Alfredo Benso

TL;DR: The second critical assessment of functional annotation (CAFA), a timed challenge to assess computational methods that automatically assign protein function, was conducted by as mentioned in this paper. But the results of the CAFA2 assessment are limited.

TL;DR: The second critical assessment of functional annotation (CAFA) conducted, a timed challenge to assess computational methods that automatically assign protein function, revealed that the definition of top-performing algorithms is ontology specific, that different performance metrics can be used to probe the nature of accurate predictions, and the relative diversity of predictions in the biological process and human phenotype ontologies.

TL;DR: The third CAFA challenge, CAFA3, that featured an expanded analysis over the previous CAFA rounds, both in terms of volume of data analyzed and the types of analysis performed, concluded that while predictions of the molecular function and biological process annotations have slightly improved over time, those of the cellular component have not.

TL;DR: The second Critical Assessment of Functional Annotation (CAFA) challenge as mentioned in this paper was the first attempt to assess computational methods that automatically assign protein function. And the results of CAFA2 showed that computational function prediction is improving.

TL;DR: In this paper, a comprehensive guide to fault injection techniques used to evaluate the dependability of a digital system is presented, along with a critical analysis of different fault injection tools and techniques.