{"id":"2110f1b2-66db-4912-b4a4-2796e95b00e7","arxiv_id":"2506.12645","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":2.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A synthesis paper restating the iterative annealing framework for chaperone action, including a model-derived prediction that moderate RNA chaperone activity maximizes in vivo splicing yield.","lead":"This paper restates a unified theory, called iterative annealing, for how protein and RNA chaperones rescue misfolded molecules. It argues that both GroEL and CYT-19 repeatedly unfold stuck states, and predicts that RNA splicing is best at moderate chaperone activity.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The in vivo optimum prediction (Eq. 9) assumes the disruption ratio κ is concentration-independent; if κ varies with [CYT-19] or [ATP], the predicted moderate-activity maximum can shift or disappear.","rationale":"The quantitative heart of the paper is the non-monotonic RNA-chaperone prediction, and that prediction is used to support the broader evolutionary-optimality claim. The reader identified the same weak spot as the weakest assumption. I do not see an internal inconsistency in the discrete-map results of Eqs. (4)–(6): those expressions are algebraically consistent and the framework has prior experimental support. The unresolved issue is whether the maximum at Eq. (9) is a real feature of the in vivo control variable. Because the paper treats κ as a fixed parameter while explicitly writing it as a function of [C] and [T], the moderate-activity optimum could be an artifact of plotting yield against k_MI^eff at fixed κ. A numerical scan with concentration-dependent unfolding rates would settle this: if the optimum survives for realistic κ([C]), the RNA prediction stands; if not, it should be presented as a schematic prediction rather than a quantitative in vivo claim. This does not change the reader's CONDITIONAL verdict, since the paper should be accepted only with the requested softening and reframing.","tokens_in":14400,"tokens_out":11266,"duration_ms":134901,"concrete_test":"Compute P_ss^SP from the Fig. 4B master equation with concentration-dependent unfolding rates, e.g. k_MI^eff = v_M[C]/(K_M + [C]) and k_NI^eff = v_N[C]/(K_N + [C]), using the parameter set from ref [20] and k_d values from the same model. Numerically locate the optimum as a function of [C] over a physiological range, scanning v_N/v_M and K_N/K_M over ratios 0.1–1. If the optimal [C] differs by more than a factor of 2 from the value implied by Eq. 9, or if no interior optimum exists, the moderate-activity prediction is not robust to realistic concentration dependence of κ.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that κ = k_NU([C],[T])/k_MU([C],[T]) in Eq. (8) is a constant (0.13) when the yield P_ss^SP is optimized over k_MI^eff. The paper itself notes that both numerator and denominator depend on [C] and [T]. In the kinetic network of Fig. 4B, with k_d small, the yield is maximized by minimizing f(x) = (κ x + k_s)(1 + k_IM/x), where x = k_MI^eff. If κ is constant, f'(x)=0 gives Eq. (9). If κ = κ(x), the extremum condition becomes κ'(x)(x + k_IM) + κ(x) - k_s k_IM/x^2 = 0, so Eq. (9) is not the correct condition. Unless the two chaperone-induced unfolding rates share the same concentration dependence (for example, both strictly linear in [C] with no saturation and no ATP dependence), the location of the optimum, and possibly its existence, changes. The manuscript provides no data or argument fixing κ across the concentration range used to define 'moderate activity.' Thus the headline RNA prediction is not yet established as a prediction about CYT-19 concentration in vivo.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes the Iterative Annealing Mechanism (IAM) as a unified framework for GroEL-mediated protein folding and CYT-19-mediated RNA folding. It derives yield expressions for repeated chaperone cycles (Eqs. 4-6), introduces a parameter κ = k_NU/k_MU to distinguish the two machines, and extends the model to in vivo self-splicing of group I intron pre-RNA, predicting that the steady-state yield of spliced RNA is maximized at a moderate chaperone activity (Eq. 9). The authors further argue that both machines have evolved to maximize native-state yield on biological timescales and that the theory quantitatively explains all available experiments.","tokens_in":14779,"tokens_out":6759,"duration_ms":75305,"significance":"If the central claims held, the paper would offer a useful unifying perspective: the same geometric-series annealing argument explains why GroEL-assisted