- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper presents RiboSBDD, an SE(3)-equivariant diffusion framework for RNA-conditioned 3D small-molecule generation, addressing data scarcity and RNA pocket flexibility. The framework combines protein-to-RNA geometric transfer, local nucleobase correction, and adaptive source–target sampling to improve ligand generation. The authors also introduce RiboSBDDBench, a curated dataset containing 8,006 RNA–ligand pocket instances with structured data splits and evaluation protocols. Experiments on 713 test pockets demonstrate improved contact recovery, reduced steric clashes, and better AnnapuRNA interaction scores compared with RNA-fine-tuned TargetDiff.
Critical strengths
S1. The paper addresses RNA-targeted structure-based drug design, an important yet relatively underexplored area. Tackling data scarcity and RNA pocket flexibility is meaningful for extending molecular generation methods beyond protein targets.
S2. The framework combines protein-to-RNA knowledge transfer with local nucleobase correction, providing a reasonable approach to leveraging protein-derived geometric priors while accounting for RNA-specific structural characteristics.
S3. RiboSBDDBench provides 8,006 curated RNA–ligand instances with grouped structural splits and multiple evaluation metrics. This dataset could serve as a useful resource for developing and evaluating future RNA-targeted molecular generation methods.
S4. The experiments demonstrate improvements in RNA–ligand spatial compatibility, particularly in contact recovery, steric clash reduction, and AnnapuRNA scores. Comparisons across four baseline architectures and the module ablations provide useful evidence supporting the proposed framework.
Critical weaknesses
W1. Local Base Correction is trained using synthetic nucleobase rotations rather than observed ligand-induced conformational changes. Although it improves interaction metrics, direct evidence that the model learns biologically meaningful RNA flexibility is still lacking. The improvements may also partly arise from additional geometric supervision.
W2. The current ablation removes coordinate composition and particle resampling simultaneously, making their individual contributions unclear. Comparisons with simpler strategies, such as fixed-weight averaging or adaptive mixing without resampling, would better demonstrate the necessity of the proposed approach.
W3. While RiboSBDDBench provides a useful dataset, further analysis of RNA target diversity, ligand interaction quality, and neighboring protein contributions would strengthen its reliability. Evaluation on an independent dataset or downstream task would also help demonstrate its practical utility.
W4. The specific methodological improvements of RiboSBDD over TargetDiff and their individual contributions could be more clearly demonstrated. While RiboSBDD achieves better spatial compatibility, it shows reduced synthetic accessibility and molecular diversity. More controlled comparisons and detailed analyses would help clarify the advantages of RiboSBDD over TargetDiff, the sources of its performance gains, and the trade-offs involved.
Questions for authors
Q1. Could the authors provide more direct evidence that Local Base Correction captures meaningful RNA conformational changes, such as angular prediction errors or comparisons with apo–holo structures? Could they also distinguish its contribution from that of the additional geometric losses?
Q2. How does the proposed adaptive sampling method compare with simpler alternatives, such as fixed-weight coordinate averaging? Could the contributions of coordinate composition and particle resampling be evaluated separately?
Q3. Could the authors provide further analyses of RNA target diversity, ligand interaction quality, and neighboring protein effects in RiboSBDDBench? Would evaluation on an independent dataset or downstream task be feasible?
Q4. Could the authors clarify which specific methodological improvements distinguish RiboSBDD from TargetDiff and provide stronger evidence of their individual contributions? Additionally, how do these modifications explain the improvements in spatial compatibility alongside the reductions in synthetic accessibility and molecular diversity?
6 comments
Response to Reviewer oPU9 (2/4)
Q2: Coordinate Composition vs. Particle Resampling in Adaptive Sampling
Q2: How does the proposed adaptive sampling method compare with simpler alternatives, such as fixed-weight coordinate averaging? Could the contributions of coordinate composition and particle resampling be evaluated separately?
Takeaway: Holding our flexible RNA checkpoint, , frozen protein source, and 125-step schedule fixed across 713 pockets, fixed source mixing, equal-density projection , and particle resampling each deliver distinct, additive gains in Table oPU9 Q2 01.
