- Rating
- 3
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
Given a sequence, an experimental starting structure and optionally a fixed protein context, this paper generates an RNA conformational ensemble with an SE(3) frame model trained by flow matching to reconstruct experimental structures from corrupted inputs; no paired conformations or MD trajectories are needed. CrossIPA lets RNA nucleotides attend jointly to RNA and protein residues, re-evaluating the RNA-protein geometry at each update, and at inference independently corrupted copies of the input are transported by the same model. REFRACT reports matched TM-scores of 0.669 on protein-bound recovery and 0.648 on binding-condition transition, closer MD agreement than RNAbpFlow and DynaRNA, and an ablation that raises AMR-R from 0.643 to 0.669 while lowering clashing structures from 32.7% to 14.1%.
Critical strengths
-
The paper conditions RNA ensemble generation on bound protein, which the evaluated ensemble generators do not do. The authors state that "neither RNAbpFlow nor DynaRNA explicitly conditions exploration on bound protein structures" (Section 1), and CrossIPA recomputes the RNA-protein geometry at each update instead of treating the protein as a fixed feature. Its consequence for my assessment is that the paper addresses a setting the two closest methods do not cover, so the results are not a re-measurement of an existing benchmark.
-
The evaluation is complete. Three complementary benchmarks come with an input-structure control, a local-perturbation control and cluster-level bootstrap intervals, and Table 6 reports all 20 RMSF profiles including the failures (2VPL free, MAE 4.61 A). The paper reports both improvements and failures, which lets me judge where the method works and where it does not.
Critical weaknesses
-
The evidence does not isolate the proposed mechanism. The protein ablation removes context only at inference, using the same trained model, and it moves AMR-R by 0.026 (0.643 to 0.669). The same model without protein at inference still exceeds the strongest baseline by 0.258 (0.643 vs 0.385). The paper does not retrain a model without protein or without CrossIPA. It therefore shows a small benefit from inference-time context; it does not show that the architecture explains the gap to baselines.
-
The setup supplies a correct experimental starting structure and a correctly placed protein. Section 2 states that the protein "is supplied with a known position and orientation relative to the starting RNA", and transport starts from 20-50% corruption of that structure (training uses t/T <= 0.5). The paper does not test three settings: no starting structure (de novo), a predicted rather than experimental start, and a protein-side holdout. The demonstrated capability is sampling around a known state.
Questions for authors
-
At inference, sample 100 structures from the protein-free variant, discard those that clash with the protein, and report AMR-R and COV-R at the remaining budget. Does CrossIPA add more than this clash filter?
-
Retrain without protein context and without CrossIPA, and report Tables 1-3. The current ablation changes only inference, so it cannot separate training-time protein exposure from inference-time conditioning.
-
Did every method use the same candidate budget? AMR-R and COV-R take a maximum over candidates and increase with N, and Section 3.1 specifies N = 100 only for REFRACT. Please state N for AF3 subMSA, RhoFold+ subMSA, RNAbpFlow and DynaRNA.
-
What base-pair input did RNAbpFlow receive, experimental or predicted? And why are AF3 and RhoFold+ absent from the MD comparison (Table 3) when they serve as ensemble baselines in Tables 1-2?
-
Replace the experimental starting structure with an AF3 or RhoFold+ prediction and report the change. Without this, the method cannot be used where no experimental structure exists.
-
For RBPTrans, is the target protein placed using the experimental coordinates of the target complex? If so, the benchmark supplies the relative RNA-protein placement that an application would need to determine, and the setup should state this.
2 comments
Response to Q4–Q6
Response to Q4 (RNAbpFlow base-pair inputs and adding AF3/RhoFold+ to the MD benchmark).
- RNAbpFlow inputs: It received predicted (not experimental) secondary structures—specifically, the ensemble of three base-pair maps predicted by RibonanzaNet, SPOT-RNA, and IPknot, matching its original publication protocol.
- AF3 and RhoFold+ on MD: They were initially omitted from Table 3 because sequence-only predictors cannot distinguish between the free and protein-bound simulation conditions of the same RNA. We have now added both to the MD benchmark:
Table R3-4: MD benchmark including AF3 subMSA and RhoFold+ subMSA. (RhoFold+ produces structure on only part of the conditions and is evaluated on that subset.)
| Evaluation Subset | Method | RMSF error (Å) ↓ | RMSF ↑ | Ens. RMSD (Å) ↓ | (Å) ↓ |
|---|---|---|---|---|---|
| All conditions | REFRACT | ||||
| All conditions | AF3 subMSA | ||||
| All conditions | RNAbpFlow | ||||
| All conditions | DynaRNA | ||||
| Conditions w/ RhoFold+ ensemble | REFRACT | ||||
| Conditions w/ RhoFold+ ensemble | AF3 subMSA | ||||
| Conditions w/ RhoFold+ ensemble | RhoFold+ subMSA |
AF3 subMSA is the strongest sequence-only baseline on MD, but REFRACT outperforms both AF3 and RhoFold+ across all four metrics on both subsets.
