Automated assay optimization with integrated statistics and smart robotics

Automated liquid handling robot for rapid lateral flow assay development

The lateral flow assay (LFA) is one of the most popular technologies on the point-of-care diagnostics market due to its low cost and ease of use, with applications ranging from pregnancy to environmental toxins to infectious disease. While the use of these tests is relatively straightforward, significant development time and effort are required to create tests that are both sensitive and specific. Workflows to guide the LFA development process exist but moving from target selection to an LFA that is ready for field testing can be labor intensive, resource heavy, and time consuming. To reduce the cost and the duration of the LFA development process, we introduce a novel development platform centered on the flexibility, speed, and throughput of an automated robotic liquid handling system. The system comprises LFA-specific hardware and software that enable large optimization experiments with discrete and continuous variables such as antibody pair selection or reagent concentration. Initial validation of the platform was demonstrated during development of a malaria LFA but was readily expanded to encompass development of SARS-CoV-2 and Mycobacterium tuberculosis LFAs. The validity of the platform, where optimization experiments are run directly on LFAs rather than in solution, was based on a direct comparison between the robotic system and a more traditional ELISA-like method. By minimizing hands-on time, maximizing experiment size, and enabling improved reproducibility, the robotic system improved the quality and quantity of LFA assay development efforts.

Graphical abstract

Cross-Platform Bayesian Optimization System for Autonomous Biological Assay Development

Current high-throughput screening assay optimization is often a manual and time-consuming process, even when utilizing design-of-experiment approaches. A cross-platform, Cloud-based Bayesian optimization-based algorithm was developed as part of the National Center for Advancing Translational Sciences (NCATS) ASPIRE (A Specialized Platform for Innovative Research Exploration) Initiative to accelerate preclinical drug discovery. A cell-free assay for papain enzymatic activity was used as proof of concept for biological assay development and system operationalization. Compared with a brute-force approach that sequentially tested all 294 assay conditions to find the global optimum, the Bayesian optimization algorithm could find suitable conditions for optimal assay performance by testing 21 assay conditions on average, with up to 20 conditions being tested simultaneously, as confirmed by repeated simulation. The algorithm could achieve a sevenfold reduction in costs for lab supplies and high-throughput experimentation runtime, all while being controlled from a remote site through a secure connection. Based on this proof of concept, this technology is expected to be applied to more complex biological assays and automated chemistry reaction screening at NCATS, and should be transferable to other institutions.

Assay Optimization: A Statistical Design of Experiments Approach

With the transition from manual to robotic HTS in the last several years, assay optimization has become a significant bottleneck. Recent advances in robotic liquid handling have made it feasible to reduce assay optimization timelines with the application of statistically designed experiments. When implemented, they can efficiently optimize assays by rapidly identifying significant factors, complex interactions, and nonlinear responses. With the use of an integrated approach called automated assay optimization developed in collaboration with Beckman Coulter (Fullerton, CA), the process of conducting these experiments has been greatly facilitated. This approach imports an experimental design from a commercial statistical package and converts it into robotic methods. The data from these experiments are fed back into the statistical package and analyzed, resulting in empirical models for determining optimum assay conditions. The optimized assays are then progressed into HTS. This tutorial will focus on the use of statistically designed experiments in assay optimization.

Design and Implementation of High-Throughput Screening Assays

HTS remains at the core of the drug discovery process, and so it is critical to design and implement HTS assays in a comprehensive fashion involving scientists from the disciplines of biology, chemistry, engineering, and informatics. This requires careful consideration of many options and variables, starting with the choice of screening strategy and ending with the discovery of lead compounds. At every step in this process, there are decisions to be made that can greatly impact the outcome of the HTS effort, to the point of making it a success or a failure. Although specific guidelines should be established to ensure that the screening assay reaches an acceptable level of quality, many choices require pragmatism and the ability to compromise opposing forces.

