ASPIRE Day 2021

We present here the value of applying 3-D printing technology as a rapid and effective prototyping strategy in a research laboratory setting. A few select examples of innovative tools for use at the bench are demonstrated to showcase the utility of 3-D printing parts to be used in automating various lab processes with the goal of increasing the efficiency of critical research tasks done by the scientists.

Current high-throughput screening assay optimization is often a manual process, although this can be enhanced by design-of-experiment approaches. An artificial intelligence-driven assay optimization was developed as part of the NCATS ASPIRE initiative to accelerate preclinical drug discovery. We describe the development of an autonomous, cloud-based Bayesian optimization system using the Kebotix adaptive optimization algorithm and a cell-free assay for papain enzymatic activity as proof of concept. The algorithm found suitable conditions and optimal assay performance with only 55 experiments with up to 20 experiments running simultaneously. This setup was sufficient for the Kebotix algorithm to identify and test the global optimum as determined from a brute-force experiment testing all 294 possible assay conditions. The algorithm could achieve a fivefold reduction in costs for laboratory supplies and high-throughput experimentation run time. This technology should enable future work toward applying autonomous optimization to more complex biological assays and automated chemistry reaction screening.

The open port sampling interface for mass spectrometry (OPSI-MS) is a constant-flow sample introduction module fitted to an ESI-source inlet.¹ The OPSI inlet can be used for direct sampling in multiple formats (e.g., via acoustically dispensed droplets, direct pipet, infusion, surface sampling). Here we describe development and integration of OPSI-MS with an EDC/ATS-100 acoustic dispense module for high-throughput MS/MS quantification. In addition, we describe development of an OPSI-MS workbench that can be used to test both acoustic and nonacoustic sampling applications.

¹Van Berkel, G. J., and Kertesz, V. (2015) An open port sampling interface for liquid introduction atmospheric pressure ionization mass spectrometry. Rapid Commun Mass Spectrom 29: 1749–1756.

The Analytical Chemistry Core within the Division of Preclinical Innovation at NCATS is collaborating with the ASPIRE program to develop an automated purification platform that can operate as a stand-alone system or an integrated module within the automated synthetic laboratory. The concepts of throughput and cycle time are of great significance as we endeavor to increase experimental capacity while also reducing the therapeutic discovery and development timeline to rapidly bring more treatments to patients. Purification is an important stage in the progression of small molecules from synthesis to biological assay. However, it poses a serious bottleneck in an automated workflow due to the serial nature of liquid chromatography (LC). A major advancement to our purification process is the design of a dual-arm fraction collection bed incorporating analytical and purification LC systems into a single automated platform to provide a more efficient and parallel technique. Standardization of system protocols in conjunction with automated method development aim to accommodate highly diverse chemotypes, as well as enable greater sample throughput. We have focused our efforts additionally on fully automating post-purification sample processing, which is an essential aspect of the initiative. This involves identity verification, purity analysis, sample consolidation, material distribution, compound registration and data management.

Quantitative nuclear magnetic resonance (qNMR) is a valuable quantification method and a compelling alternative to chromatographic techniques. Although other quantification techniques face challenges — such as specific response factors and the corresponding need to create calibration curves — qNMR is a direct method that only requires complete solubility and the presence of NMR active nuclei. At the same time as delivering quantitative information, qNMR can obtain qualitative information for identity confirmation. qNMR is thus both a comprehensive and flexible method that can be deployed rapidly to understand the result outcomes of automated chemistry operations, from simple evaporations to solid-phase extractions to assessing yield and conversion for complex reaction mixtures. The ASPIRE team and the NCATS Analytical Chemistry Core are collaborating to develop analytical qNMR techniques and workflows to aid in answering these chemistry integration questions. We will discuss specific ways that qNMR will be utilized for the ASPIRE project, as well as future opportunities and challenges to making qNMR a standard method fully integrated into the complete automated workflow.

Indigo Reactor is an automated chemical synthesis platform with eight independently controlled reactors — allowing separate thermal, reflux, stirring speed, inerting and pressure control — including interchangeable adapters that accommodate either a 10 mL or 20 mL removable glass reactor. This presentation will review the function of the Indigo Reactor and will cover hardware and software development. After conducting the benchmark reaction, we can verify the function for each reactor lane. I will introduce how to transfer a chemical reaction condition into a recipe, which is an important function for this reactor, and how to use it for reaction condition screening, recording the reaction and reproducing chemistry reaction repeatedly with the precise recipe. I also will cover the next-step upgrade for the hardware and software.

