// Experimental — provider-preview capability
Ligand-conditioned co-folding
OpenFold3-preview lets us extend the platform’s apo pocket assessment with ligand-conditioned holo prediction. Below are seven canonical targets — five RNA, two protein — where experimental holo structures exist and can serve as ground truth, plus a case study on a target whose co-folded result initially looked like a clean failure. Metrics are honest: apo baseline vs. co-folded prediction vs. experimental holo, using consistent measurement definitions per target, full replicate distributions, and no hidden failures.
// 1 of 7 — The easy case
Risdiplam / SMN2 splice-site duplex
A short, mostly Watson-Crick-paired RNA duplex — the bulged 5’-splice-site element risdiplam binds to redirect SMN2 exon 7 splicing. Sets up the rest of the page: not every RNA target needs co-folding to get the fold right.
Ligand class
Splicing modifier (approved drug — the SMA therapy risdiplam)
Apo pocket-detection (this platform’s standard pipeline)
Groove binder — cleft-based detection isn’t in scope yet; public binding site used directly.
What that means here
Our cavity-based pocket detector (fpocket) is tuned for clefts, not shallow grooves — groove-binder detection isn't in scope yet. For this cross-validation we use the public, well-studied binding site from the published structure (8R62) directly, rather than our own detection.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 0.34 Å (±0.11) | 0.22 Å (±0.06) | +0.12 Å |
| Global (whole-molecule) RMSD | 2.05 Å (±0.18) | 1.82 Å (±0.09) | +0.23 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: risdiplam_apo_seed_958682846_sample_2_model.cif (0.28 Å), co-folded: risdiplam_cofold_seed_1181241943_sample_2_model.cif (0.19 Å). Full distribution above.
// Ligand identity
- Name
- Risdiplam
- Formula / MW
- C22H23N7O · 401.2 Da
- SMILES
- CC1=CC(=NN2C1=NC(=C2)C)C3=CC(=O)N4C=C(C=CC4=N3)N5CCNC6(C5)CC6
// Ground truth reference
PDB 8R62 · 2024 · Solution NMR
"The diversity of splicing modifiers acting on A-1 bulged 5'-splice sites reveals rules for rational drug design." (2024).
// Query sequence source
Both 11-nt strands extracted directly from 8R62's own resolved chains A and B (chain A's 2 pseudouridine residues substituted with standard U for the OpenFold3 query — noted, not hidden). Chain B offset +11 applied uniformly for a single unambiguous residue-numbering space.
// Interpretation
Apo prediction already lands close to the true structure — a short, mostly Watson-Crick-paired duplex is not a hard fold to get right blind. Co-folding's improvement is real and consistent across every replicate (+0.12 Å binding-site RMSD), but small, because there isn't much conformational gap left to close. This is the expected shape of result when the apo baseline is already strong, not a limitation of co-folding itself — the same pattern shows up on FKBP12 below.
// 2 of 7 — The clean canonical win
HCV IRES IIa
A 38-nt hairpin domain of the hepatitis C virus internal ribosome entry site — a real, structured antiviral RNA target with a benzimidazole-class small-molecule binder. Shows what co-folding adds when there is genuine ligand-induced conformational change to recover.
Ligand class
Benzimidazole-class HCV IRES IIa binder
Apo pocket-detection (this platform’s standard pipeline)
Rank 1, 58% binding-site overlap — strict recovery.
What that means here
Apo pocket-detection lands the real site at rank 1, strict recovery, before any co-folding.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 5.48 Å (±0.17) | 0.36 Å (±0.06) | +5.13 Å |
| Global (whole-molecule) RMSD | 6.09 Å (±0.37) | 0.70 Å (±0.14) | +5.39 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: hcv_iresiia_apo_seed_2746317213_sample_2_model.cif (5.40 Å), co-folded: hcv_iresiia_cofold_seed_2746317213_sample_2_model.cif (0.34 Å). Full distribution above.
