Protein Structure Prediction: Selecting Salient Features from Large Candidate Pools

Weintroduce a parallel approach, "DT-SELECT," for selecting features used by inductive learning algorithms to predict protein secondary structure. DT-SELECTis able to rapidly choose small, nonredundant feature sets from pools containing hundreds of thonsands of potentially useful features. It does this by building a decision tree, using features fromthe pool, that classifies a set of training examples. The features included in the tree provide a compact description of the training data and are thus suitable for use as inputs to other inductive learning algorithms. Empirical experiments in the protein secondary-structure task, in which sets of complex features chosen by DTSELECT are used to augment a standard artificial neural network representation, yield surprisingly little performance gain, even though features are selected from very large feature pools. Wediscuss somepossible reasons for this result.1

• Type: Conference
• Year: 1993
• Where: ISMB
• Authors: Kevin J. Cherkauer, Jude W. Shavlik
• Keywords: Computational Biology, Inductive Learning Algorithms, ISMB 1993, Pools Containing Hundreds, Protein Secondary Structure