Dataset upload
Data uploaded to ToxiVerse must be in the Structure Data Format (SDF) format. Every chemical in the dataset must contain two properties. These are (1) an activity and (2) a unique compound identifier (ID). These properties can be named anything in the SDF file, however, they must be provided in the appropriate fields below. If not provided, then they are assumed to be Activity and CMP_ID.
Activity Property
The activity must be binarized or continuous.
Compound ID Property
Every chemical must have a unique compound identifier (ID). It is important to make sure this field is unique (i.e., no duplicate values).