Estimating probabilistic context-free grammars for proteins using contact map constraints

Structural annotation of proteins using probabilistic context-free grammars

Large scale structural annotation is essential to systems biology, since it contributes to functional annotation and clustering of proteins. Although the most general approach to structure annotation relies on de novo structure prediction [1], it is most often impractical due to very high demand for computational resources. Therefore, efficient methods based on sequence profiles have been propo...

Conditional Probabilistic Context-Free Grammars

Probabilistic Context-Free Grammars for Phonology

We present a phonological probabilistic contextfree grammar, which describes the word and syllable structure of German words. The grammar is trained on a large corpus by a simple supervised method, and evaluated on a syllabification task achieving 96.88% word accuracy on word tokens, and 90.33% on word types. We added rules for English phonemes to the grammar, and trained the enriched grammar o...

Statistical Properties of Probabilistic Context-Free Grammars

We prove a number of useful results about probabilistic context-free grammars (PCFGs) and their Gibbs representations. We present a method, called the relative weighted frequency method, to assign production probabilities that impose proper PCFG distributions on finite parses. We demonstrate that these distributions have finite entropies. In addition, under the distributions, sizes of parses ha...

Estimation of Consistent Probabilistic Context-free Grammars

We consider several empirical estimators for probabilistic context-free grammars, and show that the estimated grammars have the so-called consistency property, under the most general conditions. Our estimators include the widely applied expectation maximization method, used to estimate probabilistic context-free grammars on the basis of unannotated corpora. This solves a problem left open in th...