Publication in Nature Chemical Biology, June 2017 - UMR Genomique Metabolique - Genoscope
The discovery of key residues in MetX and MetA families, for which we have associated activity through our experimental screening on representatives, allows us to predict activity for all members of each family.
Thus, we screened all modeled active sites of both families. Depending on the presence of these key residues in the active site, we re-annotated all homoserine-O-acyl transfereases from MetX and MetA families following the rules described below:
Decision trees for annotating MetX and MetA . Rules for predicting enzymatic activities of (a) MetX (reference structure: MetXinfluenzae; PDB code 2B61) and (b) MetA (reference structure: MetAcereus; PDB code 2VDJ) are based on the SDP residues highlighted by the structural analysis. Sequences corresponding to InterPro IPR005697 (MetA) and IPR008220 (MetX) were submitted to homology modeling. Sequences that could not be modeled led to "Not Predicted" (NP). Models were aligned with the reference crystallographic structure of the family, and the active site residues were extracted to form a logo sequence. Function prediction was determined by inspecting the SDPs. At the bottom of the figure are indicated the number of proteins for each predicting category. The new enzyme nomenclature that is suggested to Swiss-Prot is indicated in brackets.
Download all the prediction here or submit one or a list of accession numbers (UniprotKB id sequences) below:
Please, note that your protein might not be in our database since the last update of our study is from January 2016. The next update will be in January 2018.