DNA-Encoded Libraries vs. Fragment-Based Screening: Pros and Cons
Executive Summary
DNA-encoded library (DEL) screening and fragment-based drug discovery (FBDD) have emerged as two leading alternatives to conventional high-throughput screening for small-molecule hit identification. Both strategies deliver novel chemotypes, yet they diverge in scale, logistics and the biological insight they deliver.
This article unpacks how each technology works, weighs their respective strengths and limitations, and illustrates how Vipergen’s cellular Binder Trap Enrichment (cBTE) extends the reach of DEL screening from purified protein to live-cell target engagement.
Readers working in pharmaceutical R&D will gain a clear decision framework for selecting the most time- and cost-effective hit-finding path for a given target.
1. Technology Primers
1.1 Fragment-Based Screening (FBS/FBDD)
Fragment‐based approaches purposely start small. Libraries of 1 000–2 000 low-molecular-weight compounds (typically 150–300 Da, following the Rule of Three) are screened individually by highly sensitive biophysical methods—most commonly nuclear magnetic resonance (NMR), surface plasmon resonance (SPR), thermal shift assays, or X-ray crystallography (Erlanson 2016). Fragments bind weakly (µM–mM), but because they are so small they often sample binding hot-spots that large molecules miss. Once weak binders are confirmed, medicinal chemists apply grow, merge, or link strategies, using structural guidance to build potency and selectivity.
Importantly, each fragment needs to be pre-selected for synthetic tractability: the molecules contain versatile functional handles and follow well-established chemistry rules, so medicinal chemists can rapidly elaborate any hit into dozens of analogues for systematic structure–activity relationship (SAR) exploration without having to develop new synthetic routes.
1.2 DNA-Encoded Library (DEL) Screening
DNA-encoded libraries take the opposite stance: instead of depth, they go broad. Each small molecule is covalently tagged with a unique DNA barcode that records its synthetic lineage, enabling billions of distinct compounds to be pooled in a single test tube while preserving identity (Goodnow 2017).
In a typical selection the library is incubated with the target, non-binders are washed away, the retained DNA tags are PCR-amplified, and next-generation sequencing (NGS) reveals which chemotypes were enriched. Because each compound carries its own barcode, millions can be screened in parallel. In practice we have found that libraries in the hundred-million range (Vipergen routinely deploys 400–600 million members) already capture the vast majority of accessible chemical motifs, delivering low-nanomolar binders without the diminishing returns that accompany ever-larger pools.
With Vipergen’s YoctoReactor® chemistry we routinely build libraries in the 10^8 scale; our Binder Trap Enrichment (BTE) workflow keeps selections in solution, and cellular Binder Trap Enrichment (cBTE) moves the whole screen into living mammalian cells, bypassing the purified-protein requirement through direct intracellular delivery.
2. Comparative Pros & Cons
Project teams must weigh multiple factors—throughput, protein burden, chemical starting point, data overhead and cost—when choosing a hit-finding strategy. The table below provides a concise, side-by-side comparison.
| Aspect | DNA-Encoded Libraries | Fragment-Based Screening |
|---|---|---|
| Throughput & speed | Screens ~10^8 compounds as quickly as in one month (Indsæt link til SNAP-Easy-Fast-and-Affordable); sequencing replaces manual hit picking. | ≤ 2 k measurements; several months of instrument time. |
| Protein requirement | 10–50 µg purified protein; zero with Vipergen cBTE (live-cell selection). | mg quantities of highly pure, often crystallisable protein. |
| Hit affinity window | nM–low-µM primary hits thanks to massive diversity. | mM–high-µM; potency built later by medicinal chemistry. |
| Chemical diversity | Millions of drug-like, 3D-rich scaffolds, sp³-rich motifs. | Smaller, “privileged” library; chemistry is immediately tractable. |
| Physicochemical start point | Average MW 300–600 Da (DNA linker included); re-synthesis off-DNA obligatory. | MW ≤ 300 Da; ample property space to grow into a drug. |
| Target class coverage | Soluble proteins, PPIs, RNA and—with cBTE—membrane or multiprotein complexes in situ. | Best for targets compatible with immobilization or crystallography; challenging for IDPs & membranes. |
| Data load & analytics | Millions of reads demand bioinformatics but enable robust statistics. | Dozens of spectra; interpretation is direct but manual. |
| Cost of entry | Library access + NGS are modest; CRO option eliminates capital expenditure. | High capital expenditure for NMR, SPR, X-ray and protein production pipeline. |
| IP positioning | Novel scaffolds → broad composition-of-matter claims. | Often reiterates known cores; IP strength comes from structure-guided vectors. |
Key takeaway: DEL delivers unrivalled chemical breadth with minimal protein burden, whereas FBS provides immediate structural insight when those resources are available.
