Overview
We evaluated several ensemble methods, including novel ensembles over standard benchmarks consisting of CASP 10-13 and random subsets of proteins deposited into PDB between 2016 and 2019.
The most accurate ensemble method we tested is a hybrid ensemble between Nnessy and Porter.
Accuracy
On average, the hybrid exceeds the state-of-the-art 3-state accuracy by nearly 4%, and exceeds the 8-state prediction by more than 8%.
Full accuracy measures are given in our paper.
Publication
The methods implemented in Ssylla are given in the following publication, which should be cited under noteworthy use of Ssylla
- Spencer Krieger and John Kececioglu, “Predicting protein secondary structure by an ensemble through feature-based accuracy estimation.” Proceedings of the 11th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB), 29, 1-10, 2020.