// Ligand identity
- Name
- ISH (benzimidazole-class)
- Formula / MW
- C17H27N5O · 317.2 Da
- SMILES
- CN(C)CCCn1c(N)nc2ccc3O[C@@H](CN(C)C)Cc3c12
// Ground truth reference
PDB 2KTZ · 2010 · Solution NMR
Paulsen, Seth, Swayze, Griffey, Skalicky, Cheatham, Davis. "Inhibitor-induced structural change in the HCV IRES domain IIa RNA." PNAS 107 (2010).
// Query sequence source
Extracted directly from 2KTZ's own resolved chain A (1-38, no gaps) — used as both the query-sequence source and the holo reference, so registration is exact by construction.
// Interpretation
Clean, decisive recovery of both the fold and the ligand pose. Apo prediction already finds the right region (rank-1 pocket, strict recovery) but misses the binding-site geometry by ~5.5 Å; co-folding brings that to sub-Å accuracy (0.36 Å), with the ligand itself landing within 0.75 Å of its real experimental position. This is the target where ligand-conditioned co-folding does exactly what it's meant to do — resolve a real, otherwise-unrecoverable conformational gap on a simple hairpin fold.
// 3 of 7 — The honest partial case
FMN riboswitch
A 56-nt three-way-junction FMN riboswitch aptamer — a harder RNA tertiary fold than a simple hairpin, bound to a synthetic flavin-core cofactor analog. The trust-building beat: co-folding helps, it does not solve this one.
Ligand class
Flavin-core (isoalloxazine) cofactor analog
Apo pocket-detection (this platform’s standard pipeline)
Rank 1, 22% binding-site overlap — below near-recovery threshold (rank 2: 33%, near).
What that means here
Apo pocket-detection's rank-1 cluster misses the real site (22% overlap); the real site is closer to rank-2 (33%, near-recovery). Consistent with this being a harder three-way-junction tertiary fold.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 6.85 Å (±0.53) | 4.80 Å (±0.40) | +2.06 Å |
| Global (whole-molecule) RMSD | 17.79 Å (±0.58) | 12.44 Å (±0.41) | +5.36 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: fmn_riboswitch_apo_seed_958682846_sample_3_model.cif (6.73 Å), co-folded: fmn_riboswitch_cofold_seed_958682846_sample_1_model.cif (4.75 Å). Full distribution above.
// Ligand identity
- Name
- GZG (flavin-core cofactor analog)
- Formula / MW
- C25H27N5O4 · 461.2 Da
- SMILES
- Cc1cc2N=C3C(=O)NC(=O)N=C3N(CCN(CCCC(O)=O)Cc4ccccc4)c2cc1C
// Ground truth reference
PDB 6DN2 · 2018 · X-ray crystallography
"Crystal structure of the FMN riboswitch bound to BRX1354 split RNA." (2018). Closest real match to “recently published” — no post-2023 FMN/RFN-element riboswitch-ligand structure exists in the PDB (checked directly).
// Query sequence source
Extracted directly from 6DN2's own resolved chain Y (56 nt, numbered 57-112 in the deposited file; offset -56 applied and verified before computing anything).
// Interpretation
Co-folding moves the prediction in the right direction — both the overall fold and the pocket region improve (+2.06 Å binding-site RMSD) — but neither arm reaches a usable absolute accuracy on this three-way-junction tertiary fold, and the ligand pose itself isn't reliably placed (8.1 Å mean). A harder structure-prediction problem than a simple hairpin, and co-folding alone doesn't close the gap. The honest read: this is a case for downstream refinement (e.g. short MD) on top of the co-folded output, not a final answer by itself.
// 4 of 7 — Protein, decisive
ABL1 kinase + imatinib (DFG-out)
The ABL1 kinase domain bound to imatinib (Gleevec), the classic DFG-out inhibitor complex. Shows the capability isn’t RNA-only — and isolates exactly what co-folding adds on a protein target where the pocket already exists in apo, but its conformational state does not.
Ligand class
ATP-competitive kinase inhibitor (DFG-out, approved drug)
Apo pocket-detection (this platform’s standard pipeline)
Rank 1, 76% binding-site overlap — strict recovery.