3. Vipergen’s cBTE: Turning DEL Inside-Out
Traditional DEL selections stop at purified protein. Vipergen’s proprietary cellular Binder Trap Enrichment (cBTE) enables selection inside living Xenopus laevis oocytes—large amphibian cells that can be readily microinjected:
- No purification bottleneck – screening can begin after expression has been validated in cells.
- Physiological context – the target protein is expressed and screened in a crowded cytosol with native post-translational modifications and cofactors.
- Direct intracellular delivery – microinjection bypasses membrane permeability altogether; hit recovery is driven purely by target affinity, not by passive cell entry.
A 194-million-member proof-of-concept screen identified multiple low-nanomolar chemotypes against three therapeutically relevant targets (p38α, ACSS2, DOCK5) inside live oocytes (Petersen 2021).
4. Where Fragment-Based Screening Still Shines
FBDD remains unmatched when atomic-resolution data or ultra-shallow pockets are essential:
- Structure-guided design from Month 1 – co-crystal or cryo-EM structures of fragments provide an atom-by-atom roadmap for medicinal chemistry.
- Tackling cryptic or allosteric sites – small fragments can insinuate into transient pockets that larger molecules may miss.
- Rapid SAR loops – once a fragment is validated, chemists can iterate monthly with clear biophysical feedback.
Albeit this, the workflow is still labor-intensive – each fragment must be screened and validated one-by-one across multiple biophysical platforms, and every confirmed hit then demands significant medicinal-chemistry effort for grow/merge/link optimization, often requiring many synthesis cycles before potency reaches the desired threshold.
5. Decision Framework
The table below summarizes when each technology is likely to deliver the greatest return on investment.
| Decision Question | DEL (with cBTE) | FBDD |
|---|---|---|
| Is purified protein hard or time-consuming to obtain? | Yes – proceed with cBTE | No advantage |
| Is atomic-resolution structural data required at the hit stage? | Not essential | Yes – choose FBDD |
| Target-to-hit timeline ≤ 6 months? | Faster | Slower |
| Do you already have biophysics infrastructure and expertise in-house? | Not critical | Yes – FBDD capitalises on it |
In short, if protein supply and biophysical infrastructure are readily available, then FBDD might be the optimal starting point for a drug discovery campaign. On the other hand, if the protein of interest is difficult to access in larger quantities (e.g. membrane proteins) and is difficult to crystallize, then DEL screening might be the method of choice.
Bottom line: For targets where protein supply or cellular relevance is a challenge, Vipergen’s cBTE-enabled DEL workflow offers the most direct path to potent, cell-active leads within a pragmatic month-scale timeline.
6. Regulatory & IP Considerations
Regulatory agencies evaluate small-molecule drugs on the same fundamental criteria—safety, efficacy, and manufacturing quality—regardless of how the initial hit was discovered. However, there are practical distinctions:
DEL hits and composition-of-matter (COM) claims – The massive chemical diversity inherent to DEL allows discovery of previously unreported scaffolds. Sponsors often secure broad COM patents that cover the core chemotype plus close analogues, providing a wide moat around early leads.
FBDD and structure-guided claims – Fragment-derived series frequently build on well-known heteroaromatic cores. IP strength therefore comes from claiming specific vectors and 3-D binding modes revealed by crystallography. Claims can be narrower unless novel chemistry is introduced during fragment growth.
Regulatory familiarity – Both technologies have produced clinical candidates, but agency reviewers now see DEL-origin molecules more frequently, and the first DEL-discovered drugs have entered Phase II. Meanwhile, nine FDA-approved small-molecule drugs trace roots to FBDD, demonstrating clear regulatory precedent.
Freedom-to-operate (FTO) – Because DEL libraries contain unique scaffolds synthesized in-house, FTO risk is often lower than for fragment hits that may overlap with prior art. Early IP landscaping should still be carried out in either case.
Taken together, DEL can deliver a stronger IP position out-of-the-gate, while FBDD’s structural clarity speeds claim drafting for specific binding interactions.
References
- Erlanson D. A. et al., Twenty years on: the impact of fragments on drug discovery Nat. Rev. Drug Discov. 2016, 15, 605–619. doi.org/10.1038/nrd.2016.109
- Goodnow R. A. et. al., DNA-encoded chemistry: enabling the deeper sampling of chemical space, Nat. Rev. Drug Discov. 2017, 16, 131–147. doi.org/10.1038/nrd.2016.213
- Petersen L. K. et al., Screening of DNA-Encoded Small-Molecule Libraries inside a Living Cell, J. Am. Chem. Soc. 2021, 143, 2751-2756. doi.org/10.1021/jacs.0c09213
Related Services
| Service | |
|---|---|
Small molecule drug discovery for even hard-to-drug targets – identify inhibitors, binders and modulators | |
Molecular Glue Direct | |
PPI Inhibitor Direct | |
Integral membrane proteins | |
Specificity Direct – multiplexed screening of target and anti-targets | |
Express – optimized for fast turn – around-time | |
Snap – easy, fast, and affordable |