RNA-Triggered Cell Killing with CRISPR–Cas12a2
Peer-reviewed Nature 2026; open access, 2026 · paper · Biomedicine
Shows that the CRISPR nuclease Cas12a2, once triggered by a target RNA, shreds DNA and kills yeast and human cells expressing that transcript. It is used to eliminate cells with papillomavirus, unedited cells, or a KRAS mutation, with no off-target activation observed.
Checked: PDF from nature.com, fetched 5 Oct 2026. Page numbers are those of that PDF.
9 findings
Reference or label The Competing interests statement declares an employment interest for initials that match no author: "A.G., P.S., T.F., R.B., D.S. and R.M. are employees of Akribion Therapeutics GmbH."
- p. 16 Competing interests
A.G., P.S., T.F., R.B., D.S. and R.M. are employees of Akribion Therapeutics GmbH.
- p. 1 Author list
Paul Scholz1,10 ✉, Jared Thompson2,10, Kadin T. Crosby3,10, Torsten Fauth1
- p. 16 Competing interests
A.G., P.S., T.F., R.B., D.S. and R.M. are employees of Akribion
- p. 16 Competing interests
A.G., P.S., T.F., R.B., D.S.
Only the authors can say what is right here.
Judge's note: Read crops C025-a.png, C025-c.png and C025-d.png (page 16) for the sentence, and C025-b.png (page 1) for the full byline: Paul Scholz, Jared Thompson, Kadin T. Crosby, Torsten Fauth, Nathan M. Krah, Grant Schlauderaff, Robin Back, Zachary A. Berkheimer, Alivia Jolley, Dirk Sombroek, Rebekka Medert, Christian Zurek, Oleg Dmytrenko, Emily Wilson, Friso T. Schut, Jared Rutter, Xiaoyang Zhang, Michael Krohn, Ryan N. Jackson, Chase L. Beisel, Yang Liu. No author's initials are A.G.; the other five sets in the same sentence map to authors carrying affiliation 1 (P.S. Scholz, T.F. Fauth, R.B. Back, D.S. Sombroek, R.M. Medert), and the author-contributions paragraph on the same page uses G.S. for Schlauderaff and A.J. for Jolley. The intended person cannot be determined from the files, so author_query.
Second check: another model agreed
Report a problem with this finding
Reference or label Two different GEO accessions are given for the study's sequencing data: Data availability states "RNA-seq and PFS screening data are publicly available at NCBI GEO under accession no. GSE325194", while the Reporting Summary bound into the same PDF states "All relevant sequencing data are publicly available at NCBI GEO under accession number GSE287139".
- p. 15 Data availability
RNA-seq and PFS screening data are publicly available at NCBI GEO under accession no. GSE325194.
- p. 2 Reporting Summary, Data
All relevant sequencing data are publicly available at NCBI GEO under accession number GSE287139.
- p. 27 Reporting Summary (bound into the article), Data
accession number GSE287139
- p. 27 Reporting Summary bound into the article PDF, Data section
NCBI GEO under accession number GSE287139
Only the authors can say what is right here.
Judge's note: Read crops C022-a.png, C022-e.png, C022-g.png and C022-h.png (page 15 Data availability, 'GSE325194') and the page image main-p027.png (Reporting Summary Data box, 'GSE287139'; identical text on supp-01 page 2, viewed as supp-01-p002.png). The two statements otherwise match word for word (same Hallmark/MSigDB sentence, same gRNA/target-sequence sentence), so they describe the same deposit; no note of a second series or of re-deposition appears anywhere in the files. Which accession is live cannot be checked from the files, so this is an author query rather than an evident correction.
Second check: another model agreed
Report a problem with this finding
Reference or label Reporting Summary "Antibodies used" item 1 declares "1. Rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)" (matching Methods, "rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)"), but the Validation entry 1 in the same box gives "https://www.bio-techne.com/p/antibodies/53bp1-antibody_nb100-305", a different catalogue number.
- p. 3 Reporting Summary, Antibodies (used / validation)
1. Rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)
- p. 3 Reporting Summary, Antibodies, Validation item 1
https://www.bio-techne.com/p/antibodies/53bp1-antibody_nb100-305
- p. 13 Methods, Immunofluorescence microscopy
rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)
- p. 3 Reporting Summary, Antibodies used / Validation
Rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)
Only the authors can say what is right here.
