What followed was the first generation of SS predictors, all based on exploiting statistical propensities of single AA towards specific SS conformations 6, 7, 8, 9. It was inaugurated by Pauling and Corey in 1951, when they predicted the existence of the two most common SS conformations – α-helix and β-sheet – before the first protein structure was fully determined 5. One of the most enduring open problems in Bioinformatics is Secondary Structure (SS) prediction 3, 4. Nonetheless, nearly 150,000 protein structures are now freely available, growing by roughly 10,000 per year, making proteins a central research topic in Bioinformatics 2. While we discovered more than 325 million protein sequences 1, we still lack a feasible method to experimentally fully characterize them at a large scale. Porter is available as a web server and standalone program at alongside all the datasets and alignments.įrom DNA repair to enzyme catalysis, proteins are the chief actors within the cell. When Porter 5 is retrained on SCOPe based sets that eliminate homology between training/testing samples we obtain similar results. In our tests Porter 5 is 2% more accurate than its previous version and outperforms or matches the most recent predictors of secondary structure we tested. Porter 5 achieves 84% accuracy (81% SOV) when tested on 3 classes and 73% accuracy (70% SOV) on 8 classes on a large independent set. Porter 5 is composed of ensembles of cascaded Bidirectional Recurrent Neural Networks and Convolutional Neural Networks, incorporates new input encoding techniques and is trained on a large set of protein structures. In this study we present tests on different models trained both on single sequence and evolutionary profile-based inputs and develop a new state-of-the-art system with Porter 5. In spite of this, even the most sophisticated ab initio SS predictors are not able to reach the theoretical limit of three-state prediction accuracy (88–90%), while only a few predict more than the 3 traditional Helix, Strand and Coil classes. Protein Secondary Structure prediction has been a central topic of research in Bioinformatics for decades.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |