DEL hit validation

Measure DNA-encoded library (DEL) hits on their DNA. Each hit finds its own molograms by its DNA, and you get kinetics for many hits in one run, before any off-DNA synthesis.

Illustration of a target protein in purple with a small molecule in orange bound in its pocket; a linker leads from the small molecule to a blue DNA double helix.
A DEL hit (orange) in the pocket of its target (purple), still attached to its DNA (blue).

The question

Which of my DEL hits bind the target, how strongly and how fast, and do they bind anything else?

A DEL selection gives a list of enriched barcodes, not affinities. The next step is to decide which hits deserve the work of off-DNA synthesis. That decision is easier with measured kon, koff and KD for many hits, side by side. Two published papers show how this works.

The setup

ChipDNA chip. Each oligo address sits on its own molograms, so several compounds share one chip
Hit on the chipThe hit keeps its DNA. The Oligo Adapter 5′-Phosphate is ligated to the DNA of a hit; the Oligo Adapter Amine is for on-DNA resynthesis
AnalyteThe protein flows over the chip. Study 1: carbonic anhydrase IX (CAIX). Study 2: fibroblast activation protein (FAP) and five off-target proteins. Study 3: VHL
Study 1: CAIXThe reference inhibitor AAZ and two DEL ligands, SABA and mSABA. CAIX in 11 steps from 0.36 nM to 7.1 µM, seven replicate molograms per compound
Study 2: FAPThe FAP ligand oncoFAP on DNA, in four constructs on a DNA chip with a PEG background, 16 molograms each. Code C04: purified, with a single-stranded DNA end. Codes C01 to C03: crude ligation products with a double-stranded DNA end. FAP from human (hFAP), mouse (mFAP) and dog (cFAP), 195 pM to 200 nM. CAIX, CAII, HSA, BSA and CD3 as off-target proteins, up to 100 µM
Study 3: VHL20 DNA-VHL ligand conjugates (DNA-PROTACs) that differ in their linker, pooled and immobilized together. Each takes three molograms. VHL from 16 to 500 nM
AssaySingle-cycle kinetics (SCK): rising concentrations without regeneration in between
Ligate, pool, sort, rank, check selectivity

Eight hits on an 8-plex DNA chip are shown, eight molograms each. Play the workflow or step through it.

Measured result

CAIX: two DEL ligands ranked against a reference inhibitor

Isoaffinity map with k_off on the horizontal axis and k_on on the vertical axis, both logarithmic, and dashed lines of equal K_D. Seven blue AAZ points cluster near 1 nM at the upper left. Seven orange SABA points cluster between 100 nM and 1 µM. Seven green mSABA points cluster between 1 and 10 µM at the lower right.
CAIX binding to AAZ, SABA and mSABA, single-cycle kinetics. Each point is one replicate fit, seven per compound; dashed lines mark equal KD. Data from Favalli et al., J. Med. Chem. 2026, and the app note on DEL hit validation.

FAP: on-target, cross-species and off-target binding

Heatmap with eight proteins as rows and four oncoFAP-DNA constructs as columns. Only the C04 column, with a single-stranded DNA end, lights up: 21 pg/mm² for human FAP, 25 for mouse FAP and 14 for dog FAP. All other cells, including CAIX, CAII, HSA, BSA and CD3, stay between minus 2.6 and plus 0.2 pg/mm².
Response of oncoFAP-DNA constructs at the highest analyte concentration, above the preceding baseline: 200 nM for the FAP orthologs, 100 µM for the off-target proteins. Mean of 16 molograms per construct. Data: Favalli et al., J. Med. Chem. 2026, and the app note on selectivity profiling.

VHL: 20 DNA-PROTAC linkers in one run

Sensorgrams of all 64 molograms over 78 minutes. The pooled conjugates bind during the green immobilization phase. Six VHL injections from 16 to 500 nM then raise the signal step by step, by a different amount on each conjugate. The blank address stays flat. Regeneration at the end returns the traces to their starting level.
VHL single-cycle kinetics on 20 DNA-VHL ligand conjugates, run SCK-4. Response of all 64 molograms in pg/mm²; each conjugate sits on three molograms, and 00 is a blank address. The pool is immobilized during the green phase and removed by regeneration at the end. Data from Raschke et al., RSC Chem. Biol. 2026.

How to read it

Why it works

Each mologram carries one oligo sequence on its ridges, so a hit with the matching DNA binds only to its own molograms. The protein is the analyte, so the signal comes from the protein mass that binds the hit, not from the small molecule itself. The readout counts only mass that binds in the line pattern of the mologram.

Read more: capture chemistry on ridges, matched grooves, what sets the detection limit, and from focal spot to coherent mass density.

More