Automated Lead Optimization of MMP-12 Inhibitors Using a Genetic Algorithm

Abstract

Traditional lead optimization projects involve long synthesis and testing cycles, favoring extensive structure-activity relationship (SAR) analysis and molecular design steps, in an attempt to limit the number of cycles that a project must run to optimize a development candidate. Microfluidic-based chemistry and biology platforms, with cycle times of minutes rather than weeks, lend themselves to unattended autonomous operation. The bottleneck in the lead optimization process is therefore shifted from synthesis or test to SAR analysis and design. As such, the way is open to an algorithm-directed process, without the need for detailed user data analysis. Here, we present results of two synthesis and screening experiments, undertaken using traditional methodology, to validate a genetic algorithm optimization process for future application to a microfluidic system. The algorithm has several novel features that are important for the intended application. For example, it is robust to missing data and can suggest compounds for retest to ensure reliability of optimization. The algorithm is first validated on a retrospective analysis of an in-house library embedded in a larger virtual array of presumed inactive compounds. In a second, prospective experiment with MMP-12 as the target protein, 140 compounds are submitted for synthesis over 10 cycles of optimization. Comparison is made to the results from the full combinatorial library that was synthesized manually and tested independently. The results show that compounds selected by the algorithm are heavily biased toward the more active regions of the library, while the algorithm is robust to both missing data (compounds where synthesis failed) and inactive compounds. This publication places the full combinatorial library and biological data into the public domain with the intention of advancing research into algorithm-directed lead optimization methods.KEYWORDS Lead optimization, MMP-12 inhibitors, genetic algorithm, microfluidic chemistry U sing biological data in "real time" to drive a chemistry optimization program was suggested over 10 years ago by several groups. 1-7 At GlaxoSmithKline (GSK), we have retained an interest in such approaches for a number of years and have made several attempts to drive traditional lead generation or lead optimization projects in this fashion. However, several factors contributed to only incomplete results. The traditional make/test cycle can be very long for anything but the most straightforward chemistry. This is compounded by the fact that the algorithms tend to suggest small numbers of noncombinatorial products. The extended cycle times provide plenty of time for reflection and analysis, which will inevitably compete with the suggestions of the algorithm, particularly in the early stages. In addition, other external factors come into play, such as structure-activity relationship (SAR) from related series, which may make the current template of less interest to the program.A microfluidic-based chemistry and biology platform...

Designing Lead Optimisation of MMP-12 Inhibitors

“…As a result, 90% of the simulations have been able to find greater or equal response values with respect to the best result obtained in Pickett et al [1]. This result shows the robustness of the EDO with respect to the choice of the initial population confirming the good performance of the approach.…”

“…The approach that we derived outperforms the GAO methodology developed by Pickett et al [1]. In fact selecting 120 experimental points from the whole search space, EDO is able to find the global optimum value.…”

Streamlining bioactive molecular discovery through integration and automation

“…Computation tools are now available to support the prediction of the outcome of chemical reactions, 52,53 and the design of follow-up molecules with desirable properties (including affinity, selectivity and physicochemical profiles). 2,43,54,55 However, the integration of all of these components remains an unmet challenge that is nonetheless needed to realise fully autonomous molecular discovery!…”

Autonomous Discovery in the Chemical Sciences Part I: Progress

“…rely on experimental outcomes instead [565] ). In av ery early example of iterative molecular optimization using ag enetic algorithm, Weber et al described the identification of inhibitors of the serine protease thrombin;1 6g enerations led to the identification of sub-micromolar inhibitors.…”