Computational Biology
Description
The group worked on the definition of models and interrogation languages for the representation and study of biological systems. In particular, formalisms such as Hybrid Automata, Stochastic Automata, Process Algebras, Dynamical Systems used as parameters for system modelling, Temporal and Spatial Logics, Bigraphs, and Observational Equivalences for the specific property of interest have been taken into account. Specific algorithms were conceived for the accurate or approximate verification of such properties in the models. The algorithms were implemented in several tools, among which we'll mention pyHybridAnalysis, SAHA-Tool, Sapo.
The group studied the problem of the prediction of the tertiary structure of a protein by using techniques based on constraint (logic) programming, at first employing simplified energy models on lattice spaces, then leaving the lattice and using peptide fragments whose typical positions are stored in databases. The recombination of possible fragments is discretely delimited by the available fragments in a space that is potentially non-discrete. A dedicated model, denoted 5@ (5 atoms per amino acids) was developed, together with its statistical energy potentials. Dedicated constraints were studied and implemented within the scope of a constraint solver, this too having been produced by the group. In the implementation, the parallelism of programmable graphic processing units (GPGPU) was used.
The group developed algorithms for the assembling and analysis of new-generation sequencing data. In particular, algorithms were developed for global assembling (entire higher-species genomes) and for local assembling (regions included within the paired ends of a single sequence). Also, the group studied the problem of combining different known assemblies for the same organism/species, and developed algorithms for aligning sequences treated with bisulphite (for epigenomic studies). Finally, methods were studied for the alignment of sequences on compressed indexes not requiring decompression of the text.
The group was involved in organising many conferences, workshops and international schools of systems biology and bioinformatics, including BITS, HSB, WCB, MecBIC, BCI, FMMB, CMSB. The group members have been guest editors and editors of international magazines covering this field. They have participated in national and international projects and have well-established international collaborations with NYU Bioinformatics Group, NMSU Center for Bioinformatics and Computational Biology, Institut de Biologie Paris Seine.
Research subjects
- Modelling of biological systems, including models for studying the development of some cancer types
- Identification of time/space logics and observational equivalences for the formal verification of the properties of biological models
- Definition of algorithms for assembling and representing NGS data
ERC panels
- PE6_13 Bioinformatics, bio-inspired computing, and natural computing
- PE6_4 Theoretical computer science, formal methods, automata
- PE6_7 Artificial intelligence, intelligent systems, natural language processing
- PE6_6 Algorithms and complexity, distributed, parallel and network algorithms, algorithmic game theory
Tags
- Systems Biology, Hybrid Systems, Protein Folding, Model Checking
- Automated Reasoning, Sequence Assembly Algorithms, Biological Sequences Compression