folding yield rises monotonically with added chaperone while CYT-19-assisted yield can be non-monotonic, and it ties this difference to a single dimensionless factor κ. The elementary derivations in Eqs. (4)-(6) are transparent and correct under the stated kinetic assumptions, and the explicit optimum condition in Eq. (9) is a falsifiable prediction in principle. The GroEL case with κ = 0 is effectively a parameter-free consequence of the measured partition factor Φ, which is a definite strength. However, the strong evolutionary and quantitative claims in the abstract and discussion exceed what is demonstrated here, and the in vivo optimum prediction rests on a concentration-independence assumption that is not examined.","major_comments":[{"comment":"Equations (8) and (9) are in tension. Equation (8) defines κ = k_NU([C],[T])/k_MU([C],[T]) with both rates explicitly dependent on chaperone and ATP concentrations, yet Eq. (9) is obtained by optimizing the yield with respect to k_MI^eff while holding κ fixed at 0.13 (Fig. 4C). If κ varies with [C] or [T] over the range where k_MI^eff changes, the extremum condition is not Eq. (9). With x = k_MI^eff and κ = κ(x), the stationary condition becomes κ'(x)(x + k_IM) + κ(x) = k_IM k_s / x^2, so the optimum location, and possibly its existence, changes unless the chaperone-induced unfolding rates share the same concentration dependence. The manuscript provides no data or argument that κ is concentration-independent in vivo, so the headline prediction of a moderate-activity maximum is not yet established as a prediction about CYT-19 concentration.","section":"RNA folding in cells; Eqs. (8)-(9)"},{"comment":"The claim in the abstract and Section II that 'both these machines have evolved to maximize the production of the steady state yield on biological times' is not supported by the analysis. The theory shows that for fixed κ and Φ the yield is given by Eq. (6) and increases with the number of cycles, but it does not compare evolutionary fitness across different values of κ or consider alternative optimization targets such as production rate P_N/τ, ATP cost, or substrate specificity. Indeed, the paper itself introduces the power-efficiency tradeoff, which suggests that maximizing steady-state yield is not obviously the selection target. The evolutionary statement should be framed as an interpretation or a hypothesis, not as a derived result.","section":"Discussion: Chaperones solve an optimization problem; Concluding Remarks"},{"comment":"The paper claims that IAM 'quantitatively explains all the available experiments' and 'quantitatively accounts for all the experimental observations [17]', but this manuscript contains no direct quantitative comparison of the model with experimental data: there are no theory-versus-experiment plots, no error analysis, and no table of fitted parameters. The in vivo splicing prediction uses rate constants and κ = 0.13 taken from the earlier fit in ref [20], so the moderate-activity optimum is a property of that previously fitted model rather than an independent, parameter-free prediction. The authors should either include the quantitative comparison here or explicitly present the paper as a synthesis of prior work and soften the corresponding claims.","section":"Introduction; Concluding Remarks; Fig. 4C"}],"minor_comments":[{"comment":"The citation 'works [17 ? -20]' contains a missing or malformed reference marker and should be corrected to a standard bracket format.","section":"Introduction"},{"comment":"There are several typos and grammatical slips, including 'all also by mutations' (p. 5), 'maximzes' (p. 8), 'riboyzme' (p. 9), 'out theory is allicable' (p. 16), and 'evolved to the maximize' in the abstract; these should be corrected.","section":"Throughout"},{"comment":"The star symbol and the value k_MI^eff ≈ 7.5 min^-1 should be defined in the figure caption rather than only in the text, and the source of the plotted curves relative to ref [20] should be stated in the caption.","section":"Fig. 4C"},{"comment":"Equation (7) is typeset ambiguously, with the ratio k_NU/k_MU displayed in a way that is easy to misread; it should be rewritten explicitly as P_N^κ(∞) ≈ k_UN / (κ k_UM + k_UN) with κ defined by Eq. (8).","section":"Eq. (7)"}],"recommendation":"major_revision","confidential_remarks":"This manuscript overlaps considerably with the authors' earlier PNAS 2022 paper [20] and Protein Science 2020 article [19]. The main new elements appear to be the unified presentation and the evolutionary interpretation. The technical concern about the concentration dependence of κ is the most important issue; if it cannot be resolved with data or a sensitivity analysis, the central in vivo prediction should be substantially qualified. I do not see grounds for rejection, but the overclaims should be addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short take: this is a perspective, not a new research paper. The essential equations, including the moderate-activity optimum (Eq. 9), come verbatim from the authors' earlier work (refs 17, 19, 20). The genuinely nice part is the unified presentation: one simple geometric-series argument covers GroEL and CYT-19, and the κ parameter cleanly separates a machine that leaves the native state alone (κ≈0) from one that attacks it (κ≈0.13). The power-efficiency tradeoff analogy is conceptually helpful, and the manuscript would serve as a readable entry point for graduate students.\n\nThat said, the packaging oversells. The abstract and title treat the synthesis as a fresh theory; the evolutionary phrases — 'evolved to maximize the steady state yield' — assert an optimization that the kinetics alone don't establish. Good models can explain a fitness advantage without proving evolution ran to that optimum. The paper is also loose with 'quantitatively accounts for all experiments', where the fits come from earlier papers and no error analysis is repeated here.\n\nOn the stress-test: the concern about κ being constant is real. Eq. (9) is derived by maximizing over k_MI^eff with κ fixed, but what you can actually vary in the cell (CYT-19 concentration, ATP) enters both k_NU and k_MU, and the manuscript gives no argument that their ratio is concentration-independent. If κ changes as you titrate the chaperone, the position of the optimum shifts. This is a caveat the authors should state explicitly; without it, the 'moderate activity' claim reads as a stronger experimental prediction than it is.\n\nBottom line: read it as a well-executed review of the authors' own framework, not as a new result. It deserves a serious referee — the authors are the right people to write this perspective, and the manuscript is clear enough to benefit from external checking — but it should be accepted only after the novelty claims and the κ assumption are addressed. I'd bring it to a reading group for the discussion of when a synthesis is a contribution, and cite the original papers, not this one.","headline":"A clear, well-written synthesis of the authors' own IAM framework, but not a new result; the RNA optimum prediction needs a caveat about κ.","tokens_in":15242,"tokens_out":4107,"would_cite":false,"duration_ms":45758,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that GroEL and CYT-19 chaperones work by one iterative annealing mechanism, and that RNA self-splicing yield is maximized only at moderate chaperone activity.","keywords":["molecular chaperones","iterative annealing mechanism","kinetic partitioning mechanism","GroEL","CYT-19","RNA folding","protein folding","self-splicing"],"falsifier":"Titrate CYT-19 concentration against the yield of self-spliced group I intron pre-RNA in an in vivo or reconstituted assay, holding the transcript and ATP pool fixed. If the yield rises monotonically to a plateau instead of peaking at an intermediate concentration, or if the maximum moves sharply with ATP concentration in a way inconsistent with a fixed $\\kappa$, the predicted optimum $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM}k_s/\\kappa}$ fails.","tokens_in":14220,"feed_emoji":"🧬","tokens_out":9920,"duration_ms":107490,"temperature":0.7,"pith_summary":"Molecular chaperones such as GroEL for proteins and CYT-19 for RNA are usually studied as separate machines, but this paper argues that both operate by the same iterative annealing mechanism: they burn ATP to repeatedly unfold misfolded molecules, giving the population multiple fresh chances to reach the native state instead of passively stabilizing it. The paper claims this single framework quantitatively explains the measured outcomes of experiments on both machines, and that both chaperones have evolved to maximize the steady-state yield of functional products on biological timescales rather than to approach the equilibrium Boltzmann distribution. The central prediction concerns RNA: when chaperone-facilitated folding is coupled to self-splicing, the yield of spliced pre-RNA is maximal only at a moderate level of chaperone activity, with the optimum at $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM} k_s / \\kappa}$, where $\\kappa$ is the ratio of chaperone-induced unfolding rates from native and misfolded RNA. If correct, the theory