Response to Q5 & W2 (Initializing REFRACT from AF3-predicted structures and scope of the method).
We replaced the experimental starting structure with the top-ranked AF3 prediction (AF3 top-1, full MSA), superposed onto with 's native protein context:
Table R3-5: Ensemble recovery (AMR-R ) initialized from predicted vs. experimental structures.
| Initialization / Method | RBP | RBPTrans | RNA-only |
|---|---|---|---|
| AF3 | |||
| REFRACT initialized from AF3 | |||
| REFRACT initialized from experimental (paper) |
REFRACT consistently improves upon its starting AF3 prediction across all three benchmarks ( on RBP; on RBPTrans; on RNA-only), showing that flow transport explores useful conformational variations even around a predicted start. At the same time, because single AF3 predictions on these post-cutoff complexes often have inaccurate global folds (mean to experimental on RBP), partial noising () is not designed to correct a globally misfolded topology. Accordingly, REFRACT is positioned as a structure-conditioned ensemble generator (like Str2Str and DynaRNA) that samples conformational landscapes around an experimental structure or an accurate structural model. We will include Table R3-5 and state this scope clearly in the paper.
Response to Q6 & W2 (RBPTrans protein registration and target-free placement controls).
In the paper's RBPTrans setup, the source RNA was superposed onto the target RNA to place the target protein in its experimental pose (which we will state explicitly in Section 3.1). To test whether our results depend on this target-complex registration, we reran inference using only the source complex ( paired with its own native protein in its source coordinates, carrying zero information from the target complex), as well as under rigid-body protein perturbations:
Table R3-6: Target-free protein placement controls.
| Protein Input Setting | RBP AMR-R ↑ | RBP Clash ↓ | RBPTrans AMR-R ↑ | RBPTrans Clash ↓ |
|---|---|---|---|---|
| Target complex (paper) | ||||
| Source complex only (zero target info) | ||||
| No protein () |
Using only the known source complex leaves recovery unchanged ( and ) and does not increase clash rates ( and ) (and rotating the protein by likewise preserves recovery while increasing clashes to on RBP). This confirms that the reported gains do not depend on oracle target-complex registration. In practice, protein placement can come directly from the starting complex, or from rigid-body docking / co-folding when modeling a new partner.
Overview & Response to Q1–Q3
We thank you for your constructive review. We ran all requested experiments (Q1–Q6), which address your two weaknesses:
On RBP (41 clusters) we sampled N=100 structures per cluster from the protein-free variants, discarded those clashing with the protein (any C4′ within 1 Å of a protein heavy atom), and scored the rest:
Table R3-1: Post-hoc clash filtering vs. CrossIPA on RBP.
At the same generation budget (100 trajectories), CrossIPA yields 1.17–1.27× as many sterically valid conformations (86.2 vs. 68.1 and 73.4) and higher ensemble recovery (0.668 vs. 0.621 and 0.663): steering trajectories away from occluded space during transport avoids wasting samples on protein clashes.
We retrained two ablated models from scratch with identical hyperparameters: (1) without protein, which never sees protein in training or inference; (2) without CrossIPA, where the protein enters only as a static per-nucleotide RBF distance feature (not updated across blocks via geometric cross-attention). We evaluate on RBP, on the RBPTrans transfer task (34 source→target pairs, the protocol of Fig. 4) and on MD:
Table R3-2: Retrained ablations. Transfer: AMR-R↑ / [email protected]↑ / [email protected]↑. MD: Ens. RMSD↓ / W2↓ (Å).
We will add Table R3-2 to the paper and state the two contributions separately.
REFRACT, AF3 subMSA, RNAbpFlow and DynaRNA all generate N=100 structures per cluster. RhoFold+ is deterministic, so its diversity comes only from MSA subsampling; with a singleton MSA (46/68 clusters in the default setup) it yields N=1 (mean N=22.7 on RBP). Table R3-3 compares N=1 (mean of 20 random draws) with N=100:
Table R3-3: AMR-R↑ at N=1 vs. N=100.
Even with a single candidate, REFRACT (0.547 on RBP, 0.717 on RNA-only) exceeds every baseline at N=100. Giving RhoFold+ deeper MSAs from the AF3 pipeline raises its mean N on RBP from 22.7 to 51.7 and its AMR-R from 0.349 to 0.364, leaving all conclusions unchanged. We will report N for every baseline in Section 3.1.