Science brief: a reading aid, unvalidated by any person. Not a prediction and not advice. Public data only.

Science brief: lab of Jay Shendure

As of 2025-12-31. Target mode. Literature: single-cell sequencing. 60 abstracts read, 4 claims checked, 5 papers had no abstract and were not read.

Read this first.

• Each claim was extracted by a model from one abstract. A second, separate reader then checked it against the abstracts in this pack, and a third checked that reader. Every line below carries two exact quotes that code found in the sources.
• The corpus is each lab's three most recent last-author papers at four year-ends, abstracts only. "No independent support found" means none in this corpus. It never means not replicated.
• The order is a published sort rule (scirank-v1): highest evidence tier among verified claims, then independent supports, then fewer contradictions. It is not a score and it predicts nothing.
• Evidence tier comes from the extraction's model-system field. "Randomized" is a word rule on the abstract.

Measured, once, on planted errors (an upper bound on real ones):

• One pass over the retrieved abstracts found 110 of 121 hard planted misstatements and named the paper that shows each (91%, 95% interval 85 to 96). It flagged 2 of 423 unaltered claims (false alarms).
• By type: a flipped direction 58 of 66 (88%, 81 to 94); a number swapped for another in the same abstract 35 of 37 (95%, 83 to 100); animal and human swapped 17 of 18 (too few for a rate).
• Baselines on the same claims: a read with no literature flagged 52% and could name no paper; the Lens with a string rule found 2%; fan-out over chunks found 89%, not better than one pass.
• Planted errors are an upper bound on real ones: one word or number was changed and the true paper was in the pack. No person has checked a real case. Reader and extractor are both Claude models; scoring was by code against a key.
• Supported records, the framed read and the ranking are not measured by this test.

Extraction: Measured accuracy on a fresh sample (agent readers, both must agree): output_type right 97% (92%–99%); model_system right 95% (90%–98%); scale right 97% (93%–99%); claim supported 98% (94%–99%); effect numbers supported 99% (92%–100%); limitations supported 93% (85%–97%).

Where a paper disagrees with a claim

None found in this pack.

Claims and their evidence

Jay Shendure, University of Washington

Company, PI named on the filing: none found

• C01 Global 3' UTR lengthening occurs across embryonic stages in all cell types, with shorter 3' UTRs in hematopoietic lineages and longer 3' UTRs in neuronal cell types; ELAV-like RBPs are putative APA regulators.
  • animal; verified against its paper; no independent support found in this corpus; source: The landscape of alternative polyadenylation in single cells of the developing mouse embryo (2021-08-24, P19)
  • Own paper:
    • claim: "shorter 3' UTRs in hematopoietic lineages and longer 3' UTRs in neuronal cell types" / paper: "global lengthening of 3' UTRs across embryonic stages in all cell types, although we detect shorter 3' UTRs in hematopoietic lineages" — The landscape of alternative polyadenylation in single cells of the developing mouse embryo (2021-08-24, P19). The paper reports the same global 3' UTR lengthening across stages with shorter UTRs in hematopoietic lineages. [R4115]
• C03 P3 editing links protein-protein proximity to formation of a functional CRISPR-Cas9 dual-component guide RNA, activating prime or base editing in human cells.
  • in vitro; verified against its paper; no independent support found in this corpus; source: A molecular proximity sensor based on an engineered, dual-component guide RNA (2025-02-12, P18)
  • Own paper:
    • claim: "links protein-protein proximity to formation of a functional CRISPR-Cas9 dual-component guide RNA" / paper: "which links protein-protein proximity to the formation of a functional CRISPR-Cas9 dual-component guide RNA." — A molecular proximity sensor based on an engineered, dual-component guide RNA (2025-02-12, P18). The paper describes the same P3 editing strategy and its activation of prime or base editing in human cells. [R4117]
• C04 Libraries of extant, ancestral, mutagenized or miniaturized polymerase III promoter and gRNA scaffold variants yield thousands of parts for reproducible multiplex prime editing in mammalian cells, including hundreds exceeding common sequences' activity. Reported effect: Thousands of parts identified; hundreds with greater activity than commonly used sequences; no numeric values stated in abstract.
  • in vitro; verified against its paper; no independent support found in this corpus; source: A parts list of promoters and gRNA scaffolds for mammalian genome engineering and molecular recording (2025-11-11, P08)
  • Own paper:
    • claim: "including hundreds exceeding common sequences' activity" / paper: "We identify thousands of parts for reproducible editing in human and mouse cell lines, including hundreds with greater activity than commonly used sequences." — A parts list of promoters and gRNA scaffolds for mammalian genome engineering and molecular recording (2025-11-11, P08). The paper reports thousands of parts and hundreds with greater activity than commonly used sequences. [R4118]
• C02 COMET, a combinatorial framework, tests many E3 ligases against many candidate substrates in one experiment, revealing many complex rather than 1:1 E3-substrate relationships.
  • not stated; verified against its paper; no independent support found in this corpus; source: Combinatorial mapping of E3 ubiquitin ligases to their target substrates (2025-02-01, P56)
  • Own paper:
    • claim: "revealing many complex rather than 1:1 E3-substrate relationships" / paper: "Our data suggest that many E3-substrate relationships are complex rather than 1:1 associations." — Combinatorial mapping of E3 ubiquitin ligases to their target substrates (2025-02-01, P56). The paper states the same finding for COMET. [R4116]

