Looking Back to the Future: Predicting in Vivo Efficacy of Small Molecules versus Mycobacterium tuberculosis

Abstract

Selecting and translating in vitro leads for a disease into molecules with in vivo activity in an animal model of the disease is a challenge that takes considerable time and money. As an example, recent years have seen whole-cell phenotypic screens of millions of compounds yielding over 1500 inhibitors of Mycobacterium tuberculosis (Mtb). These must be prioritized for testing in the mouse in vivo assay for Mtb infection, a validated model utilized to select compounds for further testing. We demonstrate learning from in vivo active and inactive compounds using machine learning classification models (Bayesian, support vector machines, and recursive partitioning) consisting of 773 compounds. The Bayesian model predicted 8 out of 11 additional in vivo actives not included in the model as an external test set. Curation of 70 years of Mtb data can therefore provide statistically robust computational models to focus resources on in vivo active small molecule antituberculars. This highlights a cost-effective predictor for in vivo testing elsewhere in other diseases.

A Macrophage Infection Model to Predict Drug Efficacy AgainstMycobacterium Tuberculosis

“…A major problem with the successful transfer from discovery to application for Mtb drugs is that these parameters have not been given sufficient consideration during in vitro Mtb drug screening or the early lead development process. 6,7 The majority of drug screens were performed using extracellular in vitro broth-grown single-cell culture assays, in which Mtb is actively growing and only the Mtb cell membrane acts as a potential diffusion barrier.…”

Computational Prediction and Validation of an Expert’s Evaluation of Chemical Probes

“…We have previously described the generation and validation of the Laplacian-corrected Bayesian classifier models developed for various datasets using Discovery Studio 3.5 28 . This approach was utilized with the literature probe data from PubChem.…”

Bigger data, collaborative tools and the future of predictive drug discovery

“…Surprisingly, we are still making very slow progress in finding novel therapeutics [33] for TB and the clinical pipeline is limited [34]. Ideally we should be learning from the past efforts in TB drug discovery and yet we do not appear to be doing something that is simple yet effective, learning from the data that already exists [35]. The current predominant method for identifying compounds active against Mtb is to use phenotypic high throughput screening (HTS) [36-39] and the hit rate of these screens tends to be in the low single digits.…”

“…The current predominant method for identifying compounds active against Mtb is to use phenotypic high throughput screening (HTS) [36-39] and the hit rate of these screens tends to be in the low single digits. We can estimate that upwards of 5 million compounds have been screened against Mtb over the last 5-10 years [35]. There are around 1500 Mtb hits of interest from one laboratory alone [38-41].…”

“…The in vitro, in vivo and clinical data for TB do not exist in a single database. Our own efforts to collate mouse in vivo data for modeling took many months and were recently described [35]. We see this lack of data coordination as a major limitation to progress.…”