The models used for calculating the LogP and LogD scores based on the chemical structures registered in a Vault have been updated with an extended set of training data. The values for LogP and LogD in CDD Vault might change based on these updated models. (For example, the dataset used in the new LogD models was increased considerably by incorporating data aggregated from ChEMBL by Wang et al. (2023, ChEMBL) and by reviewing and adding new data from PubChem. Additional effort was made to add logD values for peptides.)
CDD’s proprietary machine learning technology can rapidly search ChEMBL to suggest similar chemical structures for the compounds registered in your proprietary Vault. The Deep Learning algorithm now reports a similarity score for each ChEMBL compound returned. The score will appear below each structure (complete with a hover pop-up description) and is included in the exported file.
Read more about the Deep Learning feature set in this Release Note, and/or watch a recording of a webinar showcasing this technology.
Email addresses are now included in the GET Users API call. Remember, for security purposes, this API command is only available for Vault Administrators.
The JSON returned now includes a section resembling this:
[ { "id": 1234, "first_name": "Charlie", "last_name": "Weatherall", "email": "charlie@collaborativedrug.com", "role_name": "vault admin", "can_witness_eln": true, "projects": [ { "name": "CharlieW Project", "id": 8745, "can_manage_project": true, "can_edit_data": true, "hidden": false } ] }]
This knowledgebase article provides the full documentation for this API endpoint.
This blog is authored by members of the CDD Vault community. CDD Vault is a hosted drug discovery informatics platform that securely manages both private and external biological and chemical data. It provides core functionality including chemical registration, data visualization, inventory, and electronic lab notebook capabilities.