通过Docking的方法预测结构已知而底物未知的酶的底物，这不是什么新方法，不过，在这里作者在做Docking时采用的是候选底物的高能中间态形式（ high-energy intermediate forms ），这也许就是文章可以发在《Nature》上面的原因。
按照序列比对预测的底物被实验否定。按照新方法预测的结果被实验证实，并且进一步得到“酶－产物”复合物结构。
Structure-based activity prediction for an enzyme of unknown
function pp775 - 779
A computational approach is used to predict the function of an
uncharacterized enzyme by docking high-energy intermediate forms of
candidate metabolites into its purported binding site. The docking
experiments predicted that the enzyme would be able to deaminate
intermediates of 5-methylthioadenosine and S-adenosylhomocysteine, a
prediction confirmed by biochemical experiments and examination of the
X-ray crystal structure of the protein.
Johannes C. Hermann et al.
doi:10.1038/nature05981
Abstract:
http://www.nature.com/nature/journal/v448/n7155/abs/nature05981.html
Article:
http://www.nature.com/nature/journal/v448/n7155/full/nature05981.html
文章已下载到“我的文件夹”