View publication

Using genetic algorithms to select most predictive protein features.

Kernytsky A, Rost B,
Proteins (2009) 75:75-88 PublishedPSI:Phase 2  
Northeast Structural Genomics Consortium

(click to unfold)
Many important characteristics of proteins such as biochemical activity and subcellular localization present a challenge to machine-learning methods: it is often difficult to encode the appropriate input features at the residue level for the purpose of making a prediction for the entire protein. ...
metabolism chemistry methods 
Algorithms Databases, Protein Models, Molecular Proteins Computational Biology Protein Conformation Structure-Activity Relationship Serine Endopeptidases Computer Simulation Neural Networks (Computer) 
18798568  
10.1002/prot.22211  
16 (Last update: 11/10/2018 5:45:47pm)  
method