Design and practical usage of web biological databases for the annotation and classification of proteins
- Autores: Toni Hermoso Pulido
- Directores de la Tesis: Francesc Xavier Avilés Puigvert (dir. tes.), Enrique Querol Murillo (dir. tes.)
- Lectura: En la Universitat Autònoma de Barcelona ( España ) en 2015
- Idioma: catalán
- Tribunal Calificador de la Tesis: Rafael de Llorens Duran (presid.), David Reverter Cendrós (secret.), Angel Mozo Villarias (voc.)
- Materias:
- Enlaces
- Resumen
In the so-called Information society, data represent a key structural parts of knowledge generation. Likewise, present-day Bioinformatics ultimately relies on the proper management and processing of data originated from both traditional and high-throughput biological analyses.
As a primary aim of this thesis, different practical approaches for the handling of raw protein sequence data and their applied resulting Bioinformatics analyses are compiled. Nonetheless, as a preliminary step, pre-existing Bioinformatics tools and algorithms are adapted to the characteristics of current informational paradigm, basically revolving around the World Wide Web. After a proper adaptation of those applications (in this document: ProtLoc, TransMem, TranScout and Bypass), it becomes possible to face the massive processing of protein data originating from the last genome sequencing projects.
The outcomes of these analyses are made available for the world-wide scientific community in the form of user-friendly and fully accessible web-based biological databases. As examples, two cases are presented: TrSDB, a compendium of well-known and putative transcription factors from different model organisms, and ArchDB, a structural web database of protein loops.
As a further step, starting from Bypass program ¿a fuzzy-logic based tool for the re-annotation and evaluation of protein homology search alignments¿, a complete annotation and sequence management framework is deployed. As as strong point, the system is also flexible and modular enough for allowing the input of different data (e. g. Gene Ontology) or cross-communicate with other future and existing applications.
Parallely to this, and also taking advantage of the explosion of raw sequence data and database curation, a Bioinformatics characterization of a new metallacarboxypeptidase family is introduced. By using computational means, it was possible to present a first hypotheses for the enzymatic activity and phylogenetical history of these group of proteins. Notably, their actual identification may represent an enlightening focus for dealing with biological processes such as malaria or neurodegenerative disorders, where these molecules are intimately linked.
