Here is a screen shot of the OSET applet, followed by a description of its components. The target molecule in this example is LSD.

In this panel you can visualize the entire synthetic tree for the synthesis that is currently under consideration.When the tree is large the molecules may become illegible (especially if the screen resolution is small; we recommend 1024 x 768 or better), but the idea is to get an overview of the structure of the tree, not of each molecule. When possible, the heteroatoms are indicated by their chemical symbols, else they are shown as color dots according to the following code:
In this panel you can visualize the precursors from the last retrosynthetic step, as well as unanalyzed precursors from previous steps. On the left side of this panel you can see the precursors from the last step. They are relatively big and with black lines (if a precursor appears with green lines it's because it has been analyzed previously). On the right side of the panel are the orphans, precursors found on previous steps that were left unanalyzed. Thay appear here to remind the user that they must be analyzed later unless they are considered as available starting materials. The orphans appear smaller than the other precursors and with gray lines. If an orphan has been analyzed previously, it appears red.
When a compound is found in our "starting materials catalog", the name appears under its structure, like diethyl amine in the example. The catalog contains currently 70 compounds; not much, but they are the most common acceptable starting materials in textbook problems and exercises (mostly four carbon atoms or less).
The scrollbar under this panel is used to view alternative precursors for the current target. On the upper left corner you can see how many results were found for the current target. In the example, we are viewing the first result out of eleven. The results are sorted in decreasing order of preference according to our evaluation function, but the user does the best evaluation, so please check all the results. The number on the upper right indicates the depth of the analysis, that is, the number of analysis steps.
To analyze a molecule, just click over the structure. The mouse cursor should change to a hand when it is over a clickable structure, and during the analysis it should change to a hourglass or equivalent. The analysis usually takes a few seconds, but the delay depends more on the speed of the network connection than anything else. When analyzing a structure, remember to select the desired analysis mode (see below).
In this panel you can read the information regarding the last analysis step. The following information is available:
Click on this button to go up one level in the retrosynthetic tree. When you are on the first level (i.e., the main target molecule) this button is grayed.
Use this button to get the names, SMILES, and CAS registry numbers of all the compounds in the precursors panel. We currently use a catalog with more than 40,000 compounds. If a molecule is not found, it returns "unknown".
This radio button is used to select between three modes of analysis: disconnective, auxiliary, and FG0.
Important note: If you analyze a molecule, then go up, and then analyze it again using a different mode (or a different atom or bond), the new results are appended after the previous results. So if it seems like the results aren't new, try moving to the end of the list.