Judge's note: Read the page image extract/png/supp-01-p003.png: the Antibodies box reads "1. Rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)" and the Validation box directly below reads "1. https://www.bio-techne.com/p/antibodies/53bp1-antibody_nb100-305". The identical Reporting Summary page bound into the article, extract/png/main-p028.png, shows the same two strings. The Methods crop extract/crops/C031-c.png (Immunofluorescence microscopy, main p13) reads "rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306) diluted 1:1,000", so the article consistently uses -306 and only the validation link uses -305. This is an internal inconsistency visible entirely within the supplied files: the validation link is offered specifically as validation for item 1, yet encodes a different catalogue number, so the reagent identifier is ambiguous. I did not need, and did not rely on, any outside fact about whether NB100-305 and NB100-306 are related products; from the files the two identifiers simply disagree. Only the authors can say which of the two is correct, hence author_query; the data are unaffected, so S2.
Second check: another model agreed
Report a problem with this finding
Reference or label Fig. 4h legend reads 'Quantification of indel frequency by Sanger sequencing when performing FnCas12a edits followed by introduction of GeCas12a2 in HeLa-GFP cells', whereas the Methods and Extended Data Fig. 8a place the same experiment in HEK293-GFP cells.
- p. 6 Fig. 4 legend, panel h
when performing FnCas12a edits followed by introduction of GeCas12a2 in HeLa-GFP cells
- p. 14 Methods, Indel enrichment
To induce indels in the GFP transgene, 1 × 105 HEK293-GFP cells were electroporated with purified FLAG-tagged FnCas12a
- p. 24 Extended Data Fig. 8a, workflow schematic
HEK293-GFP Population
- p. 6 Fig. 4 legend, h
followed by introduction of GeCas12a2 in HeLa-GFP cells
Judge's note: Read crop C006-a and C006-d (Fig. 4 legend, page 6) — the legend plainly reads 'in HeLa-GFP cells'. Read crop C006-b and the full page main-p014.png (Methods, 'Indel enrichment'): '1 × 10^5 HEK293-GFP cells were electroporated with purified FLAG-tagged FnCas12a (WP_003040289) RNPs ... To deplete unedited cells, 1 × 10^5 cells were electroporated with 50 pmol purified FLAG-tagged GeCas12a2 RNPs'. Read the full page image main-p024.png: the Extended Data Fig. 8a workflow labels both rows 'HEK293-GFP Population', and the Methods contain only one indel-enrichment protocol. The Fig. 4i prime-editing panel is consistently HEK293-GFP in both legend and Methods. The legend cell-line label is therefore the outlier; the plotted data are unaffected, so S2 with the evident fix HEK293-GFP.
Second check: another model agreed
Report a problem with this finding
Reference or label In the FAM-labelled collateral cleavage assay Methods the buffer component BSA is expanded as a silylating reagent: "1× NEB 3.1 buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, 100 µg ml–1 bis(trimethylsilyl)acetamide (BSA), pH 7.9)", whereas the same abbreviation is used elsewhere in the Methods for a protein ("blocked with 2% BSA in 1× PBS") and the identical buffer recipe is printed later without the expansion ("100 µg ml−1 BSA, pH 7.9").
- p. 12 Methods, FAM-labelled collateral cleavage assays
1× NEB 3.1 buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, 100 µg ml–1 bis(trimethylsilyl)acetamide (BSA), pH 7.9)
- p. 13 Methods, Immunofluorescence microscopy
samples were blocked with 2% BSA in 1× PBS
- p. 12 Methods, FAM-labelled collateral cleavage assays
100 µg ml–1 bis(trimethylsilyl)acetamide (BSA)
- p. 13 Methods, Immunofluorescence microscopy
blocked with 2% BSA in 1× PBS
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the 100 µg ml−1 BSA component of NEB buffer 3.1 is bovine serum albumin and that bis(trimethylsilyl)acetamide is an unrelated water-reactive silylating reagent. I verified the printed text on crop C017-a.png (page 12), which shows the expansion verbatim, and on crops C017-b.png / C017-d.png (page 13), where '2% BSA in 1× PBS' is used as a blocking agent. Within the files the same buffer composition is restated in the off-target in vitro testing section on page 14 (viewed on main-p014.png) as '100 mM NaCl, 50 mM Tris-HCl, 1 mM MgCl2, 100 µg ml−1 BSA, pH 7.9' with no expansion, so one abbreviation is given two incompatible meanings in the same Methods. Only the authors can confirm the intended expansion, hence author_query.
