Machine learning models can predict DNA-binding affinities directly from protein sequences.
This receipt is being re-checked. Its verdict rests on a single settling source, and receipts in that position now wait for a human before they are published.