What that means here
Apo pocket-detection already finds the ATP-binding cleft at rank 1 — the cleft exists regardless of DFG-in/DFG-out state. Co-folding's value here is resolving the DFG-out conformational state itself, not finding the pocket.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 2.21 Å (±0.03) | 0.72 Å (±0.61) | +1.49 Å |
| Global (whole-molecule) RMSD | 4.97 Å (±0.10) | 1.16 Å (±0.21) | +3.81 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: abl1_apo_seed_958682846_sample_3_model.cif (2.21 Å), co-folded: abl1_cofold_seed_1181241943_sample_3_model.cif (0.40 Å). Full distribution above.
// Ligand identity
- Name
- Imatinib (Gleevec / STI-571)
- Formula / MW
- C29H31N7O · 493.3 Da
- SMILES
- CN1CCN(CC1)Cc2ccc(cc2)C(=O)Nc3ccc(C)c(Nc4nccc(n4)c5cccnc5)c3
// Ground truth reference
PDB 1IEP · 2002 · X-ray crystallography
Nagar, Bornmann, Pellicena, Schindler, Veach, Miller, Clarkson, Kuriyan. "Crystal structures of the kinase domain of c-Abl in complex with the small molecule inhibitors PD173955 and imatinib (STI-571)." Cancer Res 62 (2002).
// Query sequence source
Extracted directly from 1IEP's own resolved chain A (274 residues, numbered 225-498 in the deposited file) rather than sliced from UniProt — UniProt numbering disagrees with 1IEP's construct at the N-terminus (likely an expression-tag artifact), confirmed by direct comparison before running anything.
// Interpretation
Apo prediction already finds the ATP-binding cleft — that part of the pocket doesn't depend on ligand information. What apo prediction can't do is resolve the DFG-out conformational state that imatinib specifically selects for; co-folding recovers that almost completely (binding-site RMSD 2.21 Å → 0.72 Å mean, global RMSD 4.97 Å → 1.16 Å). One flexible P-loop residue doesn't converge well in 2 of 9 replicates — even those two still clearly beat the apo baseline, and the DFG motif itself is unaffected in either case.
// 5 of 7 — Protein, smaller effect
FKBP12 + FK506 (rapamycin-class)
FKBP12 bound to FK506 (tacrolimus), a rigid, well-packed immunophilin pocket with comparatively little room for ligand-induced movement. Rounds out the honest picture: co-folding’s effect size tracks how much conformational change there actually is to recover.
Ligand class
Immunophilin ligand (rapamycin/FK506-class, approved drug)
Apo pocket-detection (this platform’s standard pipeline)
Rank 3, 43% binding-site overlap — near recovery.
What that means here
Apo pocket-detection finds the real site at rank 3 (near-recovery), not top-2 — consistent with the platform's documented “top-3, never the top” pattern on this exact target.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 0.40 Å (±0.04) | 0.30 Å (±0.04) | +0.09 Å |
| Global (whole-molecule) RMSD | 0.46 Å (±0.03) | 0.36 Å (±0.01) | +0.09 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: fkbp12_apo_seed_958682846_sample_1_model.cif (0.40 Å), co-folded: fkbp12_cofold_seed_958682846_sample_3_model.cif (0.32 Å). Full distribution above.
// Ligand identity
- Name
- FK506 (tacrolimus)
- Formula / MW
- C44H69NO12 · 803.5 Da
- SMILES
- CO[C@@H]1C[C@@H](CC[C@H]1O)/C=C(C)/[C@H]2OC(=O)[C@@H]3CCCCN3C(=O)C(=O)[C@]4(O)O[C@H]([C@H](C[C@@H](C)C\C(=C\[C@@H](CC=C)C(=O)C[C@H](O)[C@H]2C)C)OC)[C@H](C[C@H]4C)OC
// Ground truth reference
PDB 1FKJ · 1995 · X-ray crystallography
Wilson, Yamashita, Sintchak, Rotstein, Murcko, Boger, Thomson, Fitzgibbon, Black, Navia. "Comparative X-ray structures of the major binding protein for the immunosuppressant FK506 (tacrolimus) in unliganded form and in complex with FK506 and rapamycin." Acta Crystallogr D 51 (1995).
