion, integrated multivariate data-mining of miRNA and mRNA profiles in a large clinical series could identify independent prognostic miRNAs associated with key BC progression pathways.

His work has covered the physical mapping and gene identification in the human Major Histocompatibility Complex. In 1992 he became Lecturer in Molecular Genetics at the Division of Medical Molecular Genetics of the United Medical and Dental Schools of Guy?s and St. Thomas London. He became a Senior Lecturer in 1995. In 2001 he moved to the University of Oxford working at the Wellcome Trust Centre for Human Genetics, where he is Head of Genomic Research. He is also University Reader in Genomics. His work covers the application of microarray and next generation sequencing technologies in genetics, epigenetics and regulation of transcription. Since 2010 he has been elected Researcher (Professor Level) in Genomics and Director of the Genomics Facility at BSRC Al Fleming in Athens.

This revolution is enabled with breakthroughs in technologies and computer sciences that are together leading to an incredible increase in complex information, over time, for millions of individuals. This information must be accessed, analyzed and interpreted in real time to provide effective proactive health care. Analysis of this information, together with parallel studies in model organisms, is creating a deep understanding of the function of biological systems as well as discovery of new ways to maintain health. I will summarize the key elements of this ongoing revolution, and will provide examples of the ways in which an understanding of genetics, systems biology and information sciences can be used to prevent diseases such as cancer and obesity.

He was a founding member of the International Mammalian Genome Society and a founding editor of Mammalian Genome and of Systems Biology and Medicine. He was founder and director of the Mouse Genome Informatics Project and founder of the Mouse Gene Expression Database Project. He has served on review panels and advisory groups at the National Institutes of Health, the National Science Foundation and the Human Genome Database, and has consulted for several biotech and major pharmaceutical companies. His research interests include cancer, metabolic disease and development, with an emphasis on genetic, genomic, computational, bioinformatics, and systems studies of mouse models of human disease. He has won several awards for his work, is an Elected Fellow of the American Association for the Advancement of Science, and was recently recognized with a Pioneer Award from the National Institutes of Health. He is currently Director of Research and Academic Affairs at the Institute for Systems Biology.

Nature being efficient, protein docking can be formulated as a the problem of minimizing the Gibbs free energy of the complex. Optimization is performed over translations and rotations of the ligand with respect to the receptor (a nonlinear manifold), as well as, over conformational changes (especially side-chains at the interface). However, the free-energy functional is very complex having multiple deep funnels and a huge number of local minima of less depth that are spread over the domain of the function. We present a systematic multi-stage method for performing this optimization. The entire conformational space of ligand translations and rotations is explored using simplified energy potentials to produce clusters of promising conformations. Optimization of side chains is formulated as a combinatorial optimization problem for which we propose a new distributed algorithm based on graph-theoretic ideas. The energy landscape in the space of ligand translations and rotations is explored using dimensionality reduction approaches, which reduces the domain of further optimization.
Finally, cluster refinement is done using a new stochastic global optimization method we have developed, the so called Semi-Definite programming based Underestimation (SDU) method. We will discuss the algorithms, provide convergence guarantees, comparisons with related work, and an array of computational results illustrating our approach.

He obtained a Diploma (1991) from the National Technical University of Athens, and an M.S. (1993) and a Ph.D. (1996) from the Massachusetts Institute of Technology (MIT), all in Electrical Engineering and Computer Science.
In September 1996 he joined Boston University where he has been ever since. His current research interests lie in the fields systems and control, networking, applied probability, optimization, operations research, computational biology, and bioinformatics.
Prof. Paschalidis' work on communication networks has been recognized with a CAREER award (2000) from the National Science Foundation and the second prize in the 1997 George E. Nicholson paper competition by INFORMS. He was an invited participant at the 2002 Frontiers of Engineering Symposium, organized by the National Academy of Engineering. His recent work on protein docking has been recognized by a 1st prize in the Protein Interaction Evaluation Meeting (2007) and an invitation to a select workshop at the Institute for Mathematics and Its Applications (IMA) on Molecular and Cellular Biology (2008). He has served in the program/organization committees of many conferences, has been a past associate Editor of the IEEE Trans. on Autom. Control and of the Operations Research Letters, and is currently an associate editor of the SIAM Journal on Control and Optimization.