Combining evolutionary information and neural networks to predict protein secondary structure

By Burkhard Rost, Christopher Sander. 1994

How researchers see this paper

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

Citations in the corpus

ProtTrans: Towards Cracking the Language of Life’s Code Through Self-Supervised Deep Learning and High Performance Computing (2020, methodology): “New protein representations may complement existing solutions, most successful when combining evolutionary information and machine learning [68, 69, 34, 70].”
A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction (2014, methodology): “Neural networks have been effectively used in a variety of prediction algorithms, including speech and face recognition as well as the aforementioned usage in protein secondary structure prediction[11-22, 24, 33-36].”
HMMSTR: a hidden Markov model for local sequence-structure correlations in proteins. (2000, methodology): “…some improvement in prediction quality might be produced by a postprocessing step analogous to the second layer in the PHD secondary structure prediction method, which reassigns the predicted secondary structure at a position based on the predictions at surrounding positions (Rost & Sander, 1994).”
Comprehensive review of methods for prediction of intrinsic disorder and its molecular functions (2017, methodology): “Output: Predicted disordered residues from each selected input predictor and the consensus score for each residue in the input sequence; predicted secondary structure with PROFsec [102] and PSIPRED [97]; predicted secretion signal peptides with SignalP [103]; predicted transmembrane regions with TMH”
The human genome encodes 10 α-crystallin–related small heat shock proteins: HspB1–10 (2003, background): “The secondary structure was predicted from the 10 aligned sequences with PHD (Rost and Sander 1994); predicted ß strands are numbered according to the approximate location of the corresponding ß strands in the crystal structures of M jannaschii Hsp16.”
ProtTrans: Towards Cracking the Language of Life’s Code Through Self-Supervised Learning (2021, methodology): “Secondly, we compared the effects of auto-regressive and auto-encoding pre-training upon the success of the subsequent supervised training, and compared all LMs trained here to existing state-of-the-art (SOA) solutions using evolutionary information (EI) [40].”
DFLpred: High-throughput prediction of disordered flexible linker regions in protein sequences (2016, unknown): “The two conservation scores are derived using ClustalX (Thompson et al., 1997) and PHDsec (Rost and Sander, 1993; Rost and Sander, 1994), and the hydrophobicity score is assigned using the Kyte and Doolittle’s hydropathy index (Kyte and Doolittle, 1982).”
Better prediction of sub‐cellular localization by combining evolutionary and structural information (2003, background): “…Bank of experimentally determined 3D structures of proteins; 88 PHD, proﬁle-based neural networks for predicting secondary structure (PHDsec) 62,66,54 , and solvent accessibility (PHDacc); 53,54 PredictNLS, prediction of nuclear proteins through nuclear localization signals; 22,75 HSSP,…”
Length-dependent prediction of protein intrinsic disorder (2006, methodology): “In addition, three secondary structure prediction scores by the PHDsec predictor [50] and another three by the PSIPRED predictor [48] were also included.”