Why BGPT?
logo

Assess an author's data and outputs

See the raw experimental evidence behind an author's publications and reproducibility signals.







Press Enter ↵ to solve



    Fuel Your Discoveries




     Quick Explanation


    Yi Li — scientific-leaning author snapshot (evidence-based, skeptical)
    Based on the provided set of papers/excerpts, the strongest signals are: (i) mechanistic breadth across molecular/cellular biology and computational biology, and (ii) multiple works with high stated scientific depth/rigor and substantial experimental detail (e.g., semisynthetic autophagy protein acetylation; photosystem I cryo-EM; TME multi-omics modeling; orthogroup/genome resources; and mechanistic T6SS toxin biology).
    Main caution: “Yi Li” is an extremely common name; without an unambiguous author identifier (ORCID/affiliation disambiguation) and without full publication list + author-position metadata, bibliometrics and “author credit” attribution can be confounded.



     Long Explanation


    Author Review: Yi Li
    Date context: April 22, 2026
    Evidence basis: only the information explicitly provided in the prompt (paper excerpts + metrics + DOIs) and the cited paper pages.

    1) Bibliometrics & identity caveat (skeptical)

    The prompt provides citation metrics for “Author Y. Li” (h-index=3; total citations=22; paper count=8; papers listed). No source link is provided for those metrics inside the prompt, so I can’t independently verify them here.
    Additionally, OpenAlex results in the prompt show many different “Yi Li / Li Y / Y. Liu / etc” records, underscoring a major name-disambiguation risk. This makes any direct claim like “Yi Li did X” sensitive to whether the correct individual is identified.
    Therefore: the rest of this review focuses on the scientific signals present in the provided paper excerpts/DOIs, not on unverified global credit.

    2) Visual summary of the provided paper-evidence set

    Note: the bar chart uses only the paper_scientific_quality_score values present in the prompt’s extracted objects.

    3) Cross-paper scientific strengths (what shows up repeatedly)

    (A) Mechanistic modeling + quantitative assays
    • Autophagy mechanism: Atg3 acetylation at K19/K48 is analyzed via semisynthesis, with measured changes in membrane binding and enhanced Atg8 lipidation, including binding affinities via SPR/MST and functional reconstitution into PE liposomes.
    • Structural mechanistic insight: a 2.23 Å cryo-EM reconstruction of Euglena PSI-LHCE links pigment composition to antenna architecture and reports quantitative energy-transfer metrics (e.g., trapping time and quantum efficiency).
    (B) Bioinformatics & data integration capability
    • Survival prediction from integrated single-cell TME + multi-omics is supported by explicit dataset sizes (CRC scRNA-seq atlas; bulk cohorts; spatial platforms) and quantified metrics (C-index and AUCs).
    • A genome resource for fission yeasts integrates orthology, alignments, phylogenies, and local synteny across nine chromosome-level assemblies and reports SOG counts and category breakdowns.
    (C) Experimental discipline appears in multiple domains
    • The Atg3 study includes careful binding assays (SPR/MST) and reconstituted liposome systems, and explicitly discusses in vitro vs lysate differences.
    • The SPARP-seq work is framed around measurement technology (single-molecule long-read) tied to translation states and poly(A) tail length estimation, validated by comparison to ribosome profiling.

    4) Critical blind spots / uncertainty drivers (what could weaken the inference)

    • Name disambiguation: the prompt’s OpenAlex snippet suggests multiple similarly named authors exist; without ORCID/affiliation-to-paper mapping, “Yi Li did this” can be wrong even if the papers themselves are solid.
    • Attribution mismatch: many provided excerpt DOIs correspond to different likely authors (not provably the same Yi Li). So this review evaluates scientific signals within the provided set, not a guaranteed author-level track record.
    • Publication bias & narrative bias: some items are explicitly narrative reviews (e.g., the lung cancer landscape and acupuncture review). Narrative syntheses are useful for orientation but are less directly evidential than primary mechanistic work.
    • Reproducibility vs “score optimism”: several prompt-provided fields separate novelty/quality/generality from reproducibility. For example, some mechanistic works rate high in scientific quality but only moderate/reduced reproducibility score (as provided in the prompt), which may reflect limited public deposition or complex experimental replication.
    • Cross-species generalization: some systems are yeast, plants, mice, or cell lines. Transfer to human biology can be nontrivial.

    5) Reproducibility vs novelty vs usefulness (2nd visual)

    Interpretation constraint: x/y/size come from prompt-provided numeric fields, not independently measured.

    6) What I would look for to validate Yi Li’s authorship and strengthen evidence

    • Author identity: ORCID and/or a disambiguated author-to-DOI mapping.
    • Contribution clarity: author position (first/last/corresponding) and stated contributions.
    • Reproducibility: public code/data deposition links (not “available on request”) where possible; full methods and accession numbers.
    • Replication: independent cohorts/orthogonal assays for the core claims (especially for predictive models and in vivo-to-human extrapolations).

    Evidence citations used (from provided prompt DOIs)

    Included representative citations across domains: autophagy semisynthesis , PSI-LHCE cryo-EM , CRC TME prediction model , and SPARP-seq translation-tail coupling .


    Feedback:    

    Updated: April 22, 2026

     Top Data Sources ExportMCP


     Hypothesis Graveyard


    The claim that “membrane binding affinity increases alone explains functional activation” for Atg3 acetylation is weakened by the reported modest Atg3–Atg8 affinity changes; thus a simpler affinity-only mechanism is less favored than the co-localization/tethering model.


    A strong-but-incorrect simplification would be that “more novelty automatically implies better reproducibility.” The prompt’s own reproducibility-vs-novelty tradeoffs (as provided) suggest novelty and reproducibility are separable and must be evaluated independently.

     Science Movie


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




     Discussion








    Get Ahead With Science Insights

    Custom summaries of the latest cutting edge Science research. Every Friday. No Ads.


    My BGPT






     Trending