Cellular Assay Optimization

This report describes the implementation of an automated work cell with commercially available hardware and software, capable of handling up to 15 separate reagents for performing 96-well or 384-well assays but with a small footprint and only a single liquid dispenser and two plate washers. Extremely flexible software was used to enable this simple work cell to perform processes that would traditionally require a much larger, more expensive automation platform. With the development of the C-Myc assays for the targets DYRK, BMX, PERK, and FAK, the authors describe a software solution to multibatch assays to run simultaneously, reducing reagent dead volume and increasing the efficiency of running multiple assays such that the time to generate data across multiple targets was significantly shortened. Although a larger automated system with multiple robotic arms and extensive equipment would also be able to process multiple assays simultaneously, the work cell we have described represents an inexpensive and flexible, easily upgradable option suitable for a wider range of labs.

Screening for inhibitors of low-affinity epigenetic peptide-protein interactions: an AlphaScreen-based assay for antagonists of methyl-lysine binding proteins

The histone code comprises many posttranslational modifications that occur mainly in histone tail peptides. The identity and location of these marks are read by a variety of histone-binding proteins that are emerging as important regulators of cellular differentiation and development and are increasingly being implicated in numerous disease states. The authors describe the development of the first high-throughput screening assay for the discovery of inhibitors of methyl-lysine binding proteins that will be used to initiate a full-scale discovery effort for this broad target class. They focus on the development of an AlphaScreen-based assay for malignant brain tumor (MBT) domain-containing proteins, which bind to the lower methylation states of lysine residues present in histone tail peptides. This assay takes advantage of the avidity of the AlphaScreen beads to clear the hurdle to assay development presented by the low micromolar binding constants of the histone binding proteins for their cognate peptides. The assay is applicable to other families of methyl-lysine binding proteins, and it has the potential to be used in screening efforts toward the discovery of novel small molecules with utility as research tools for cellular reprogramming and ultimately drug discovery.

Statistics and decision making in high-throughput screening

Screening is about making decisions on the modulating activity of one particular compound on a biological system. When a compound testing experiment is repeated under the same conditions or as close to the same conditions as possible, the observed results are never exactly the same, and there is an apparent random and uncontrolled source of variability in the system under study. Nevertheless, randomness is not haphazard. In this context, we can see statistics as the science of decision making under uncertainty. Thus, the usage of statistical tools in the analysis of screening experiments is the right approach to the interpretation of screening data, with the aim of making them meaningful and converting them into valuable information that supports sound decision making.In the HTS workflow, there are at least three key stages where key decisions have to be made based on experimental data: (1) assay development (i.e. how to assess whether our assay is good enough to be put into screening production for the identification of modulators of the target of interest), (2) HTS campaign process (i.e. monitoring that screening process is performing at the expected quality and assessing possible patterned signs of experimental response that may adversely bias and mislead hit identification) and (3) data analysis of primary HTS data (i.e. flagging which compounds are giving a positive response in the assay, namely hit identification).In this chapter we will focus on how some statistical tools can help to cope with these three aspects. Assessment of assay quality is reviewed in other chapters, so in Section 1 we will briefly make some further considerations. Section 2 will review statistical process control, Section 3 will cover methodologies for detecting and dealing with HTS patterns and Section 4 will describe approaches for statistically guided selection of hits in HTS.

Design and implementation of high-throughput screening assays

HTS is at the core of the drug discovery process, and so it is critical to design and implement HTS assays in a comprehensive fashion involving scientists from the disciplines of biology, chemistry, engineering, and informatics. This requires careful analysis of many variables, starting with the choice of assay target and ending with the discovery of lead compounds. At every step in this process, there are decisions to be made that can greatly impact the outcome of the HTS effort, to the point of making it a success or a failure. Although specific guidelines should be established to ensure that the screening assay reaches an acceptable level of quality, many choices require pragmatism and the ability to compromise opposing forces.