unifies protein and RNA chaperone biology and gives a quantitative, testable target for chaperone activity in the cell.","feed_headline":"RNA splicing peaks at moderate chaperone activity, theory says","feed_subtitle":"A unified iterative-annealing framework says both GroEL and CYT-19 evolved to maximize native-state yield on biological times.","key_machinery":"The load-bearing objects are the kinetic partitioning factor $\\Phi$ and the relative disruption factor $\\kappa$. $\\Phi = k_{UN}/(k_{UM}+k_{UN})$ is the fraction of molecules that fold directly to the native state on a rugged landscape, and it is the parameter the chaperone acts on by repeatedly unfolding the misfolded fraction. $\\kappa = k_{NU}([C],[T])/k_{MU}([C],[T])$ measures how strongly the chaperone disrupts the native state relative to misfolded states; $\\kappa = 0$ recovers the GroEL case, while CYT-19 has $\\kappa \\approx 0.13$. These two parameters enter the iterated-annealing yield formula, and together with the splicing rate $k_s$ and the folding rate $k_{IM}$ they produce the optimum condition $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM}k_s/\\kappa}$ that carries the paper's central RNA prediction.","core_discovery":"The central discovery claimed here is that the iterative annealing mechanism, originally introduced for chaperonin-assisted protein folding, also describes RNA chaperones such as CYT-19, so that a single nonequilibrium principle covers both. In this mechanism, the fraction $\\Phi$ of molecules that fold directly is set by kinetic partitioning on a rugged landscape; the remaining fraction is trapped in misfolded states. The chaperone repeatedly unfolds trapped molecules, so after $n$ rounds the fraction folded is $P_N(n) = 1 - (1 - \\Phi)^n$ when only misfolded states are disrupted. The paper generalizes this to RNA chaperones by introducing $\\kappa = k_{NU}/k_{MU}$, the ratio of chaperone-induced unfolding rates from native and misfolded states, giving $P_N^\\kappa(\\infty) = \\Phi/[\\kappa + (1-\\kappa)\\Phi]$; for CYT-19, $\\kappa \\approx 0.13$, so the steady-state yield stays below unity, matching the measured loss of native ribozyme at high CYT-19 concentration. Extending the network to include self-splicing yields the non-monotonic prediction that spliced pre-RNA is maximized at an intermediate chaperone activity, $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM} k_s/\\kappa}$, because the chaperone must rescue misfolded RNA without destroying the native state before splicing can occur. The paper's broader assertion is that both machines have evolved to maximize native-state production on biological times, and that this is why chaperone action is a far-from-equilibrium, ATP-consuming process.","pith_inferences":["A direct test of the paper's central RNA prediction would be to titrate CYT-19 concentration in a group I intron splicing assay and look for an interior maximum in spliced RNA yield; the paper states this prediction awaits experimental tests.","If the optimum condition survives in vivo, it suggests a design principle for RNA chaperone expression: cells should tune DEAD-box protein levels to the ratio of folding and splicing rates, not simply make more chaperone.","By extension, a similar moderate-dosage optimum should appear for any chaperone that can disrupt the functional native state of its substrate, including DEAD-box proteins involved in ribosome assembly, a direction the paper already gestures toward with CsdA.","Because $\\kappa$ is a ratio of rates that each depend on chaperone and ATP concentration, the theory's sharpest prediction assumes that ratio stays fixed; measuring $k_{NU}$ and $k_{MU}$ separately would show whether the optimum exists across concentrations or only in a narrow window."],"forward_implications":["For GroEL-type chaperones, the formula $P_N(n) = 1 - (1 - \\Phi)^n$ implies that even a substrate with a tiny partition factor such as Rubisco ($\\Phi \\approx 0.05$) is driven to near-complete native yield after enough rounds, with full rescue predicted in roughly 40 seconds at equal chaperone and substrate concentrations.","For RNA chaperones with $\\kappa > 0$, the steady-state native yield is strictly below unity, so the theory explains why ribozyme activity falls at high CYT-19 concentration rather than saturating.","When folding is coupled to self-splicing, the yield of spliced pre-RNA is predicted to be non-monotonic in chaperone activity, with a maximum set by $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM}k_s/\\kappa}$; this gives a concrete target for chaperone expression levels in