// Query sequence source
UniProt P62942 canonical sequence (108 aa incl. initiator Met). 1FKJ's crystallized construct omits the initiator Met, so predicted residue N corresponds to 1FKJ residue N−1 — confirmed by direct residue-identity comparison at both chain termini after an initial off-by-one was caught independently (see /findings).
// Interpretation
FKBP12's pocket is small, rigid, and well-packed — there's comparatively little room for a ligand to induce conformational change in the first place. Co-folding still helps, consistently across all 9 replicates (+0.09 Å binding-site RMSD), but the effect size is an order of magnitude smaller than ABL1's. Read together, the two protein targets show the same pattern as the RNA ones: co-folding's benefit tracks how much real conformational change there is to recover, not a fixed capability level independent of the target.
// 6 of 7 — RNA, no free pass
DM1 / CUG repeat duplex
A short (CUG)n repeat duplex — the RNA structural expansion implicated in myotonic dystrophy type 1 — bound to a small-molecule CUG binder. The least flattering result on the page: no cherry-picking, even on the platform’s own home turf.
Ligand class
Amidine-benzimidazole CUG-repeat binder
Apo pocket-detection (this platform’s standard pipeline)
Groove binder — cleft-based detection isn’t in scope yet; public binding site used directly.
What that means here
Same groove/duplex scope gap as risdiplam/SMN2 above — the public, well-studied binding site from 9CPD is used directly rather than our own cavity-based detection.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 2.03 Å (±0.05) | 2.45 Å (±0.05) | -0.42 Å |
| Global (whole-molecule) RMSD | 1.90 Å (±0.08) | 3.88 Å (±0.29) | -1.98 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: dm1_cug_apo_seed_2746317213_sample_1_model.cif (2.03 Å), co-folded: dm1_cug_cofold_seed_958682846_sample_3_model.cif (2.46 Å). Full distribution above.
// Ligand identity
- Name
- MQC (amidine-benzimidazole CUG binder)
- Formula / MW
- C17H18N4O · 294.1 Da
- SMILES
- COc1cccc(Nc2cc(C)nc3ccc(cc23)C(N)=N)c1
// Ground truth reference
PDB 9CPD · 2024 · Solution NMR
Taghavi, Frank, Fountain, Choudhary, Roy, Childs-Disney, Disney. "Structures of small molecules bound to RNA repeat expansions that cause Huntington's disease-like 2 and myotonic dystrophy type 1." Bioorg Med Chem Lett (2024).
// Query sequence source
Both 13-nt strands extracted directly from 9CPD's own resolved chains A and B. Chain B offset +13 applied for a single unambiguous residue-numbering space, matching the convention used for the risdiplam/SMN2 duplex above.
// Interpretation
A short, mostly Watson-Crick-paired duplex — the kind of fold apo prediction already gets close to right (2.03 Å binding-site RMSD blind). Co-folding places a chemically plausible, non-clashing ligand pose, but doesn't clearly improve the RNA fold itself over the apo baseline here (2.45 Å vs. apo's 2.03 Å), and the ligand pose doesn't converge tightly to the crystal position (3.65 Å mean across replicates). We also tested downstream MD refinement (500 ps, ligand-aware) on top of the co-folded structure: real but modest further improvement (best frame 2.34 Å), not a full recovery. Included here as an honest demonstration case, not a clean win — a small, already-rigid duplex like this one leaves little conformational gap for co-folding or refinement to close, the same pattern seen on FKBP12 above.
// 7 of 7 — RNA, right chemistry, imprecise pose
HIV-1 TAR + RBT-550
HIV-1 TAR — the Tat-binding hairpin, and specifically its flexible trinucleotide bulge — bound to RBT-550, a rationally designed synthetic inhibitor that targets an alternative, "flipped-out" bulge conformation rather than the resting state. Closes the standard set on a genuinely hard case: a real pharmacophore, real conformational selection, and an imprecise atomic pose.
