Why BGPT?
logo

Review papers by their claims

Assess a manuscript by extracting its claims, linked experiments, exact results, and limitations for reproducible review.Know what the science actually supports before you trust the answer.

Press Enter ↵ to extract claims



    Explore by Goal









     Quick Explanation


    Across 681 single-site and 988 double-site fX174 G-protein variants, the study finds that large destabilization (FoldX-predicted DDG) strongly predicts inviability, while FoldX-based stability explains only a weak fraction of fitness variation among viable mutants (e.g., multivariate fitness regression shows R² ≈ 0.013). The authors’ multistage binomial model yields “half-max” survival around DDGfold ≈ 3 kcal/mol and DDGbind ≈ 6 kcal/mol, with non-100% maximal survival attributed to additional unobserved factors limiting determinism.


     Long Explanation


    Evidence for the key claim (threshold on inviability, weak prediction of fitness among viable mutants)

    Reported quantitative patterns: Single-site variants (n=681 after exclusions) show 27% inviability when only observed at T0, vs 73% viable when observed after T0, and MBM estimates imply each +1 unit DDGfold decreases log-odds of survival substantially (b1≈-1.4047) with similar but smaller slope for DDGbind (b1≈-0.385), with an MBM-predicted maximal survival probability capped at ~0.87 for singles.

    Fitness vs stability (among viable variants): Linear regression on viable mutants (n=498) yields weak negative coefficients (DDGfold ≈ -0.158, p≈0.023; DDGbind ≈ -0.062, p≈0.427) and very small overall explanatory power (R² ≈ 0.013 unadjusted), leading the authors to conclude stability alone is not a reliable fitness predictor within the viable set.

    Most important alternative explanations / missing discriminators

    The MBM asymptote (<1 maximal survival) is interpreted as “unobserved factors,” but the paper does not quantify (i) how much of the capped probability comes from FoldX-model systematic bias vs experimental sampling/stochasticity vs other biological steps (e.g., assembly, host interactions). The double-mutant analysis assumes additive DDG for doubles; if non-additivity (epistasis) is common, the inferred thresholds may be partially model-driven.

    Practical implications

    For variant classification, the work supports a “stability-threshold” intuition: highly destabilizing capsid spike mutations are reliably lethal, but within viability, many mutations may remain phenotypically tolerated through compensating biological mechanisms not captured by DDGfold/DDGbind alone.



    Feedback:    

    Updated: July 19, 2026

     BGPT Paper Review


    Study Novelty

    70%

    Novelty is moderate: it applies deep mutational scanning to fX174 G for stability/viability and foregrounds a multistage binomial model with explicit probability caps and FoldX error integration; the novelty is primarily methodological/modeling and system-specific rather than entirely new biology.


    Scientific Quality

    70%

    Strengths: large DMS variant coverage (681 singles; 988 doubles), explicit statistical modeling (MBM) with uncertainty, and direct comparisons to linear regression. Red flags/uncertainty: major dependence on FoldX+MD snapshot DDG predictions, additive DDG assumption for doubles, and limited ability (from the provided text) to disentangle model bias from experimental stochasticity in producing the capped survival probability.


    Study Generality

    60%

    The study is centered on one phage capsid protein (fX174 G) and specific folding/binding proxies; extending the exact threshold values or predictive performance to other proteins requires additional datasets/tests.


    Study Usefulness

    70%

    Practical value is strongest for classification of strongly destabilizing substitutions; weaker for ranking fitness within the viable set.


    Study Reproducibility

    60%

    Without accessible model/code/data, independent reconstruction of exact coefficients/curves may be difficult.


    Explanatory Depth

    80%

    The paper provides a mechanistic-statistical story: multistage failure (folding/binding) yields threshold-like inviability, while non-100% survival reflects additional steps not captured by DDGfold/DDGbind; this is internally coherent and quantifies effect sizes.


    🎁 Authors: Collect 197 Free Science Tokens (≈ $19.7 USD)

    Claim My Author Tokens

    Use for 49 days of free BGPT access (4 tokens = 1 day) or trade/sell (≈ $19.7 USD)

     Top Data Sources ExportMCP


     Analysis Wizard



    Reformat reported MBM and regression outputs into structured tables; generate a threshold-focused plot of survival probability vs DDG using only provided coefficients and probability limits.



     Hypothesis Graveyard


    Assuming additive DDG for doubles is “good enough” may be too optimistic: without quantifying epistatic deviation, it risks attributing epistasis to the unobserved-factor asymptote rather than to explicit stability non-additivity.


    The notion that DDGfold/DDGbind are sufficient to explain both inviability and fitness is falsified by the reported R²≈0.013 among viable mutants; stability-only explanations cannot account for fitness variation within viability.


    novel_experiments single-site multiple replicates would strengthen inference

     Science Art


    Paper Review: Modeling the Relationship between the Capsid Spike Protein Stability and Fitness in ϕX174 Bacteriophage Science Art

     Science Movie


    Make a narrated HD Science movie for this answer ($32 per minute)




     Discussion


    Stay current without chasing every paper.

    Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.


    My BGPT






     Trending