Assay Optimization: A Statistical Design of Experiments Approach

With the transition from manual to robotic HTS in the last several years, assay optimization has become a significant bottleneck. Recent advances in robotic liquid handling have made it feasible to reduce assay optimization timelines with the application of statistically designed experiments. When implemented, they can efficiently optimize assays by rapidly identifying significant factors, complex interactions, and nonlinear responses. This article focuses on the use of statistically designed experiments in assay optimization.

Advances in nasal drug delivery through tight junction technology

New approaches for enhancing intranasal drug delivery based on recent discoveries on the molecular biology of tight junctions (TJ) are significantly improving the bioavailability of 'non-Lipinsky' small molecules, and peptide, protein and oligonucleotide drugs. As knowledge of the structure and function of the TJ has developed, so has the ability to identify mechanism-based TJ modulators using high-throughput molecular biology-based screening methods. The present review focuses on recent developments on the TJ protein complex as a lipid raft-like membrane microdomain, the emerging role of unique endocytic pathways in regulating TJ dynamics, and the utility of techniques such as RNA interference and phage display to study TJ components and identify novel peptides and related molecules that can modulate their function. Experimental and statistical methodologies used for the identification of new classes of TJ modulators are described, which are capable of reversibly opening TJ barriers with broad potential to significantly improve intranasal and, eventually, oral drug delivery. The development of an advanced intranasal formulation for the obesity therapeutic PYY(3-36), the endogenous Y2 receptor agonist is also reviewed.

A comparison of assay performance measures in screening assays: signal window, Z' factor, and assay variability ratio

In this article, the authors compare the assay performance measures, signal window, Z' factor, and assay variability ratio. They examine their mathematical formulae for similarities and differences, describe their statistical sampling properties using the results of a computer simulation, and illustrate their use with example data. Based on these results, the authors recommend the Z' factor as a preferred measure of assay performance for screening assays and point out that none of these measures are adequate for characterizing concentration-response assays.

The Minimum Significant Ratio: A Statistical Parameter to Characterize the Reproducibility of Potency Estimates from Concentration-Response Assays and Estimation by Replicate-Experiment Studies

The authors show by illustration that procedures used to validate the reliability of single-concentration high-throughput screens such as the signalwindowand Z' factor do not ensure sufficient reliability in potency estimates fromconcentration response assays. They develop theminimumsignificant ratio as a statistical parameter to characterize the fold change between 2 compounds run in the same experiment that can be considered a real difference and use this parameter to characterize the reliability of the assay. They adaptmethods described by Bland and Altman to develop a simple set of 2 experiments to estimate theminimum significant ratio and show that this protocol can identify assays that lack reproducibility. The methods are then extended to validate the equivalency of the same assay run by multiple laboratories.

Thrombin inhibitors identified by computer-assisted multiparameter design

Here, we present a series of thrombin inhibitors that were generated by using powerful computer-assisted multiparameter optimization process. The process was organized in design cycles, starting with a set of randomly chosen molecules. Each cycle combined combinatorial synthesis, multiparameter characterization of compounds in a variety of bioassays, and algorithmic processing of the data to devise a set of compounds to be synthesized in the next cycle. The identified lead compounds exhibited thrombin inhibitory constants in the lower nanomolar range. They are by far the most selective synthetic thrombin inhibitors, with selectivities of >100,000-fold toward other proteases such as Factor Xa, Factor XIIa, urokinase, plasmin, and Plasma kallikrein. Furthermore, these compounds exhibit a favorable profile, comprising nontoxicity, high metabolic stability, low serum protein binding, good solubility, high anticoagulant activity, and a slow and exclusively renal elimination from the circulation in a rat model. Finally, x-ray crystallographic analysis of a thrombin-inhibitor complex revealed a binding mode with a neutral moiety in the S1 pocket of thrombin.