Ligand class
Synthetic TAR inhibitor (RiboTargets series, bulge-conformation-selective)
Apo pocket-detection (this platform’s standard pipeline)
Groove binder — cleft-based detection isn’t in scope yet; public binding site used directly.
What that means here
Same groove/duplex scope gap as risdiplam and DM1 above — TAR's binding site is a shallow, flexible bulge-adjacent groove, not a cavity our fpocket-based detector is tuned for. Public binding site used directly rather than our own detection.
// Apo vs. co-folded vs. experimental ground truth
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
// Metrics
| Metric | Apo | Co-folded | Improvement |
|---|---|---|---|
| Binding-site RMSD | 2.87 Å (±0.51) | 2.38 Å (±0.37) | +0.49 Å |
| Global (whole-molecule) RMSD | 5.63 Å (±0.60) | 3.93 Å (±0.36) | +1.71 Å |
Replicate distribution (9 apo / 9 co-folded runs)
Default view shows the representative replicate closest to each arm’s median binding-site RMSD — apo: hiv_tar_apo_seed_2746317213_sample_1_model.cif (2.91 Å), co-folded: hiv_tar_cofold_seed_1181241943_sample_3_model.cif (2.39 Å). Full distribution above.
// Ligand identity
- Name
- RBT-550 (synthetic TAR inhibitor, chemical component P13)
- Formula / MW
- C24H33N5O · 407.6 Da
- SMILES
- NCCCOc1ccc(cc1CNCCN2CCNCC2)c3ccc4[nH]ccc4c3
// Ground truth reference
PDB 1UTS · 2004 · Solution NMR
Murchie, Davis, Isel, Afshar, Drysdale, Bower, Potter, Starkey, Swarbrick, Mirza, Prescott, Vaglio, Aboul-ela, Karn. "Structure-based drug design targeting an inactive RNA conformation: exploiting the flexibility of HIV-1 TAR RNA." J Mol Biol 336(3) (2004). Apo reference: Aboul-ela, Karn, Varani. "Structure of HIV-1 TAR RNA in the absence of ligands reveals a novel conformation of the trinucleotide bulge." Nucleic Acids Res 24 (1996) (PDB 1ANR).
// Query sequence source
29-nt TAR construct extracted directly from 1UTS's own resolved chain (author chain B, native numbering 17-45) — identical sequence and numbering to the true apo reference structure 1ANR, so apo prediction can also be checked directly against a real unliganded experimental structure, not only against the holo one.
// Interpretation
TAR is a genuinely harder target than a simple duplex — a flexible bulge whose conformation the ligand itself selects for (RBT-550 was rationally designed against an alternative, 'flipped-out' bulge state, not the resting one). Co-folding moves the RNA fold consistently toward the true bound-state geometry (binding-site RMSD 2.87→2.38 Å, real improvement on every replicate) — uniquely for this target we also have a true experimental apo structure (1ANR) to check blind prediction against directly, where it scores 1.67 Å. The ligand-pose RMSD looks poor in isolation (7.98 Å mean) but checking what it actually contacts tells a better story: all 9 replicates land on the exact real pharmacophore residues (A22, U23, G26, C39, U40) plus the flexible bulge itself (C24, U25) that RBT-550 specifically exploits — the right site, the right chemistry, an imprecise atomic pose. This matches published independent work on this same system (Panei, Gkeka & Bonomi, Nat Commun 2024, SHAMAN) that flags TAR as harder than a rigid riboswitch pocket for exactly this reason — flexibility, not detection failure.
// Case study — when a solved crystal structure isn’t the whole story
C9orf72 (GGGGCC) repeat RNA
A 4-strand (G2C4)₂ tetramer from the C9orf72 hexanucleotide repeat expansion — the RNA target implicated in the most common genetic form of ALS and frontotemporal dementia — co-folded with ANP77, a rationally-designed bivalent ligand. The first result looked like a clean failure: across 60 independent co-folding replicates, the predicted ligand pose landed 3.6–22 Å from its position in the single published crystal structure (8QMH), and neither 60× resampling nor downstream MD refinement moved it closer. That resistance to correction is what made it worth checking properly rather than writing off.