Table oPU9 Q2 01: Separate evaluation of fixed mixing, adaptive , and resampling across 713 pockets from Table 1 and Table 18.
| Setting | Valid% ↑ | QED ↑ | SA ↑ | BA JS ↓ | Contact ↑ | Clash% ↓ | IMP% ↑ | Div. ↑ |
|---|---|---|---|---|---|---|---|---|
| Target only: w/o Sampling | 89.59 | 0.30 | 0.59 | 0.498 | 0.78 | 9.20 | 60.31 | 0.67 |
| Fixed mixing , no resampling | 91.42 | 0.33 | 0.61 | 0.489 | 0.76 | 6.24 | 65.38 | 0.71 |
| Adaptive mixing , no resampling | 91.79 | 0.34 | 0.61 | 0.486 | 0.76 | 5.51 | 67.74 | 0.71 |
| Fixed mixing + resampling | 92.04 | 0.33 | 0.61 | 0.486 | 0.76 | 5.46 | 68.18 | 0.71 |
| RiboSBDD Full: + resampling | 92.22 | 0.34 | 0.62 | 0.483 | 0.76 | 5.00 | 70.12 | 0.70 |
- Stepwise mechanism: Static mixing restores protein-pretrained steric avoidance by cutting Clash% from 9.20% to 6.24%; equal-density projection from Skreta et al. [2] in Eq. 10 cancels shared noise along on 94.3% of steps down to 5.51%; and Feynman-Kac resampling from Skreta et al. [3] in Eq. 11 with triggers when on 18.4% of Phase I steps, reaching 5.00% Clash% and 70.12% IMP% in 14.2 s vs. 9.46% and 56.42% for 315-NFE TD P→R at 14.3 s in Table oPU9 Q4 01.
↳ Manuscript updates: Secs. 3.3, 4.3, Table 1, App. B.6, D.2 (Table 18); evaluated fixed mixing, adaptive , and particle resampling separately.
Response to Reviewer oPU9 (3/4)
Q3: RNA Diversity, Neighboring Proteins, External Pairs, and Lead Selection
Q3: Could the authors provide further analyses of RNA target diversity, ligand interaction quality, and neighboring protein effects in RiboSBDDBench? Would evaluation on an independent dataset or downstream task be feasible?
Takeaway: RiboSBDD leads all evaluated baselines on Valid%, Contact, Clash%, and IMP% across four separately evaluated OOD subsets, all five RNA classes, full-assembly and 250-pocket protein-free checks, 14 external SMARTFlexDB [11] pairs, and post-hoc lead selection; prospective wet-lab binding assays remain future work.
- OOD strata & neighboring proteins: In Table oPU9 Q3 01, across separately evaluated OOD subsets covering vs. the RNA+protein union, RNA-unseen scaffolds, RNA-unseen Rfam/Infernal [7, 8] families, and US-align , plus the 250 protein-free pockets and all five RNA classes in Table 24B where RiboSBDD reaches 61.2%–71.8% IMP% vs. 57.7%–65.7% for TD P and 47.5%–56.1% for TD P→R, RiboSBDD leads Valid% at 91.5%–93.1% and Contact at 0.75–0.77 while reaching 0.41–0.43 QED and 0.66–0.67 SA across the OOD and protein-free subsets. Furthermore, 521/713 pockets lack protein atoms within 5 Å in Table 9, and RiboSBDD achieves the lowest Clash% across all 15 configurations on
pocket.pdbat 5.00%, full RNA at 5.07%, and RNA+Protein at 9.09% vs. 9.50% for TD P and 14.29% for TD P→R in Table 21.
Table oPU9 Q3 01: Clash% ↓ and IMP% ↑ across protein-free and OOD subsets, where DiffSBDD denotes P→R; full metrics in Tables 22–24.
| Subset | DiffSBDD Clash↓ | DiffSBDD IMP↑ | TD P Clash↓ | TD P IMP↑ | TD P→R Clash↓ | TD P→R IMP↑ | RiboSBDD Clash↓ | RiboSBDD IMP↑ |
|---|---|---|---|---|---|---|---|---|
| Protein-Free, | 58.14% | 46.58% | 5.63% | 61.18% | 9.64% | 52.18% | 4.86% | 68.43% |
| Tanimoto , | 62.58% | 39.18% | 7.21% | 57.18% | 12.78% | 46.48% | 6.14% | 64.18% |
| RNA-Unseen Scaff., | 62.72% | 39.58% | 6.94% | 54.79% | 12.94% | 45.62% | 6.28% | 63.84% |
| RNA-Unseen Rfam, | 59.68% | 46.98% | 6.58% | 64.32% | 11.38% | 55.14% | 5.68% | 72.84% |
| , | 60.48% | 41.24% | 6.74% | 58.12% | 11.86% | 48.64% | 5.84% | 65.42% |
- External validation & lead ranking: Beyond the 14 SMARTFlexDB [11] pairs in Tables oPU9 Q1 02a–02b, Eq. 30 [4, 5] batch ranking under default <5% clash in Table oPU9 Q3 02a improves unselected Contact to 0.80 and independent LigandRNA [10] IMP% to 66.8% unranked and 82.6% Top-1 across all 589/713 parser-compatible pockets, including 79.7% on unseen Rfam.
