- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes PepRAE for peptide generation given the protein target, binding site, and peptide length, which architecture-wide is a partially latent diffusion model. Its main contribution is developing a framework that can learn feature latents for protein targets and peptides from pretrained protein models via representation reconstruction and peptide reconstruction. PepRAE achieves competitive performance in amino acid recovery, reconstruction rmsd, and interface qualities.
Critical strengths
- The problem is well-defined and the partially latent architecture is well-motivated for joint modeling of sequence and structure (similar to proteina-complexa (https://research.nvidia.com/labs/genair/proteina-complexa/) ). The framework is carefully designed to solve the mismatch between complete-context representations available during training and target-only information available at inference.
- The benchmarks are carefully set to ensure fair comparison. PepRAE results are relatively strong compared with those baselines, showing the effectiveness of the proposed framework. And ablations on different loss terms and components also show the usefulness of different modules.
Critical weaknesses
- One of the main motivations for this work is claimed to be using the learned representation from pretrained protein model (e.g. Boltz-2) for this generation task. But the ablation result in Table 4 shows that even training PepRAE from LightEnc, a lightweight sequence encoder trained from scratch, can achieve fairly good performance, which wins in L-RMSD, M-RMSD, DockQ and ∆G, compared with other much stronger pretrained models. So this raises a question of how important it actually is to use those much stronger representations.
- The overall method description and framing lead me to view this work primarily as a peptide design model. In the benchmark, the authors show that PepRAE achieves strong native-recovery performance, including the best amino-acid recovery and lowest structural RMSDs relative to the reference peptides. However, the evaluation does not assess whether the generated peptide sequences are self-consistent with the generated structures, for example through independent refolding with an external structure predictor. Such a self-consistency check is a common and important validation in protein design, because native-recovery metrics alone do not establish that the designed sequences intrinsically support the generated conformations.
Questions for authors
Clarification:
- Can the author clarify whether the purpose of this work is for de novo design of protein-binding peptides, or is it just for native recovery? If it's the former, have the authors done any self-consistency evaluation to see whether the generated peptides are designable? An example reference is https://www.nature.com/articles/s42004-023-01029-7
Suggestion:
- It would be useful to provide a more detailed analysis of which specific information encoded in pretrained protein representations contributes to performance on this task, especially given the competitive results of the lightweight LightEnc baseline.
- It might be worth checking how similar the test peptide sequences are to peptides in the training set to exclude some possibilities for peptide-level leakage and memorization, which would also give a false-positive recovery.
10 comments
Response to Reviewer adDD — W2/Q1: Design task and consistency analysis
W2/Q1. Design task and sequence–structure consistency.
PepRAE is a de novo target-conditioned peptide sequence–structure generator. Given a target structure, binding site, and prescribed peptide length, it generates peptide feature and spatial states from Gaussian noise and decodes them into a sequence and bound conformation (Sections 3.1 and 3.4). The evaluated setting uses bound target coordinates, but does not supply the native peptide sequence or coordinates as peptide inputs. Following the peptide-design benchmark protocol in Section 4.1 and Appendix D, Tables 1 and 10 report native recovery alongside Rosetta interface scores, clashes, dihedral-distribution distances, and diversity.
Following your suggestion, we evaluated whether sequence-based complex prediction reproduces each candidate's own generated conformation, using Boltz-2. The protocol and results are below.
Candidate selection. For each of the 92 targets, we selected the candidate with the lowest Rosetta ΔG from each model's original 100-candidate pool. ΔG is the mean over three relaxations per candidate (Appendix D). Candidate identities were fixed before structure prediction, without using native-recovery scores or prediction results. PepRAE uses the same three diffusion runs as Table 1, evaluated separately; baselines use the models and candidate pools evaluated in the submission.
Complex prediction. We used Boltz-2 v2.2.1 with receptor MSAs and single-sequence peptide inputs, keeping the receptor construct identical across methods for each target. Boltz-2 is also the representation source for the main PepRAE configuration; in this experiment, it receives sequences and MSAs, without generated peptide coordinates, structural templates, or contact/pocket constraints. Each input received five predictions with seed 12, three recycling steps, and 200 sampling steps. We selected the prediction with the highest official confidence_score, rather than by agreement with the generated structure, and applied no relaxation to the predictions.