Second check: another model agreed
Report a problem with this finding
Reference or label In the Reporting Summary the validation link for the anti-53BP1 antibody points to a different catalogue number than the antibody used: "https://www.bio-techne.com/p/antibodies/53bp1-antibody_nb100-305" against "Rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)" listed directly above it and in the Methods.
- p. 28 Reporting Summary, Antibodies
53bp1-antibody_nb100-305
- p. 13 Methods, Immunofluorescence
Novus Biologicals, NB100-306
Judge's note: Read the page image main-p028.png, where the Antibodies box lists 'NB100-306' and the Validation box immediately below gives the URL ending 'nb100-305'; the digits are legible on the image. Crop C028-b.png confirms the Methods text 'rabbit anti-53BP1 antibody (Novus Biologicals, NB100-306)' on page 13. Two independent statements give -306 and only the link gives -305, so the pointer is the odd one out.
Second check: another model agreed
Report a problem with this finding
Reference or label The Extended Data Fig. 2 legend attributes the branch-support method to a patent: "Branch confidence is shown on a 1-100 scale as ultrafast bootstrap values (UFBoot)76", where reference 76 is "Scholtz, P., Zurek, C. & Krohn, M. Novel, non-naturally occurring CRISPR-Cas nucleases for genome editing. World Patent no. WO/2022/017633 (2022)."
- p. 15 Reference list, ref. 76
Scholtz, P., Zurek, C. & Krohn, M. Novel, non-naturally occurring CRISPR-Cas nucleases for genome editing. World Patent no. WO/2022/017633 (2022).
- p. 16 Competing interests
nuclease sequence (WO 2022/017633) and its use for cell depletion (WO 2023/139096), with P.S., C.Z. and M.K. as inventors
- p. 18 Extended Data Fig. 2 legend, panel a
Branch confidence is shown on a 1-100 scale as ultrafast bootstrap values (UFBoot)76.
- p. 18 Extended Data Fig. 2 legend
ultrafast bootstrap values (UFBoot)76
Only the authors can say what is right here.
Judge's note: Read crops C023-c.png and C023-d.png (page 18 legend, superscript 76 printed immediately after '(UFBoot)') and C023-a.png / C023-e.png / C023-f.png (page 15 reference list, entry 76 verbatim as quoted); also viewed main-p018.png. Within the files the mismatch is evident from the reference's own title: a patent on nuclease sequences cannot be the source for an ultrafast-bootstrap branch-support method. The Methods (crop C018-b.png, page 11) state the phylogeny was reconstructed with IQ-TREE v2.0.3 (ref. 62) and that 'branch confidence was reported as ultrafast bootstrap values', with no citation at that point, and the same patent number is declared in Competing interests (crop C023-b.png) as the GeCas12a2 nuclease-sequence patent. Only the authors can say whether ref. 76 was meant as a method citation misplaced here or as a source credit for the sequence; note also that the Methods say the scale runs 0 to 100 whereas the legend says 1-100. The misspelling of the first author's surname in the same reference is judged separately under C024.
Second check: another model agreed
Report a problem with this finding
Wording Results state 'we targeted six endogenous transcripts across four cell lines derived from melanoma, lung, and head and neck cancers (Fig. 1h and Supplementary Table 2)', but the four lines keyed in Fig. 1h are HeLa-GFP, SK-MEL-30, COR-L23 and SCC-4.
- p. 3 Results, Cas12a2 eliminates human cells
we targeted six endogenous transcripts across four cell lines derived from melanoma, lung, and head and neck cancers
- p. 2 Fig. 1h, legend key
HeLa-GFP
- p. 3 Results, 'Cas12a2 eliminates human cells'
we targeted six endogenous transcripts across four cell lines derived from melanoma, lung, and head and neck cancers (Fig. 1h and Supplementary Table 2)
- p. 2 Fig. 1h, panel key
HeLa-GFP SK-MEL-30 COR-L23 SCC-4
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: HeLa is a cervical adenocarcinoma line, not a melanoma, lung or head-and-neck line. What I verified in the files: crop C004-a shows the sentence as quoted on page 2 (continuing on page 3); crop C004-b shows the Fig. 1h key listing exactly four lines — HeLa-GFP, SK-MEL-30, COR-L23, SCC-4 — and the Fig. 1 legend in the text layer reads 'across four different cell lines'. So the three tumour types named cover only three of the four lines plotted. The Methods culture section lists the same four lines. Rated S3 because the panel and legend are correct and only the text's summary description is imprecise.