Table oPU9 Q3 02a: Benchmark-wide selection under <5% clash across 713 pockets, 125 PDBs, and 589-pocket LigandRNA from Table 17A.
| Method & Tier | Valid%↑ | Pock. QED↑ | PDB QED↑ | SA↑ | Cont.↑ | Clash%↓ | Pock. IMP%↑ | PDB IMP%↑ | LigandRNA↑ |
|---|---|---|---|---|---|---|---|---|---|
| DiffSBDD P→R Top-1 | 44.50% | 0.49 | 0.51 | 0.78 | 0.72 | 9.82% | 65.74% | 64.18% | 58.9% |
| TD P Top-1 | 88.58% | 0.53 | 0.55 | 0.75 | 0.74 | 1.65% | 80.15% | 78.64% | 73.8% |
| TD P→R Top-1 | 88.74% | 0.52 | 0.54 | 0.76 | 0.76 | 2.76% | 72.18% | 70.62% | 67.5% |
| RiboSBDD Top-3 | 92.22% | 0.52 | 0.54 | 0.73 | 0.79 | 1.43% | 85.16% | 83.42% | --- |
| RiboSBDD Top-1 | 92.22% | 0.56 | 0.58 | 0.77 | 0.80 | 1.18% | 88.64% | 87.12% | 82.6% |
- Riboswitch clash gates: On 6QN3 in Fig. 17c and Table oPU9 Q3 02b, we explicitly distinguish default <5%-clash Top-1 #13 at composite ranking score 0.610 from Top-1 under the strict 0%-clash filter #03 at score 0.559; on 3NPQ and 6FZ0, default <5% Top-1 already achieves 0.00% Clash%, so both gates select the identical pose.
Table oPU9 Q3 02b: Unseen-Rfam riboswitch Top-1 leads under default <5% versus strict 0% clash gates from Table 17B.
| Target & Rfam | Gate | Model & Pose | Valid%↑ | QED↑ | SA↑ | Clash%↓ | Score ↓ | Crystal |
|---|---|---|---|---|---|---|---|---|
| 6QN3, RF01704 | Default <5% | TD P→R | 88.0% | 0.442 | 0.70 | 0.00% | −107.57 | −165.48 |
| 6QN3, RF01704 | Default <5% | RiboSBDD #13 | 94.0% | 0.331 | 0.58 | 3.80% | −241.32 | −165.48 |
| 6QN3, RF01704 | Strict 0% | TD P→R | 88.0% | 0.442 | 0.70 | 0.00% | −107.57 | −165.48 |
| 6QN3, RF01704 | Strict 0% | RiboSBDD #03 | 94.0% | 0.475 | 0.72 | 0.00% | −180.15 | −165.48 |
| 3NPQ, RF01057 | <5% & 0% | TD P→R | 88.0% | 0.472 | 0.78 | 0.00% | −142.34 | −218.62 |
| 3NPQ, RF01057 | <5% & 0% | RiboSBDD | 92.0% | 0.528 | 0.80 | 0.00% | −234.86 | −218.62 |
| 6FZ0, RF01826 | <5% & 0% | TD P→R | 90.0% | 0.465 | 0.78 | 0.00% | −162.58 | −289.44 |
| 6FZ0, RF01826 | <5% & 0% | RiboSBDD | 94.0% | 0.514 | 0.79 | 0.00% | −329.88 | −289.44 |
↳ Manuscript updates: Secs. 4.2, 5, Tables 2a–2b, App. C.1–C.3 (Table 17), D.5 (Tables 21–24); added RNA diversity/OOD analyses, protein checks, and <5%/0% riboswitch ranking.
Response to Reviewer oPU9 (1/4)
We thank Reviewer oPU9 for the detailed review and for recognizing RiboSBDD's motivation, cross-biomolecule transfer design, and benchmark curation. Below we address Q1–Q4 on conformational Δχ validation, separate sampling ablations, OOD and lead-selection studies, and backbone transfer and atom-count priors.
Q1: Direct Conformational Evidence for Local Base Correction vs. Geometry Losses
Takeaway: Holding 81k updates, pψ, the frozen protein source, and the contact/steric loss schedule fixed across 713 pockets, Local Base Correction Δχ delivers over half of the IMP% gain and nearly half of the Clash% reduction, and directly recovers glycosidic torsion shifts on synthetic perturbations and 14 split-disjoint SMARTFlexDB [11] Apo-to-Holo pairs.
Table oPU9 Q1 01: Separating geometry losses from local base flexibility across 713 pockets, matching Table 18.
Table oPU9 Q1 02a: Direct Δχ recovery on synthetic perturbations across N=713 pockets and SMARTFlexDB Apo–Holo across N=14 pairs and 58 bases, matching Table 19A,C.
Table oPU9 Q1 02b: Apo-conditioned generation on Capo scored in the Holo frame across N=14 pairs from Table 19C.