We started with a protein sequence that did not give any good hits in the BLAST search. However, through the use of Prosite, Interpro, and SMART we now have some potential leads, which could help us with the design of experiments in the lab to explore the function of the protein.

Prosite suggested the presence of:

Interpro suggested the presence of:

SMART suggested:

If these facts are combined, it suggests that the protein is transmembrane spanning with the C-terminal outside the cell (or in the lumen of an organelle), and the N-terminal inside the cell.

Working out that the protein could be membrane-spanning is easy, as the presence of a transmembrane spanning domain is suggested from the SMART and Interpro results. The presence of the N-glycosylation site (at 7), before the transmembrane spanning domain (47-69) suggests the N-terminal is outside the cell as N-glycosylation only occurs on sections of protein that are outside the cell. This is supported by the the predicted presense of two possible PKC phosphorylation sites at positions 149 and 286 and two possible tyrosine kinase phosphorylation sites at positions 137 and 274, as these kinases will only phosphorylate parts of proteins in the cytoplasm.

However, the above is contradicted by the Interpro result which suggests the C-terminal outside the cell, and the N-terminal is inside the cell. And, as we know, you don't get N-glycosylation inside the cell, or phosphorylations by PKC and tyrosine kinases outdie the cell.

This contradiction of findings demonstrates that you should always question the results and not just accept what the computer tells you.

Peronally, I favour the Interpro result for protein membrane orientation because we turned off the 'Exclude motifs with a high probability of occurrence from the scan' filter on Prosite. This means that the scan picked up the kinase and N-glycosylation site motifs (N-glycosylation site and kinase motifs tend to be short, and so there is a good chance they will occur at random). In this sense, the Prosite scan can be misleading.

These simple observations from the three searches would now allow us to design a series of experiments to test the ideas. We have formed a hypothesis.