DNA-Encoded Library (DEL) Screening

Each unique small molecule in a DEL is covalently linked to a distinct, amplifiable DNA sequence that serves as a molecular barcode. This DNA tag records the entire synthetic history of the compound. After affinity selection against an immobilized protein target, the identity of bound molecules is not determined by traditional analytical chemistry but by PCR amplification and high-throughput DNA sequencing of the attached barcodes. This transforms a complex chemical identification problem into a simple, massively parallel genetic readout.

DELs are constructed using a highly efficient split-and-pool synthesis strategy. The process involves:

• Split: Dividing a bead-bound precursor into multiple wells for a first round of distinct chemical coupling.
• Ligate: Attaching a short DNA tag to record that specific coupling event.
• Pool: Recombining all the beads. Repeating this cycle for 3-4 steps generates a combinatorial explosion of diversity. A library built from 1,000 building blocks over 3 cycles yields a theoretical library size of 1 billion (10^9) unique compounds, each with its own unique genetic barcode.

The screening process is a physical, in vitro selection rather than a computational one. The entire DEL library—often containing billions of compounds—is incubated with an immobilized protein target of interest. Non-binding molecules are washed away under stringent conditions, while target-binding molecules are retained. The bound fraction is then heat-eluted, and the DNA barcodes are sequenced. The relative enrichment of specific barcode sequences after selection directly correlates with the binding affinity of the corresponding small molecule, identifying high-affinity hits from a sea of inactives.

DEL technology overcomes the one-compound-per-well limitation of traditional high-throughput screening (HTS). A single DEL screen can interrogate billions of compounds in a single test tube within a few days, at a fraction of the cost and time of HTS. This massive scale allows for the exploration of vast tracts of chemical space that are inaccessible to other methods, dramatically increasing the probability of discovering novel chemical matter for challenging or previously undruggable targets.

A key constraint of DEL technology is that all chemical reactions must be compatible with the presence of the DNA barcode. The DNA tag must remain stable and amplifiable under the synthesis conditions, which limits the reaction palette to DNA-compatible chemistry. This predominantly includes aqueous-phase reactions at near-neutral pH, such as amide bond formations, Suzuki couplings, and reductive aminations. The development of new DNA-compatible reactions is a critical area of research to expand the drug-like chemical diversity accessible by DELs.

The primary output of a DEL screen is a ranked list of enriched DNA barcode sequences, which are computationally translated into the corresponding small molecule structures. Because the selection is performed on DNA-tagged compounds, the initial hits must be resynthesized off-DNA—without the DNA tag—to confirm their binding and biological activity in orthogonal assays. This validation step is crucial to rule out false positives arising from DNA intercalation or non-specific binding of the DNA tag itself to the protein target.

The NGS-based readout step that decodes the identity of enriched binders after affinity selection. Following incubation with an immobilized protein target and stringent washing, the DNA tags from retained compounds are amplified via PCR and sequenced. The relative frequency of each barcode sequence in the output sample compared to the input library quantifies enrichment, directly correlating to binding affinity. This massively parallel readout is what enables DELs to screen billions of compounds in a single tube.

The foundational combinatorial chemistry method for constructing DELs. The process iteratively splits a bead or solution pool into aliquots, reacts each with a unique building block and a corresponding DNA tag, and then re-pools them. This creates a one-to-one correspondence between a compound's synthetic history and its attached DNA sequence. A 3-cycle library using 1,000 building blocks per step yields 1 billion (1,000³) distinct compounds, each with a unique amplifiable barcode.

An umbrella term encompassing all screening platforms that use a chemical tag to record synthetic history, including DNA-encoded libraries (DELs), peptide nucleic acid-encoded libraries, and DNA-encoded dynamic libraries. ELT extends the concept beyond standard DNA barcodes to other information-rich polymers. The core principle remains: genotype-phenotype linkage at the single-molecule level, enabling pooled screening and amplification-based readout of chemical structures.

The constrained set of chemical reactions that can be performed in the presence of unprotected DNA without causing depurination, strand scission, or cross-linking. Reactions must be aqueous, operate near neutral pH, and avoid strong oxidizers or electrophiles that modify nucleobases. This limitation historically restricted DELs to amide bond formations and simple heterocycle syntheses, but recent advances in micellar catalysis and photoredox chemistry have dramatically expanded the accessible reaction space.

The computational pipeline that translates raw sequencing data into actionable hit lists. Key steps include: barcode counting to measure enrichment, normalization against library-wide background, statistical confidence scoring (e.g., z-scores or p-values), and structure enumeration from decoded synthetic routes. Advanced analysis uses machine learning to identify structure-activity relationships (SAR) directly from the selection fingerprint, predicting compounds that were never physically synthesized.