I stumbled upon a new table “targetsPriorisation” which seems to have appeared in the 23.06 release (Index of /pub/databases/opentargets/platform/23.06/output/etl/json/targetsPriorisation). I didn’t see any mention of this table in the 23.06 release blog or any of the other released documentation.
This table seems really cool because it has some information such as which proteins are predicted to be secreted but having some more information about this table would be helpful.
Thanks and great work maintaining and extending this valuable platform!
Director of Discovery Data Science
Calico Life Sciences LLC
This new table is a feature currently being tested with the Open Target partners, but we decided to release the data and API for convenience. There will be more news about Target Prioritisation (with the typo in the filename fixed ) later this month.
As you guessed, the dataset contains an assessment of target properties that might make them more amenable to follow-up in a drug discovery program. The features broadly cover different aspects of the drug discovery process: clinical precedence, safety, tractability or doability. All of them are standardised so they can be interpreted as a traffic light system in the range [-1, 1], but they are so far designed to raise awareness of the underlying information, not necessarily as a fully quantitative assessment. We are still performing different benchmarks on the data and we keep iterating on some of the factors, so please use it at your own risk. For specific documentation on how every feature is built please stay with us.
Thanks for reaching out
Thanks for the context! I’m looking forward to hearing more when the file is officially released.
All the best,