HER is a new drug disposition mechanism, where intestine is the metabolic organ and liver is the recycling organ (eLife. 2021). HERpred is based on graph representations to characterize chemical structures and integrated with a supervised Octave residual convolutional neural network (ORCNN) method to predict compound HER rates. The current model is primarily built upon a diverse set of internally tested phenolic and flavonoid analogues, with intensive studies of their phase II metabolism by enzymes such as uridine 5'-diphospho-glucuronosyltransferases (UGT) and their colon exposure that are affected by gut microbiota. This helps understand how compounds are distributed to the colon and which are more bioavailable in the gut, thus providing new insight into identification and design of better agents with desired interactions with gut microbiome and increased local exposure to colonocytes. Ultimately, based on such knowledge, we will be able to design significantly improved treatment and prevention strategies for colon diseases (e.g., colorectal cancer) by enhancing drug local bioavailability and reducing on-target-off-tissue (e.g., systemic) toxicities. This web site is implemented with the Flask framework, and has also been optimized for friendly use on mobile devices.

Input for Prediction (Cmpd Name and SMILES apart by space)

Output of Predicted HER