High-Throughput Screening for Kinase Inhibitors

Following G protein-coupled receptors (GPCRs), protein kinases have become the second most important class of targets for drug discovery over the last 20 years. While only four kinase inhibitors have reached the market to date (Fasudil for rho-dependent kinase, Rapamycin for TOR, Gleevec for BCR-Abl, and Iressa for EGFR), many more are already in clinical development. A historical overview of kinase inhibitors was recently published by Cohen. [1] After the previous successes, protein kinases are now regarded as attractive, well-drugable targets, and the analysis of the human genome has yielded 518 protein kinases. [2] We can thus expect screening for protein kinase inhibitors to become even more important in the future. In this review we will focus on the early steps of drug discovery programs producing new lead compounds. We will guide the reader through efficient state-of-the-art assay development and high-throughput screening of large chemical libraries for protein kinase inhibitors.

Engineered protein function by selective amino acid diversification

Almost all protein engineering methods rely upon making changes to naturally occurring proteins that already possess some of the desired properties. This will probably remain the case as long as we lack a complete understanding of the way that an amino acid sequence gives rise to a protein with a precisely defined biological function. Common to all methods for altering an existing protein is the selection of a subset of amino acids in the protein for variation and a choice of which substitutions to make at each position. Variants are then tested empirically and further variants are created based upon their performance. Differences between protein engineering methods are the ways in which amino acids are chosen for variation, the protocols followed for creating the variants, and how information regarding variant properties is used in creating subsequent variants. In this article, we describe these differences and provide examples of how the experimental parameters of specific projects determine which method is most suitable.

Application of statistical ‘design of experiments’ methods in drug discovery

The use of 'design of experiments' (DoE) is a revolutionary approach to optimisation and screening of experimental parameters. Simple experimental designs and statistical tools for data analysis can provide much information about the system under investigation after only a few experiments. Such information can be key in decision-making for further experiments and can enable the development of robust and reliable protocols for chemical synthesis, analytical methods or biological assays. Coupling of design of experiments with modern high-throughput automation systems has the potential to maximise the capabilities of these systems and give increased productivity for many drug discovery applications.

Serotonin antagonist profiling on 5HT2A and 5HT2C receptors by nonequilibrium intracellular calcium response using an automated flow-through fluorescence analysis system, HT-PS 100

Characterization of the potencies of agonists and antagonists in cell-based assays can be complicated by nonequilibrium conditions of functional response. We assessed the potencies of a series of serotonin (5HT) antagonists by inhibition of intracellular calcium response in HEK 293 cells expressing 5HT(2A) or 5HT(2C) receptors. An automated system, HT-PS 100, was used to profile the antagonists in two experimental setups: coadministration of agonist and antagonist to cells and preincubation of the cells with antagonist prior to agonist administration. We showed that the antagonist potencies (pIC(50) values) determined in the preincubation configuration were close to or exceeded those measured in the coadministration configuration. Closeness of the potencies determined in the two configurations supposedly reflected a rapid antagonist-receptor equilibration, whereas a significantly higher preincubation potency implied slow antagonist dissociation from the receptor. Schild analysis of the inhibition of serotonin-induced cell response by a competitive 5HT(2A) antagonist, spiperone, showed a typical competitive inhibition pattern when both the agonist and antagonist were applied simultaneously. Contrary to this, an insurmountable diminishing of the maximal cell response to serotonin was observed when the cells were preincubated with spiperone. We conclude that a combination of the coadministration and preincubation experimental setups is necessary for appropriate mechanistic interpretation and quantitative assessment of the antagonist activity when using transient functional readouts.

Miniaturized HTS technologies - uHTS

The transition from slow, manual, low-throughput screening to industrialized robotic ultra-high throughput screening (uHTS) in the past few years has made it possible to screen hundreds of thousands of chemical entities against a biological target in a short time-frame. The need to minimize the cost of screening has been addressed primarily by reducing the volume of sample to be screened. This, in turn, has resulted in the miniaturization of HTS technology as a whole. Miniaturization requires new technologies and strategies for compound handling, assay development, assay adaptation, liquid handling and automation in addition to refinement of the technologies used for detection systems and data management. This review summarizes current trends in the field of uHTS and illustrates the technological developments that are necessary to enable the routine application of miniaturized uHTS systems within an industrial environment.