Species-specific splice site prediction for Physcomitrella has been developed using an HMM (Hidden Markov Model) and a SVM (Support Vector Machine).
Both models predict less false positives than NetPlantGene, an ANN (Artificial Neural Network) trained on Arabidopsis data.
Because of the good overall performance (high recall and precision) we suggest to use the SVM as first choice for Physcomitrella splice site prediction.

Performance of HMM and SVM as compared to NetPlantGene

HMM and SVM trained and predicted on Physcomitrella patens sequences.
NetplantGene trained with Arabidopsis and predicted on Physcomitrella sequences.

donor siteacceptor site

Details / Citation

Rensing SA, Fritzowsky D, Lang D and Reski R (2005)
Protein encoding genes in an ancient plant: analysis of codon usage, retained genes and splice sites in a moss, Physcomitrella patens.
BMC Genomics 6:43