Herbalog: A tool for target-based identification of herbal drug efficacy through molecular docking

Traditionally, molecular docking is primarily employed to screen pure compounds; the top-ranking chemicals are subsequently selected for experimental validation. Unlike synthetic chemicals, most natural products are commercially unavailable. The isolation and purification of each natural product is extremely time-consuming, which has restricted the screening of lead compounds from natural products.

We developed a protocol, Herbalog, to facilitate the identification of bioactive phytochemicals through molecular docking.

We wrote a script using Python and Autodock Vina for docking; ligand displacement and adipolysis assays were used to determine the anti-fatty acid binding protein (FABP) 4 activity of bioactive extracts. An ultraperformance liquid chromatography quadrupole time-of-flight mass spectrometry system was applied for identifying major peaks of bioactive extracts.

Herbalog, a natural product database, contains 5,112 phytochemicals from 197 common herbs and a script that counts the number of hits from docking in each herb and calculates the hit rate of herbs. Herbalog prioritizes herbs according to their hit rates, and top-ranking herb candidates contain a large repertoire of hits. We used Herbalog as a screening tool and identified labdane diterpenoids from Andrographis paniculata as leading candidates of FABP4 inhibitors.

Herbalog facilitates the discovery of herbs that possess the highest number of inhibitors or activators against target proteins, which reduces the sample preparation time for preliminary validation.