The Comprehensive Directory of Bacterial Promoters comprises a database of in-silico predicted bacterial promoter sequences, based on an Extreme Gradient Boosting (XGBoost) model trained with 15,654 promoter sequences from diverse bacterial organisms. In addition, a second classificatory instance in which the misclassification of the first method is further trained with a second XGBoost method to capture distinctive promoter signals, was considered.

CDBProm includes over 12 million promoter sequences from more than 6,000 unique bacterial genomes. Each promoter predicted is mapped to the transcriptomic data of the organism, identifying the function of the gene regulated by each predicted promoter.

Visitors