vivo.","The product of steady-state yield and inverse relaxation time, $P_N/\\tau$, is an increasing function of chaperone concentration for both GroEL and CYT-19, even where the yield itself decreases for RNA, so the theory identifies the relevant biological objective as production rate rather than yield alone."],"supporting_citations":[{"why":"introduces the iterative annealing mechanism and the yield formula for chaperonin-facilitated protein folding that the paper builds on","marker":"[6]"},{"why":"supplies the kinetic network for folding, CYT-19 action, and self-splicing that yields the moderate-activity optimum","marker":"[20]"},{"why":"shows that chaperones drive substrates out of equilibrium to maximize native-state yield on biological times","marker":"[17]"},{"why":"gives the earlier unified iterative annealing treatment of GroEL and RNA chaperones","marker":"[19]"},{"why":"establishes the kinetic partitioning value $\\Phi \\approx 0.08$ for the Tetrahymena ribozyme, setting the need for an RNA chaperone","marker":"[10]"},{"why":"identifies CYT-19 as an ATP-dependent RNA chaperone in group I intron splicing","marker":"[24]"},{"why":"provides the experimental finding that native ribozyme yield decreases with increasing CYT-19 concentration, which $\\kappa > 0$ explains","marker":"[74]"}],"fun_headline_variants":["Moderate RNA chaperone maximizes splicing yield","Unified theory: one mechanism for protein and RNA chaperones","Iterative annealing: common principle for chaperones","Splicing peaks at intermediate chaperone activity","Theory unifies protein and RNA chaperone action"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The prediction that a moderate level of RNA chaperone activity is optimal assumes that the ratio $\\kappa = k_{NU}/k_{MU}$ of chaperone-induced unfolding rates from native and misfolded RNA stays fixed when chaperone or ATP concentration changes; if $\\kappa$ varies, the location or even the existence of the predicted optimum changes.","fun_headline_variants_meta":{"raw":{"variants":["Moderate RNA chaperone maximizes splicing yield","Unified theory: one mechanism for protein and RNA chaperones","Iterative annealing: common principle for chaperones","Splicing peaks at intermediate chaperone activity","Theory unifies protein and RNA chaperone action"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000169,"raw_usage":{"total_tokens":1312,"prompt_tokens":1043,"completion_tokens":269,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":659,"completion_tokens_details":{"reasoning_tokens":196}},"tokens_in":659,"tokens_out":269,"duration_ms":4059,"temperature":1.0,"reasoning_tokens":196,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T00:44:25.648983+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Titrate CYT-19 concentration against the yield of self-spliced group I intron pre-RNA in an in vivo or reconstituted assay, holding the transcript and ATP pool fixed. If the yield rises monotonically to a plateau instead of peaking at an intermediate concentration, or if the maximum moves sharply with ATP concentration in a way inconsistent with a fixed $\\kappa$, the predicted optimum $k_{MI}^{\\rm eff} \\approx \\sqrt{k_{IM}k_s/\\kappa}$ fails.","supporting_citations":[{"cited_title":"Moderate activity of RNA chaperone maximizes the yield of self-spliced pre-RNA in vivo","cited_arxiv_id":null,"evidence_quote":"supplies the kinetic network for folding, CYT-19 action, and self-splicing that yields the moderate-activity optimum"},{"cited_title":"Molecular Chaperones Maximize the Native State Yield on Biological Times by Driving Substrates out of 18 Equilibrium","cited_arxiv_id":null,"evidence_quote":"shows that chaperones drive substrates out of equilibrium to maximize native-state yield on biological times"},{"cited_title":"Iterative annealing mechanism explains the functions of the groel and rna chaperones","cited_arxiv_id":null,"evidence_quote":"gives the earlier unified iterative annealing treatment of GroEL and RNA chaperones"},{"cited_title":"Folding of rna involves parallel pathways","cited_arxiv_id":null,"evidence_quote":"establishes the kinetic partitioning value $\\Phi \\approx 0.08$ for the Tetrahymena ribozyme, setting the need for an RNA chaperone"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"identifies CYT-19 as an ATP-dependent RNA chaperone in group I intron splicing"},{"cited_title":"Bhaskaran and R","cited_arxiv_id":null,"evidence_quote":"provides the experimental finding that native ribozyme yield decreases with increasing CYT-19 concentration, which $\\kappa > 0$ explains"}],"review_version":1}