Prediction of protein secondary structure at better than 70% accuracy.

By Burkhard Rost, Christopher Sander. 1993

Subfields: Computational Biology & Bioinformatics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Biology & Bioinformatics

1. Jinbo Xu
2. Yang Zhang
3. David T. Jones
4. Sheng Wang
5. Burkhard Rost
6. David Baker
7. Jianlin Cheng
8. Björn Wallner

Citations in the corpus

• Modeling aspects of the language of life through transfer-learning protein sequences (2019, methodology): “Per-residue performance: Toward this end, we used the standard three-state per-residue accuracy (Q3 = percentage correctly predicted in either helix, strand, other [2]) along with its eight-state analog (Q8).”
• Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction (2014, methodology): “Most significant improvement to secondary structure prediction has been achieved by leveraging evolutionary information by using sequence profiles from multiple-sequence alignment (Rost & Sander, 1993) or position-specific scoring matrices from PSI-BLAST (Jones, 1999).”
• ProtTrans: Towards Cracking the Language of Life’s Code Through Self-Supervised Deep Learning and High Performance Computing (2020, background): “New protein representations may complement existing solutions, most successful when combining evolutionary information and machine learning [34], [69], [70], [71].”
• rawMSA: End-to-end Deep Learning Makes Protein Sequence Profiles and Feature Extraction obsolete (2018, background): “The sequence profile describes the available evolutionary information of the target protein and is better than a single sequence representation, often allowing for a significant increase in the prediction accuracy [31,32].”
• Application of multiple sequence alignment profiles to improve protein secondary structure prediction (2000, methodology): “17,25,23,13,6 In our previous study of algorithms that use multiple sequences as the basis for prediction, neural network prediction methods were found to be the most accurate.”
• Toward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networks (2014, methodology): “In the absence of a reliable solution to the protein structure prediction problem, some research groups have focused on solving simplified problems such as the prediction of protein structural features [12-14].”
• What makes an Escherichia coli promoter σS dependent? Role of the −13/−14 nucleotide promoter positions and region 2.5 of σS (2001, methodology): “Protein secondary structure prediction Prediction of protein secondary structure was performed using algorithms described previously (Rost and Sander, 1993; 1994; Rost et al., 1994) with the help of a server provided by EMBL (http://www.embl-heidelberg.de/ predictprotein/). b-Galactosidase…”
• Constraint Logic Programming approach to protein structure prediction (2004, background): “Secondary structure can be predicted with good approximation (e.g., [36]).”
• High Quality Prediction of Protein Q8 Secondary Structure by Diverse Neural Network Architectures (2018, background): “ wide variety of statistical approaches [5, 12, 8, 15, 13, 20]. To ease the prediction challenge, these studies typically merged the eight DSSP classes to only three. In the early 90s Rost and Sander [26, 27], augmented protein sequences by profiles, derived from multiple sequence alignment of homolo”
• Deep learning methods in protein structure prediction (2020, methodology): “The third generation of PSA predictors has been characterised by the adoption of evolutionary information [56] in the form of alignments of multiple homologous sequences as input to the predictive systems, which are almost universally Machine Learning, or Deep Learning algorithms.”