Comparison and metrics. Each selected prediction is compared with that candidate's original, unrelaxed generated structure. Native-sequence controls use the same prediction settings and are compared with their experimental structures. Both RMSDs measure peptide Cα coordinates: pocket fit aligns a fixed set of receptor-pocket Cα atoms without subsequently fitting the peptide; peptide fit aligns the peptide itself. The pocket residue set is fixed before prediction and used only for evaluation. Thus, pocket fit assesses agreement in the receptor frame, while peptide fit assesses the peptide's internal conformation.
Table R8(a). Boltz-2 geometric agreement.
| Method | Peptide RMSD: pocket fit (Å) ↓ | Peptide RMSD: peptide fit (Å) ↓ | Pocket-fit RMSD ≤2 Å (%) ↑ |
|---|---|---|---|
| PepGLAD | 9.172 | 3.795 | 1.09 |
| PepFlow | 10.504 | 2.469 | 3.26 |
| PepTri | 9.113 | 2.724 | 2.17 |
| UniMoMo | 9.380 | 2.361 | 7.61 |
| PRIME | 9.031 | 2.926 | 7.61 |
| PepRAE | 7.328 ± 0.064 | 2.613 ± 0.060 | 8.70 ± 1.09 |
| Native control | 0.812 | 0.384 | 79.35 |
Table R8(b). Confidence of the same selected predictions.
| Method | Peptide pLDDT ↑ | Pocket–peptide PAE (Å) ↓ |
|---|---|---|
| PepGLAD | 72.19 | 4.412 |
| PepFlow | 71.48 | 4.559 |
| PepTri | 70.32 | 4.259 |
| UniMoMo | 72.47 | 4.593 |
| PRIME | 74.21 | 4.239 |
| PepRAE | 73.78 ± 1.18 | 3.918 ± 0.105 |
| Native control | 88.80 | 2.660 |
Each run covers all 92 targets. RMSDs are cross-target medians. For each target, we compute peptide-averaged pLDDT (0–100) and mean bidirectional pocket–peptide PAE, then average these values across targets. The ≤2 Å column reports the percentage of targets meeting that geometric criterion. PepRAE reports mean ± sample SD of the three separately calculated run-level statistics; percentage SDs are in percentage points. Bold marks the best reported central value among the six design methods.
PepRAE has the lowest reported pocket-fit RMSD and pocket–peptide PAE among the evaluated design methods. Its pocket-fit RMSD is 7.328 Å, compared with 9.031 Å for PRIME, the next-lowest baseline. It also has the highest reported fraction within 2 Å: 8.70%, versus 7.61% for UniMoMo and PRIME; the native-sequence control reaches 79.35%. UniMoMo has the lowest peptide-fit RMSD, and PRIME the highest peptide pLDDT.
Response to Reviewer adDD — S1: Backbone information and decoding responses
PepRAE generates both peptide sequences and bound conformations, so we examined backbone geometry. The backbone angle ψ, defined by the atoms , describes rotation around . We compared predictions using source features, residue identity alone, or a sequence window containing the central residue and two neighbors on each side. To account for amino-acid-specific angle preferences, we also predicted residuals around identity-conditioned label means. Finally, we tested decoding after changing the autoencoder's feature posterior means, with equally large random changes as controls. Table R12(a) reports both prediction analyses in separate columns; Table R12(b) reports the intervention.
Predicting backbone angles. With source models frozen, we fitted linear models on 9,113 training complexes and evaluated them on the original 1,064-complex validation split. Regularization was selected within training by receptor-cluster splits.
For source model , we use (Eq. (1)); is residue identity. Direct models predict . For residual prediction, let index fitting residues and count residues of identity . We compute each identity's mean label, subtract it, and fit a linear residual predictor :
Angles are recovered by atan2. The identity-only reference uses a fitted identity model for direct prediction and alone for residual prediction.
Table R12(a). ψ prediction error (degrees) ↓.
| Input | Direct prediction | Identity reference + residual |
|---|---|---|
| Residue identity only | 69.77 | 68.36 |
| Sequence window (±2 residues) | 61.16 | 60.99 |
| Boltz-2 source features | 39.45 | 39.66 |
| ESM-2 source features | 50.41 | 50.51 |
| LightEnc source features | 55.23 | 55.26 |
All sources have lower errors than both references; Boltz-2 has the lowest reported error in both analyses.
