Previous work in predicting protein localization to the chloroplast organel
le in plants led to the development of an artificial neural network-based a
pproach capable of remarkable accuracy in its prediction (ChloroP). A commo
n criticism against such neural network models is that it is difficult to i
nterpret the criteria that are used in making predictions. We address this
concern with several new prediction methods that base predictions explicitl
y on the abundance of different amino acid types in the N-terminal region o
f the protein. Our successful prediction accuracy suggests that ChloroP use
s little positional information in its decision-making; an unexpected resul
t given the elaborate ChloroP input scheme. By removing positional informat
ion, our simpler methods allow us to identify those amino acids that are us
eful for successful prediction. The identification of important sequence fe
atures, such as amino acid content, is advantageous if one of the goals of
localization predictors is to gain an understanding of the biological proce
ss of chloroplast localization. Our most accurate predictor combines princi
pal component analysis and logistic regression. Web-based prediction using
this method is available online at http://apicoplast.cis.upenn.edu/pclr/.