Page Comparison

Note: Before doing this practice it is best to first do the practical on NR ligand binding/wiki/spaces/DOC/pages/327684, which is the introduction to 3DM.

Note that sometimes you need to make a panel of a subset of sequences. For instance, say you want to find the most active enzyme with a certain specificity. Then you should first make a subset that contains only sequences that have this specificity. This sounds simple, but due to wrong notation of proteins is tricky. The best way to do this is to first do a keyword search to find enzymes likely to have the correct specificity and make a subset of this set of sequences. Then use this subset to make a motif with 4 to 7 amino acids that is specific for this subset. It is best to use the "subset specific residues" plot (consult the OAH questions about this plot). Make a new subset that contains all sequences that have this motif. With this approach you will not only find sequences with the correct specificity but that are annotated as "hypothetical protein", but you will also delete the sequences which are wrongly annotated. This approach doesn't make sure you have all sequences with the correct specificity, but it does maximise the chance that the once ones that are in your subset all indeed have the correct specificity.