If you have already explored the functionalities of Althotas, then I appreciate your taste — you are on a good track. If you haven’t, then tell us what are you waiting for?
Today I am posting to make my fellow researchers familiar with the merits of Pgp substrate prediction, as it is the first in the list of virtual experiments that you can perform using Althotas. Human Pgp (P-glycoprotein) is a large multipass transmembrane protein that belongs to ABC (ATP-binding cassette) multidrug transporter superfamily. Pgps transports of xenobiotics and endogenous compounds out of the cells using energy derived from ATP hydrolysis. P-gp substrates are mainly hydrophobic and weakly amphipathic substances. Pgps transport an extensive number of drugs like, antibiotics, steroid hormones, chemotherapeutics, immunosuppressants, HIV protease inhibitors. Our software will help you in knowing the prospective and existing substrates that can follow a pgp mediated pathway for their translocation in cellular environment.
You can calculate the SVM predictions as well as create a docking geometry of your selected substrate/ligand with different PDB models of pgps. There are two mouse X-Ray structures and a human model structure available so far in Althotas. You can use these models to dock your ligands to Pgp. Just follow the following simple steps:
- Type in a job name
- Choose SVM predictions and select the pgp model that you want to do the docking experiment with.
- Specify a ligand. Ligands can be uploaded or search by text. You can of course select multiple ligands and save them in a file and then upload them all in one go (Check out the tutorial for details)
Give the server a few seconds and your results will be uploaded to your dashboard from where you can view them and they will also be saved permanently for later use! To put it briefly, your results will tell you if the ligand can potentially be a pgp substrate (based on SVM predictions) and it will also create a docking calculation in order to calculate the geometry of your ligand-receptor complex which can be edited to change view/colours and different functions can be applied on it in order to calculate angles and lengths.
Results for pgp substrate prediction for drugs namely; Allopurinol, Azapropazone, Diiodosalicylic acid and Busulfan. As it turns out none of them can be substrates for pgps based on SVM predictions, the binding energy in kcal/mol for each substrate-pgp pair is also given. You will be able to view the docking by clicking on the magnifier buttons. For more help, check our tutorial.
So this is just one example, I will be posting on substrate prediction with albumins and cytochromes very soon, stay tuned!
Predicting P-Glycoprotein-Mediated Drug Transport Based On Support Vector Machine and Three-Dimensional Crystal Structure of P-glycoprotein. (2011) Zsolt Bikadi, Istvan Hazai, David Malik, Katalin Jemnitz, Zsuzsa Veres, Peter Hari, Zhanglin Ni, Tip W Loo, David M Clarke, Eszter Hazai, Qingcheng Mao. PLoS One. 6 (10):e25815