Changing feature means before decoding. We perturb the autoencoder's eight-dimensional peptide feature posterior means entering . Following Appendix I.2, is the amplitude and contains the training standard deviations of these means, one per channel. A linear classifier fitted on training data predicts helix, strand, or coil from standardized means on active channels, with training mean . Here denote elementwise division/multiplication. With active channels out of , its helix/coil weights define
We use . Four random perturbations satisfy in original latent coordinates. Target feature rows, spatial inputs, and autoencoder parameters stay fixed. Eq. (4) gives
We reassign secondary structure from decoded coordinates. Let be the helix fraction under perturbation , with from unperturbed decoding. In percentage points, ; the excess response is
Table R12(b). Boltz-2 source: helix-fraction change (percentage points).
| Perturbation | Change |
|---|---|
| Helix–coil direction | +2.73 |
| Mean of four norm-matched random directions | +0.80 |
| Excess response | +1.93 |
We evaluate 63 complexes with complete Cα coordinates from a fixed 64-complex validation subset. After averaging random responses within complexes, we bootstrap the paired contrast over 56 PDB groups (1,000 draws). The pointwise 95% percentile-bootstrap interval is 0.41–3.74 percentage points. Decoded residue identities change at 71.1% of positions relative to unperturbed decoding, so this is a joint sequence–structure response.
The helix–coil response exceeds the average random response.
Response to Reviewer adDD — W1: Representation sources and adaptation objectives
Thank you for recognizing the task formulation, the treatment of the training–inference mismatch, and the supporting ablations. We first clarify the role of the representation sources, then report the additional structure-prediction and peptide-overlap analyses in the accompanying comments.
W1. Both pretrained sources and LightEnc support competitive generation.
Table 4 directly examines how generation performance depends on the representation source. It compares separately trained Boltz-2, ESM-2, and LightEnc variants with matched feature-latent widths, downstream architecture, objectives, and optimization settings (Section 4.5). Different sources lead on different metrics: ESM-2 has the highest reported AAR (40.41%), Boltz-2 the lowest C-RMSD (2.432 Å), and LightEnc the lowest L-RMSD and M-RMSD (1.187 and 1.408 Å), highest DockQ (0.746), and lowest Rosetta ΔG (−30.269 REU).
These results show that a large pretrained source is not required for competitive generation in the evaluated setting. The paper's claim is that the same adaptation formulation can learn feature latents for generation from the tested representation sources. LightEnc is itself trained: Appendix C specifies its masked-language-model training from random initialization on the training split, followed by freezing before PepRAE training.
The role of the adaptation objectives is tested separately in Table 2, with the Boltz-2 source and architecture held fixed. Removing representation reconstruction, No Lrep, changes C-RMSD from 2.432 to 2.612 Å, L-RMSD from 1.213 to 1.343 Å, and Rosetta ΔG from −30.161 to −27.946 REU. Residue reconstruction accuracy remains almost unchanged, at 99.89% and 99.81%, respectively (Table 13). Near-identical reconstruction accuracy therefore coexists with different reported generation results.
Restricting representation-decoder updates to representation reconstruction also yields less favorable generation means (Table 2). This control blocks peptide-loss updates to the representation-decoder parameters, while retaining those gradients through the decoder to the feature latents and encoder (Appendix G.1). Each configuration retrains its autoencoder and then trains three diffusion models with its selected autoencoder fixed.
Table 3 separately tests target-adapter supervision with the autoencoder fixed. PepRAE trains the adapter by latent alignment and stops denoising gradients at its output. No supervision removes alignment while retaining this gradient stop; Denoising-supervised removes alignment and allows denoising gradients to update the adapter (Section 4.4.2; Appendix G.2). C-RMSD is 2.432 Å for PepRAE, 2.767 Å for No supervision, and 2.780 Å for Denoising-supervised. The complete configuration has the best reported three-run mean on all seven generation metrics in each ablation table. These are means of run-level benchmark statistics; within each run, RMSDs are medians of per-target minima (Appendix D.1).
Tables 2–3 support the evaluated adaptation and conditioning objectives; Table 4 establishes applicability across the tested sources. LightEnc's competitive performance is consistent with the stated cross-source finding.