Medicinal and organic chemists commonly synthesize molecules that have never been made before. However, finding optimal or even working reaction conditions to make these compounds is often a time- and resource-consuming task involving manually setting up and monitoring a handful of conditions. Automated, high-throughput reaction screening accelerates the process of preparing, analyzing and standardizing a higher volume of reaction conditions. The ASPIRE team has developed a reaction screening platform with the goal to provide chemists with same-day solutions to their synthetic challenges. For setting up and executing reactions, we use an Unchained Labs Junior, which is capable of liquid handling and solid loading for hundreds of reactions in parallel at multiple temperatures. The screens are analyzed quickly using Virscidian analysis software connected to a dedicated UPLC (Ultra-high Performance Liquid Chromatograph). We have used this platform successfully to determine optimal reaction conditions for ongoing medicinal chemistry projects. Furthermore, we are using this platform to design and prepare reaction screening kits for common medicinal chemical reactions to be used by chemists or ourselves on demand. Our platform is ready to assist users and provide expedient solutions.

Novel synthetic methodologies that enable late-stage modifications of small molecules during hit-to-lead or lead optimization are highly desirable to accelerate the drug discovery process. Additionally, such robust methodologies provide an opportunity to explore a new chemical space during lead optimization. Nitrogen-containing compounds, particularly heterocycles, share a major chemical space in drug discovery. Because the introduction of the fluorine (F) atom into small molecules is a powerful approach to modulate the pharmacological properties, late-stage integration of the trifluoromethyl (CF3) group into nitrogen heterocycles would be an interesting, rational strategy to optimize the lead compounds. Moreover, it is essential to understand the effect of the N-trifluoromethyl (NCF3) group on activity and pharmacological properties because it provides valuable information for other and future drug discovery programs. Surprisingly, limited studies have been done in this area mainly because of scarcity of synthetic methodologies. This study aims at developing a new method for late-stage N-trifluoromethylation of heteroarenes and amines by employing electro/photochemical strategies and thereby investigating the in vitro activity, ADME properties and in vivo pharmacokinetic studies in comparison to their methyl (NCH3 vs. NCF3) counterparts. Optimization of experimental methods is currently underway by utilizing the ASPIRE reaction screening resources and automation platform.

Recently, NCATS has started a new initiative that will transform chemistry from an individualized craft to a modern, information-based science through the ASPIRE program. ASPIRE seeks to advance preclinical drug discovery by enabling a more rapid exploration of biologically relevant chemical space. To facilitate ASPIRE, we have developed an extramural component of the program that will enable and support effective collaborations between the greater scientific community and the NCATS DPI Intramural ASPIRE Laboratory. The program’s goals are to build upon the DPI ASPIRE infrastructure and capabilities through unique contributions from extramural scientists facilitated by the exchange of ideas, to develop co-funding opportunities between multiple government agencies, and to engage a wide spectrum of stakeholders including, but not limited to, professional societies, private industry and patient advocacy groups.

The ASPIRE program was initiated in 2018 with funding from The Helping to End Addiction Long-term^SM Initiative (or NIH HEAL Initiative). The funding was used to develop and support 2018 NCATS Aspire Design Challenges and a follow-up 2020 NCATS Reduction-to-Practice Challenge. These challenges were designed to address critical development of new treatments for pain, opioid use disorder and overdose, using cutting-edge solutions in artificial intelligence (AI) and automation. The grand prize winner of this competition will be determined by DPI ASPIRE laboratory as to which of the innovative solutions has achieved an integrated approach combining data science and data mining, biological assays, and AI and machine learning (ML) toward novel analgesics and pain/overdose medications. In addition to the challenges, funding opportunities recently were developed to support extramural investigators in taking advantage of the unique research opportunities available at the NCATS ASPIRE Laboratory and develop innovative, automated modules that will facilitate identification of novel chemical entities targeted toward currently undrugged biological space.

Here we present an update on the ASPIRE program and lessons learned that continue to shape the ASPIRE program’s ultimate goal: to make the process of drug discovery and development cheaper and faster and to bring treatments to all who need them.

The poster on Internet of Things (IoT) infrastructure will be an in-depth discussion regarding the fundamentals of our entire IoT-based laboratory dashboard. This presentation is for anyone interested in learning about the engineering details behind laboratory automation using IoT.

Modern instruments are often web enabled, app based, wireless and autonomous. Yet many of the instruments used in the laboratories these days are systems of the past and, therefore, do not have these modern features of convenience and automatization. Because these instruments are fully functional and play crucial roles in laboratory processes, we have become stagnant in our research methods and unable to update our traditional methodologies to reduce some of our in-person involvement. However, what if these older instruments could be modernized (i.e., web enabled), opening us to a new realm of possibilities. In this presentation, we will discuss the potential of transforming our outdated systems to increase researcher convenience.