Checked against the source paper’s own solution-state biophysics rather than against the crystal alone: independent ITC measurements report two real binding events on this tetramer (Kd 976 nM and 7.6 μM), and mass spec confirms both 1:1 and 1:2 ligand:RNA stoichiometry — the paper’s own authors attribute this to structural heterogeneity in solution. ANP77 itself is a SELEX-derived bivalent binder for a specific recognition motif (two neighboring cytosines next to a guanine), not a generic RNA binder. Checked at residue-and-base resolution, all 60 co-folded replicates contact that exact motif — most extend it to a second, chemically identical copy of the same motif on the tetramer’s symmetric neighboring strand. The model isn’t missing the real chemistry; it’s finding more of it than the one crystal structure shows.
// 1. Co-folded at the crystallographic site
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
RNA backbone at the binding site: median 0.25 Å (min 0.22, max 1.73, n=9) — the local fold is recovered essentially exactly.
Ligand pose vs. the crystal position: mean 14.62 Å (min 3.62, max 22.23, n=9) — the fold is right, the crystal’s specific pose isn’t reproduced.
// 2. Co-folded at the second, literature-predicted site
Slate: apo prediction (no ligand information). Teal: co-folded prediction (sequence + ligand SMILES). Gold: experimental holo structure. Magenta marks the binding-site residues on every structure that shows them, so alignment or mismatch is visible directly. All three are pre-aligned on the binding-site region — toggling doesn’t require re-orienting.
Representative replicate contacts the CC+G motif on chains C/D — the tetramer’s symmetric copy of the exact motif ANP77 is designed to recognize — without touching chain B at all.
Ground truth shown for chain/spatial reference only — this site has never been crystallized, so there is no experimental pose to compare against here. This is a testable prediction, motivated by independent ITC/mass-spec evidence that a second site exists, not a validated structure.
// Why the standard ranking would have discarded site 2
In this platform’s standard apo pocket-detection pass, the crystallographic site is found in 1 of 5 ensemble frames and the second site in 2 of 5 — both below the persistence floor (≥2 frames) our default ranking requires before surfacing a candidate. On a small RNA with few real candidate sites, we test an alternative: rank by how many spatially distinct candidates exist rather than by how persistent any one of them is, and surface a representative of each for cheap co-fold validation. On this target, that alternative recovers both the known site and the literature-predicted second site that the default ranking silently drops. Prototype, validated on this one target so far — not yet in production.
// Why this matters for a customer
A team with one crystal structure and a compound series that underperforms in solution has a real, expensive, common problem: the static structure may not be the only state the target populates. Co-folding candidates and checking whether engagement reproducibly bridges to a second site — backed here by independent ITC and mass-spec evidence — is a structural explanation for exactly that kind of gap, surfaced automatically rather than found after a failed campaign.
Ligand: W53 / ANP77 (bivalent CC+G-motif binder) · C23H26N8O · 446.2 Da. Ground truth: PDB 8QMH · 2024 · X-ray crystallography. Błaszczyk, Ryczek, Das, Mateja-Pluta, Bejger, Śliwiak, Nakatani, Kiliszek. "Antisense RNA C9orf72 hexanucleotide repeat associated with amyotrophic lateral sclerosis and frontotemporal dementia forms a triplex-like structure and binds small synthetic ligand." Nucleic Acids Research 52(11) (2024). PDB 8QMH (holo), 8QMI (apo). Query sequence: 4 identical 6-nt GGCCCC strands, extracted directly from 8QMH's own resolved chains A–D (numbered 1-6 in each chain's own native numbering).
This is a scoping-gate result, not a claim that ligand-induced conformational change is solved on either RNA or protein. See the platform’s core, benchmarked capability on /demos.
Get in touch