Framed read (unvalidated: one model's synthesis of the records above)

Claim that carries the most weight: C04. C04 is the claim most tied to a reusable, countable asset (a parts library), and C03 about editing in human cells is only useful at scale if such parts perform as stated. Its only record, R4118, comes from the claim's own paper, so everything leaning on it rests on a single non-independent source.

What would have to hold:

• The hundreds of parts with greater activity than commonly used sequences must hold up outside the originating paper; the only record is R4118, and no independent paper in this pack addresses it.
• Reproducibility of the parts in human and mouse cell lines (R4118) must extend to the multiplex prime editing setting in mammalian cells that C04 names; no record covers multiplex performance directly.
• The P3 editing link between protein-protein proximity and a functional dual-component guide RNA (R4117) must work beyond the originating paper; no independent paper in this pack addresses it.
• The finding in R4116 that many E3-substrate relationships are complex rather than 1:1 must not depend on the one framework that produced it; no record covers an orthogonal method.
• The 3' UTR pattern in R4115 is supported only for shorter UTRs in hematopoietic lineages; no record covers longer UTRs in neuronal cell types or the ELAV-like regulator role in C01.

Questions a data room would have to answer:

• (C04) Which comparator sequences were used to define 'commonly used', how was activity measured for each, and can the raw per-part data for the parts above that comparator be shared?
• (C04) Were the parts tested in a multiplex prime editing format in mammalian cells, or only singly, and has any part been re-tested by a group outside the originating lab?
• (C03) What is the editing efficiency of P3 editing versus standard prime or base editing at the same loci in human cells, and what is the background activity when the proximity partners are absent?
• (C02) How many E3-substrate pairs in COMET were confirmed by a second, non-combinatorial assay, and what fraction of the complex relationships could be artifacts of overexpression?
• (C01) What evidence in the 3' UTR dataset supports the ELAV-like RBPs as APA regulators beyond correlation, and was the neuronal lengthening checked against an independent dataset?

Cannot say:

• Whether any claim here is reproduced by a different lab: all four records are from the claim's own paper and none is independent.
• Whether the effect sizes in C04 are large, since the records give only the words thousands and hundreds with no numeric values.
• Whether any of the four claims has commercial or therapeutic relevance, or how the lab's pieces fit together as a company.
• Whether the neuronal half of C01 or the ELAV-like regulator role is supported, since R4115 quotes only the hematopoietic side.

What was not shown

• Claims with no checked record: none.
• Records proposed: 4; refused by code (a quote not found, or a number of the reader's own): 0; removed by the checker: 0.
• Papers without an abstract, not read: 5.

Run record

• run: sci-e07e08d2a1; reader arm: B1R; spec: sci-0.1; rank rule: scirank-v1
• reader prompt sha256: c04daba6743bf488eaeb3427cec02860a257d2c06ef9f58cf307195206620e03; model: claude-sonnet-5-5, medium effort
• extraction: claude-haiku-5-5, prompt extract_v0.3; no human has verified it