A collaborative web application for supporting researchers in the task of generating protein datasets

ARMANO, GIULIANO;MANCONI A.

2011-01-01

---

Abstract

The huge difference between known sequences and known tertiary structures has fostered the development of automated methods and systems for protein analysis.When these systems are learned using machine learning techniques, the capability of training them with suitable data becomes of paramount importance. From this perspective, the search for (and the generation of) specialized datasets that meet specific requirements are prominent activities for researchers. To help researchers in these activities we developed ProDaMa-C, a web application aimed at generating specialized protein structure datasets and fostering the collaboration among researchers. ProDaMa-C provides a collaborative environmentwhere researcherswith similar interests can meet and collaborate to generate new datasets. Datasets are generated selecting proteins through user-defined pipelines of methods/operators. Each pipeline can also be used as starting point for building further pipelines able to enforce additional selection criteria. Freely available as web application at the URL http://iasc.diee.unica.it/prodamac , ProDaMa-C has shown to be a useful tool for researchers involved in the task of generating specialized protein structure datasets.