Computational prediction tool for Leafy Cotyledon Proteins (LEC) in oil palms and date palms

Abstract

Members of the palm family (Arecaceae) constitute one of the most economically essential crops used by mankind. Even though genome sequences of oil palm (Elaeis guineensis) and date palm (Phoenix dactylifera) are available, genome wide analysis has not been undertaken yet in these crops. In the present study we have analyzed the genome sequence of Elaeis guineensis and Phoenix dactylifera for the presence of protein Leafy Cotyledon (LEC), which has been implicated in the control of embryo development and maturation. The resultant LEC protein sequences from the two palms were then used to create predictive model using different computational algorithms like Naive Bayes, SMO, MLP and Random Forest based on their motif pattern and amino acid properties such as charged amino acids and basic amino acids. Performance testing of the computational models developed in this study resulted in 100% accuracy for motif feature using Naive Bayes, MLP and Random Forest algorithm.