Join MySLAS Social

DNA-Encoded Library Technology: A Quarter Century of Progress

“DELT [DNA-Encoded Library Technology] provides an economical and feasible means to make progress toward the exploration of larger chemical spaces. Instead of a focusing on one to two million compounds in a typical high-throughput screen (HTS), a DELT library may present some hundred million compounds for selection by a biological target of interest. The concepts of drug-likeness, chemical diversity and compound collection are still important, but the efficiency of the DELT process shifts the balance of opportunity costs in a unique manner relative to the traditional HTS compound collection and process.” – Robert Goodnow, Ph.D., SLAS Discovery Guest Editor

Goodnow, vice president of chemistry innovation for Pharmaron (Boston, MA), has been working in DNA-encoded chemistry for a “good number of years now” and feels the door is wide open for DELT solutions.

“DELT specifically addresses the problem of exploring the interface of chemistry and biology with hundreds of millions of small molecules in an efficient manner that has been neither possible, nor feasible before,” says Goodnow. “In a high-throughput screening (HTS) format, it’s not unusual to work through some one, two or even three million compounds,” he explains. “But at that point you start to calculate the cost to produce a compound. What will it cost per well, per assay? How much time will it take? If you go from a million to a hundred million or a billion, suddenly those costs are prohibitive. DELT changes this.”

In simple terms, Goodnow says, it’s taking what’s been discovered before and building upon it.

“DELT involves use of DNA oligos that are alternately concatenated as a means to record the introduction of small molecule reagents or building blocks,” Goodnow says. “Once a cycle of chemistry and DNA tag ligation has been completed, the separate reactions are mixed and re-distributed in a classical split-and-pool process before beginning the next cycle. A typical DELT library will have between three and five cycles with hundreds to thousands of chemical inputs and corresponding tags, thereby generating millions to billions of compounds in a short time and on a small scale. This mixture is then incubated with a protein target of interest to identify compounds that bind with relatively higher affinity, followed by isolation of that enriched mixture. The information encoded by the very small quantity of enriched samples is then amplified by PCR methods prior to sequencing of the DNA tags with next generation sequences (NGS) technologies. The sequence information is then decoded to indicate which building blocks have been enriched (i.e., have bound with a higher relative affinity) by a target protein.”

SLAS Discovery Special Collection

The SLAS Discovery Special Collection on DNA-Encoded Chemical Library Technologies: Screening and Hit Identification is well timed according to Guest Editor Goodnow, as 2017 was the 25th anniversary of a fundamental paper that laid out the concept of DNA encoding and chemical synthesis by Brenner and Lerner. He adds that the Brenner and Lerner paper describes several key aspects that are still central to the practices of this technology: orthogonal chemical and oligonucleotide bond formation, split-and-pool methodology, encoding with DNA sequences, affinity selection methods, amplification with PCR, installation of restriction sites to avoid primer dominance in selective hybridization and off-DNA re-synthesis.

“This technology has been around, and it has emerged from its initial phase to more involved detail or second generation and is now being widely practiced,” Goodnow says. “Exploring the topic in SLAS Discovery is an opportunity to gain new insight and spread the word beyond the usual audience of only chemists. After reading the special collection, I would like people to understand that this is not a chemistry-only story. There are aspects of biology and automation and information sciences that are very important. While DNA-encoded library technology was discovered, practiced and popularized by chemists, there are different types of expertise that need to get involved to move it forward.”

The DELTA Platform

A team (Jesús Castañón, José Pablo Román, Theodore C. Jessop, Jesús de Blas, Rubén Haro) from Eli Lilly and Company (Madrid, Spain and Indianapolis, IN) collaborated to publish “Design and Development of a Technology Platform for DNA-Encoded Library Production and Affinity Selection” in the SLAS Discovery Special Collection.

“As part of an effort to implement DEL (DNA-encoded library) technology, we have developed the DELTA platform, an automation and informatics solution for the design, production and affinity selection of DELs,” the authors state. “To our knowledge, this is the first report of a platform applied to DELs that builds on the concept of capturing and tracking the experimental information of the library synthesis and affinity selection as it is generated. This information is used to command the laboratory instrumentation, maximizing scientific productivity and enhancing the overall quality of DEL processes. Through this approach, we maximize reproducibility of the work by ensuring that experimental protocols captured by scientists reflect the actual manipulations carried out in the laboratory. In addition, we improve scientific productivity by automating a number of sample and analytical data gathering and processing tasks and enhance the overall quality of our DEL by reducing the number of errors resulting from manual execution of the complex liquid handling routines that are required.”

The paper describes the production and affinity selection of DEL-6, a three-cycle library encoded in double-strand DNA and incorporating a triazine central core. The authors demonstrate DELTA’s utility as a model for others planning to start their own DEL program.

“DELT is not a single technology method,” says Goodnow. “The Eli Lilly authors describe a means to automate the creation of multiple libraries in an efficient process. To date, Eli Lilly has not published a lot in this area, so it is quite interesting to see their approach.”

Improve Identification of DNA Tags

“tagFinder: A Novel Tag Analysis Methodology That Enables Detection of Molecules from DNA-Encoded Chemical Libraries” reports a methodology to “deconvolute encoding oligonucleotides, thus optimizing the sequencing power regardless of the library size, design complexity or sequencing technology chosen.” Authors are from Instituto de Investigaciones Sanitarias (IDIS) (A Coruña, Spain), Centro Singular de Investigación en Medicina Molecular y Enfermedades Crónicas (CIMUS) (A Coruña, Spain), Eli Lilly and Company (Alcobendas, Madrid, Spain) and Lilly Corporate Center (Indianapolis, IN).