Retrosynthesis is a technique for solving problems in the planning of organic syntheses. This is achieved by transforming a target molecule into simpler precursor structures regardless of any potential reactivity or interaction with reagents, which is one of the most complex issues in the field of organic chemistry. With a huge number of available reactions accumulated — such as around 1.8 million reactions available from the USPTO — how to find the most appropriate reactions for the target molecule with consideration of critical functional groups surrounding the reaction center is one of the main challenges in this field. To address this problem, one possible solution is to transform and represent such a large amount of reaction data in a more meaningful way that is suitable for retrosynthesis design computationally. Inspired by the three-layer reaction model designed by InfoChem for the purpose of reaction classification, we propose to develop a multilayer reaction knowledgebase to capture fundamental information for retrosynthesis. More specifically, we will define a reaction data model composed of three layers, namely the layers of reaction center (RC), maximum common structure (MCS) and specific reaction (SR), to capture the full spectrum of reaction information. In addition, not only will we consider the backbone structure of RC and MCS, but also the rest of structures will be represented as R-groups to allow R-group similarity comparison while reaction matching with the target molecule. Subsequently, we will be able to develop the multilayer reaction knowledgebase with accommodation of USPTO reaction data and, from there, build up retrosynthesis applications.

Dissolved oxygen (DO) and pH are factors impacting the growth characteristics of the cell. Glucose is the most important supporting factor in rapid proliferation of cells because it is the main nutrient used by the cells. High levels of carbon dioxide are toxic to cell culture, such that it acts as an inhibitory factor affecting cell metabolism. Small-scale cell culture studies in academia, as well as in industry, currently are conducted in single-use vessels that are not equipped with systems monitoring the aforementioned factors despite the fact that they play a major role in cell culture condition. As a result, findings from small-scale cell culture studies are not as useful from an analytical point of view. This fact makes these kinds of experiments less repeatable and reliable. The Center for Advanced Sensor Technology has developed noninvasive pH and DO sensors (featuring a patch attached to the bottom of the vessel), as well as dissolved carbon dioxide (DCO2) and glucose sensors to monitor cell culture environment. These sensors are suitable for various kinds of bioreactors because of their low profile. Currently, the DCO2 sensor is integrated with the T flask (featuring a sampler mounted outside of the vessel). The evaluation of the noninvasive monitoring system for DCO2 shows promising results. In future work, standard cell culture flasks equipped with sensors for DO, pH, DCO2 and glucose will provide continuous monitoring. Application of these sensors will improve the understanding of the small-scale cell culture microenvironment and provide real-time information on the nutrients and metabolites. The analytical data from the monitoring system will be used to interpret the effect of microenvironmental conditions on cell behavior.

We have created a database of 1.75 billion compounds predicted to be easily synthesizable: the Synthetically Accessible Virtual Inventory (SAVI). SAVI was generated via a set of transforms based on an adaptation and extension of the CHMTRN/PATRAN programming languages describing chemical synthesis expert knowledge, which originally stem from the Lhasa project. The chemoinformatics toolkit CACTVS was used to apply a total of 53 transforms to about 150,000 readily available building blocks from Enamine. Only single-step, two-reactant syntheses were calculated for this database, even though the technology can execute multistep reactions. The possibility to incorporate scoring systems in CHMTRN allowed us to subdivide the database into sets rated by their predicted synthesizability, with the most-synthesizable class comprising 1.09 billion synthetic products.

Point-of-care (POC) technologies have brought medical diagnostics and treatments to patients who would otherwise go without medical care. Currently, POC technologies are focused mainly on medical diagnostics — such as COVID-19 antigen test kits — and little investment has been made to manufacture therapeutics at the point of care. Additionally, COVID-19 has highlighted the genuine utility of POC technologies by displaying the urgency of providing medicines in pandemic hot spots. For this reason, a POC manufacturing platform that seeks to produce therapeutics is an essential part in future pandemic preparedness. Here, we report the utility of cell-free protein synthesis (CFPS) coupled with a POC manufacturing platform (BioMOD) capable of expression and purification of a variety of therapeutics ranging from monoclonal antibodies to insulin. Specifically, CFPS systems are the preferred method of POC expression of these therapeutics because they provide a rapid, scalable and versatile platform for synthesis of a variety of different proteins. Moreover, the capacity of CFPS reactions to be lyophilized and subsequently hydrated to synthesize novel proteins allows them to be a novel technology for POC diagnostics and treatments. With recent advances in microfluidics, these products can then be purified rapidly in continuously automated purification processes, generating a final product on the time scale of hours. To this end, we have selected the known broad-spectrum antiviral lectin Griffithsin as an ideal candidate to pilot the BioMOD system due to preliminary studies demonstrating its ability to function as an inhibitor of SARS-CoV2 infections. Currently, we are able to successfully express soluble Griffithsin within an E. coli CFPS system and plan to investigate the application of this protein on the BioMOD system.