Retro Drug Design: From Target Properties to Molecular Structures.

To generate drug molecules of desired properties with computational methods is the holy grail in pharmaceutical research. Here we describe an AI strategy, retro drug design, or RDD, to generate novel small molecule drugs from scratch to meet predefined requirements, including but not limited to biological activity against a drug target, and optimal range of physicochemical and ADMET properties. Traditional predictive models were first trained over experimental data for the target properties, using an atom typing based molecular descriptor system, ATP. Monte Carlo sampling algorithm was then utilized to find the solutions in the ATP space defined by the target properties, and the deep learning model of Seq2Seq was employed to decode molecular structures from the solutions. To test feasibility of the algorithm, we challenged RDD to generate novel drugs that can activate µ opioid receptor (MOR) and penetrate blood brain barrier (BBB). Starting from vectors of random numbers, RDD generated 180,000 chemical structures, of which 78% were chemically valid. About 42,000 (31%) of the valid structures fell into the property space defined by MOR activity and BBB permeability. Out of the 42,000 structures, only 267 chemicals were commercially available, indicating a high extent of novelty of the AI-generated compounds. We purchased and assayed 96 compounds, and 25 of which were found to be MOR agonists. These compounds also have excellent BBB scores. The results presented in this paper illustrate that RDD has potential to revolutionize the current drug discovery process and create novel structures with multiple desired properties, including biological functions and ADMET properties. Availability of an AI-enabled fast track in drug discovery is essential to cope with emergent public health threat, such as pandemic of COVID-19.

Robust biofilm assay for quantification and high throughput screening applications.

Bacterial biofilms are populations of bacteria within a self-produced adherent extracellular matrix that are notoriously resistant to treatment. Existing methods for biofilm quantification are often limited in their dynamic range of detection (signal-to-background), throughput, and require modifications to the protocol depending on the bacterial species. To address these limitations, a broad utility, high-throughput (HTP) method was required. Using a fluorescent dye, FM1-43, we stained the biofilm, followed by solvent extraction and quantitation of biofilm employing a fluorescent plate reader. Utilizing eight different bacterial pathogens, we demonstrate that this method is widely applicable for biofilm quantification. Depending on the species, this biofilm assay offered a large dynamic range of 8-146 fold change compared to 2-22 fold for crystal violet staining under similar conditions. In addition to routine biofilm quantification using this new assay, as a proof-of-concept, 1200 compounds were screened against two different bacterial species to identify biofilm inhibitors. In our HTP screens we successfully identified compounds rifabutin and ethavarine as potential biofilm inhibitors of Burkholderia pseudomallei Bp82 and Acinetobacter baumannii biofilm production respectively. This newly validated biofilm assay is robust and can be readily adapted for antibiofilm screening campaigns and can supplant other less sensitive and low throughput methods.