This article provides a thorough description of a methodology for identification of DNA tags independent of the sequencing technology chosen. The authors state it is more accurate than previously described methods, requires less computing resources, shows an important decrease of existing running times while reducing the overall error and conclude “its high degree of customization allows the identification of libraries with a different number of codification cycles, different sequencing lengths and the discrimination of degenerated regions to identify the occurrence of unique sequences based on the specific combination of codifying and degenerated sequences. This algorithm could be easily implemented in the analysis workflow of any DEL platform for the fast, efficient and accurate detection of codifying DNA tags.”

“DELT does not consist of a single DNA encoding method,” says Goodnow. “Rather, the robustness of DNA as an information storage medium has allowed for the evolution of different DNA architectures and constructs; thus, it is instructive to learn of tagFinder.”

Calculate for Success

A team from GlaxoSmithKline (Cambridge, MA) studies the robustness of DEL selection by examining the sequencing readouts of warheads and chemotype families among a large number of experimentally repeated selections. As reported in “Randomness in DNA Encoded Library Selection Data Can Be Modeled for More Reliable Enrichment Calculation,” the results reveal that the output of DEL selection is intrinsically noisy but can be reliably modeled by the Poisson distribution, and that Poisson noise is the dominating noise at low copy counts and can be estimated even from a single experiment.

Authors Letian Kuai, Thomas O’Keeffe and Christopher Arico-Muendel discuss the “shortcomings of data analyses based on directly using copy counts and their linear transformations and propose a framework that incorporates proper normalization and confidence interval calculation to help researchers better understand DEL data.”

“Kuai and other scientists at GSK explain that despite the noisiness of selection informatics, results can be modeled accurately according to a Poisson distribution,” Goodnow explains. “Although the Poisson distribution concept has been noted for DELT selections before, this thorough analysis increases confidence in the reproducibility of the selection outputs. GSK scientists also report on information methods for decoding of DEL selections.”

Z’ for DNA-Encoded Selection

“Robustness of In Vitro Selection Assays of DNA-Encoded Peptidomimetic Ligands to CBX7 and CBX8” finds collaborators from Purdue University and Purdue University Center for Cancer Research (West Lafayette, IN) and the University of Victoria (Victoria, British Columbia, Canada) studying the statistical details of selection. Kyle E. Denton, Sijie Wang, Michael C. Gignac, Natalia Milosevich, Fraser Hof, Emily C. Dykhuizen and Casey J. Krusemark report “a method for optimizing and utilizing affinity selection assays to identify potent and selective peptidic ligands to the highly related chromodomains of CBX proteins. To optimize affinity selection parameters, statistical analyses (Z′ factors) were used to define the ability of selection assay conditions to identify and differentiate ligands of varying affinity.”

“I’d often said it was important that for the technology to mature, we needed to have a Z’ calculation,” says Goodnow. “Z’ is a measure of assay quality. I’m very happy the special collection includes a paper with focus around calculation of a Z’ for DNA-encoded selection. To me that means the approach is moving from concept phase to multi-disciplinary professional phase.”

The authors conclude their work lays “the groundwork for future studies exploring additional modifications to improve potency and selectivity toward CBX8 against other CBX ChD proteins. Broadly, our results show how carefully conducted in vitro selection assays can be useful not only for ligand discovery but also for finer-grained determination of quantitative structure-activity relationships.”

Working with Modified Protein Targets

“Many DELT publications focus on the identification of high-affinity ligands for targets of biological interest,” says Goodnow. “Such targets must usually be soluble proteins. In an intersection of cutting edge technologies, authors at X-Chem, working with AstraZeneca and Heptares scientists, report on selection against a mutant stabilized PAR2 GPCR. Such targets are normally outside the reach of DELT selections, but these authors show the promise of working with such modified protein targets.”

“Agonists and Antagonists of Protease-Activated Receptor 2 Discovered within a DNA-Encoded Chemical Library Using Mutational Stabilization of the Target” also includes collaborators from Novartis Institute for BioMedical Research (Cambridge, MA and Basel, Switzerland), École Polytechnique Fédérale de Lausanne (Lausanne, Switzerland), Morphic Therapeutic (Waltham, MA) and Entasis Therapeutics (Waltham, MA).

The authors believe this report “illustrates the power of thermostabilized G-protein-coupled receptors (StaRs) for affinity-mediated selection of DNA-encoded chemical libraries. Selection using a protease-activated receptor 2 StaR yielded both the first example of a DECL-derived G-protein-coupled receptor (GPCR) agonist and a novel series of allosteric antagonists. We suggest that these results indicate that future use of StaRs with DECL technology will be a fruitful approach to the discovery of novel modulators of GPCRs.”

Multi-Disciplinary; Multi-Beneficial

Goodnow is excited about DELT and its future as a multi-disciplinary science. “You can’t be just a chemist or just a biologist or just a technician; you have to understand different aspects of all the fields,” he says. “Project coordination efforts appeal to me – I want to be able to bring it all together. DELT changes the way in which people can explore chemical space to find new hits, new compounds and new molecules for therapeutic targets for unmet medical needs. It can make an impact. Pharma has engaged in a very long-term effort toward finding solutions. I get really excited when I see opportunities to make things better.”

Goodnow hopes the stories of DELT successes reported in this special collection spur more attempts and, therefore, more successes.

“Case stories of successes are really important; that’s what energizes interest and investment,” he says. “We’ve shown examples of successes in this special collection but continued examples would be helpful to continue our movement forward.”

May 21, 2018