How to create a computation job for GPCR-specific Chemogenomics Knowledgebase for DA?
To issue a new computation job, you first need to click on the yellow Create a new job button from the ALL JOBS page. You can find the link to ALL JOBS wherever you are on this web site.
Name your job and draw your molecule with the Molecular Editor. You can also paste in any pre-existing molecule structure in either MOL, SDF or SMILES format via the dropdown menu of the blue double-triangle icon on the toolbar. In this way, please be aware that the pasted code should NOT contain any leading or trailing spaces. Hit Accept to continue.
Verify the graphic depiction of your molecule presented at the Molecular Editor canvas area. You can use the editor to modify it as needed and then hit the yellow Create Job button below when it is ready to go. Your job will be created immediately and enqueued to be processed.
How to inspect prediction results?
Once your job is created and enqueued, you will be brought to the home page of the newly created job where you are presented useful functionalities like Molecular Fingerprint, Molecular Model Download and BBB Prediction for the input molecule. Scroll down the page to access all the tools. Links to OUTPUT and SPIDER PLOT can be found from the context-aware navigation bar where you will find prediction results of your job.
The OUTPUT page can display results of your job in either table mode or cards mode for your conveniences. Toggle between views with the top right switch. In table mode you are provided multiple options to customize the view such as defining the pagination length and the sorting column. Results that are finalized will be presented with links to see detailed information about the protein, to download models or to compare between best matched known molecule. Also note that the sorting headers support multi-sorting by clicking multiple column headers one after another while holding SHIFT key at the same time. In card mode you can instinctly compare 3D models of the targets. Same functionalities are provided but view customization is limited.
Switch to SPIDER PLOT and you will see prediction result in graphics. A bunch of customization tools are provided. You can define molecule/target circle colors and sizes, text colors and sizes, border sizes and colors. You can also toggle visibility of interaction labels, display of molecular depiction. All nodes on the graph are also manipulatable. You can define position of every single nodes and by holding SHIFT or CTRL key you can even move the entire graph and multi-select nodes. Additional functionalities including exporting the plot to an image are coming soon. At the moment with Chrome you can Save image as by right clicking on the image.