2026 Harvey - AF-M nanobody binders to GPCR
AF-M plus interface confidence metrics can enrich true GPCR nanobody binders from a synthetic VHH library, but the paper’s own benchmark says this does not yet transfer cleanly to soluble proteins or non-GPCR membrane proteins. For STRC, this is therefore a method caution and possible secondary interface-scoring gate, not a new therapeutic hypothesis.
Citation
Harvey EP, Smith JS, Hurley JD, et al. “In silico discovery of nanobody binders to a G-protein coupled receptor using AlphaFold-Multimer.” Nature Communications, 2026. DOI: 10.1038/s41467-026-72093-5.
Numbers that matter
| Parameter | Value | Units | Source | STRC use |
|---|---|---|---|---|
| GPCR benchmark positives | 32 | nanobody-target pairs | Results, benchmarking set | Positive-control scale for method confidence |
| GPCR benchmark negatives | 127 | permuted non-binder pairs | Results, benchmarking set | Negative-control construction pattern |
| Non-GPCR membrane positives | 17 | nanobody-target pairs | Results, benchmarking set | Shows weak transfer to non-GPCR membrane proteins |
| Non-GPCR membrane negatives | 376 | permuted non-binder pairs | Results, benchmarking set | Negative-control construction pattern |
| Soluble-protein positives | 49 | nanobody-target pairs | Results, benchmarking set | Shows weak transfer to soluble targets |
| Soluble-protein negatives | 1469 | permuted non-binder pairs | Results, benchmarking set | Negative-control construction pattern |
| Interface contact cutoff | ⇐10 | A C-alpha distance | Results, metric parser | Reusable for AF3/AF-M interface feature extraction |
| GPCR AUROC, average pTM | 0.73 | AUROC | Results, Fig. 2 / Supp. Table 2 | Best single GPCR discriminator reported |
| GPCR AUROC, LCF | 0.71 | AUROC | Results, LCF paragraph | Composite score is robustness-oriented, not best single metric |
| GPCR top-5% precision | 0.93-1.0 | AUC5% | Results, Fig. 2 | Useful only in high-rank screening regime |
| Soluble/non-GPCR AUC5% | ⇐0.22 | AUC5% | Results, Fig. 2 | Direct warning against applying as STRC-binder discovery evidence |
| Prospective library size | 10000 | VHH sequences | Results, MRGPRX2 screen | Scale required for enrichment |
| Liability-filtered hit threshold | 179 / 10000 | candidates | Results, MRGPRX2 screen | 1.79% exceeded best GPCR negative-control LCF |
| Extracellular predicted hits | 177 / 179 | candidates | Results, MRGPRX2 screen | Topology filter for receptor-targeting screens |
| Expressed candidates | 10 | nanobodies | Results, MRGPRX2 screen | Wet-lab validation scale after compute triage |
| Validated top binders | 3 | nanobodies | Results / Table 1 | Ranks 1, 5, 7 validated as binders |
| Sim8619 Kd | 200 +/- 20; 100 +/- 10 | nM | Table 1, HEK293T; ROSA | Nanomolar validation anchor |
| Sim9877 Kd | 160 +/- 30; 20 +/- 4 | nM | Table 1, HEK293T; ROSA | Nanomolar validation anchor |
| Sim4784 Kd | 80 +/- 30; 50 +/- 10 | nM | Table 1, HEK293T; ROSA | Nanomolar validation anchor |
Method essentials
- AF-M predictions used local ColabFold / AF-M v2 or v3, 3 recycles, no templates, no Amber relaxation.
- Metrics combined into LCF: average pTM, best model pTM, average interface PAE, best model interface PAE, average interface pLDDT, best model interface pLDDT. Best pDockQ was reported but excluded from LCF as redundant with pLDDT.
- Negative controls were generated by permuting nanobody-antigen pairs, then checking that antigens were not too closely related by BLAST.
- Developability liabilities were filtered before expression: CDR glycosylation sites and predicted polyreactivity.
- Code:
kruselab/MRGPRX2-AF-M-screen, MIT license, main commit observed 2025-03-18.
Limitations
- Target-class limitation is explicit: GPCR works; soluble and non-GPCR membrane targets do not show useful enrichment in the paper’s benchmark.
- The GPCR advantage likely depends on the large and stereotyped GPCR-nanobody structure set in PDB.
- The method is a binder-enrichment/ranking workflow, not an affinity predictor for arbitrary protein-protein interfaces.
- It requires a large VHH design library and wet-lab expression/cell-binding assays to validate hits.
Relevance to STRC
- h09: applicable only as a secondary scoring discipline for AF3/AF-M interface predictions and negative controls. It should not override the current Phase 4m/4l/4k gates because h09 is peptide/actin/TMEM145, not GPCR nanobody discovery.
- h26: not actionable until a structural candidate passes the homodimer gate. It does not rescue current h26 failures because the paper says non-GPCR transfer is weak.
- New VHH therapy hypothesis: not warranted. A nanobody against TMEM145/STRC would add delivery, immunogenicity, size, and validation burdens while relying on a method the paper does not validate for this target class.
Connections
[part-of]avidity-and-dimers[applies]STRC Synthetic Peptide Hydrogel HTC[applies]STRC Engineered Homodimer Avidity[see-also]MRGPRX2 AF-M screen[see-also]2026-sang-af3-turboab-antibody-structural-prediction[source]Harvey 2026 - In silico discovery of nanobody binders to a G-protein coupled rec