E140 Topology - Issues

[Jaguiar]

Hi Protter, First I want to thank your organization for providing this service or predictions on proteins. Second, I want to raise a potential problem with our antigen.

We are working with a malaria antigen called E140. This antigen is expressed in all species of the malaria parasite named Plasmodium, i.e. orthologs of paralogs (see link: https://plasmodb.org/plasmo/app/record/gene/PF3D7_0104100#category:evolutionary-biology.

These ortholog antigens vary by a few amino acids and some degree of aa variation BUT the structure of the protein is the same; membrane protein around 800 aa with 5 transmembrane domains.

The problem is that when we look at the topology for the P. falciparum E140 it predicts 2 small loops to be extracellular (2 and 4) and 3 loops intracellular (1, 3, and 5). For the other ortholog E140 antigens (P. yoelii, P. knowlesi, P. vivax, etc) it shows the opposite; loops 1, 3, and 5 extracellular. This is very unusual to say the least. (see graphs attached).

In fact, our lab data for P. falciparum E140 shows that loops 1, 3, and 5 are extracellular.

I assume that your algorithm is predicting these 2 models due to variation on the sequences BUT I wonder if this is “real” and in fact this protein has different topologies in different parasites albeit from the same genus (Plasmodium).

Is this issue something you could address? I can send you the sequences if you want.

I appreciate your efforts.

joao