Second check: another model agreed
Report a problem with this finding
Wording Results state 'indel formation was again evaluated after seven days (Extended Data Fig. 8a)' and Extended Data Fig. 8a shows 'After 2 days Transfer to PDL coated culture dishes' then '7 days Grow Cells', whereas the Methods state for the GeCas12a2 step 'After 10–12 days, the indel analysis was performed.'
- p. 7 Results, Cas12a2 enriches gene-edited cells
and indel formation was again evaluated after seven days (Extended Data Fig. 8a)
- p. 14 Methods, Indel enrichment
After 10–12 days, the indel analysis was performed.
- p. 24 Extended Data Fig. 8a, workflow
7 days Grow Cells
- p. 7 Results, Cas12a2 enriches gene-edited cells
indel formation was again evaluated after seven days
Only the authors can say what is right here.
Judge's note: Read crop C010-a and C010-d (page 7 Results), crop C010-b (page 14 Methods, 'Indel enrichment') and the full page image main-p024.png. The Methods give one week for the FnCas12a stage ('One week after electroporation, cells were detached and subjected to indel analysis') and then, after GeCas12a2 electroporation and a 48 h transfer, 'After 10–12 days, the indel analysis was performed'. The Extended Data Fig. 8a schematic shows 2 days + 7 days for the GeCas12a2 arm, i.e. 9 days, and the Results say seven days. Neither reading of the Methods window (10–12 days from electroporation, or 10–12 days after the 2-day transfer) matches 2 + 7 = 9 or 7. Rated S3 because the discrepancy is a culture-duration detail that leaves the protocol ambiguous rather than changing any reported value; only the authors can say which timing is right.
Second check: another model agreed
Report a problem with this finding
11 minor (typos and the like; most checked by the judge only)
Reference or label The Methods send readers to Supplementary Table 8 for the cell lines ("refer to Supplementary Table 8 for a complete list of cell lines and origins") while the Reporting Summary published with the article places that list in another table ("We included a detailed Table S11 in the Supplementary Information, describing all yeast and bacterial strains, as well as the mammalian cell lines used in this study, along with their sources"); Supplementary Table 8 is also cited for the PFS library construct ("a target sequence (Supplementary Tables 8 and 9)").
- p. 12 Methods, Mammalian cell culture
refer to Supplementary Table 8 for a complete list of cell lines and origins
- p. 11 Methods, PFS library preparation
constructed to include a target sequence (Supplementary Tables 8 and 9)
- p. 3 Reporting Summary, Eukaryotic cell lines / Cell line source(s)
We included a detailed Table S11 in the Supplementary Information, describing all yeast and bacterial strains, as well as the mammalian cell lines used in this study, along with their sources.
- p. 11 Methods, PFS library preparation
a target sequence (Supplementary Tables 8 and 9)
Only the authors can say what is right here.
Judge's note: Verified the Methods pointers on crops C018-a.png (page 12, 'refer to Supplementary Table 8 for a complete list of cell lines and origins') and C018-b.png / C018-d.png (page 11, 'Supplementary Tables 8 and 9'), and the Reporting Summary wording on the page image main-p028.png, which shows 'Table S11' in the Cell line source(s) box (the same text appears in supp-01 page 3). The Methods elsewhere cite 'Supplementary Tables 11 and 12' for the yeast nuclease/gRNA plasmid and HDR template, which conflicts with Table S11 holding strains and cell lines. I downgraded the reviewers' main argument: a single Table 8 listing strains/cell lines could legitimately be co-cited with Table 9 (sequences) for the CBS-6873 library construct, so 'one table cannot be both' is not established. The supplementary tables themselves are not among the files, so which numbering is right cannot be settled here; what is verifiable is that the article and its Reporting Summary point to different tables for the same cell-line list.
Second check: another model agreed
Report a problem with this finding
Wording Two versions of the acquisition software are reported for the 53BP1 DNA-damage-foci imaging: the Methods say "Raw images were exported from Nikon NIS-Elements software (v.5.41.02)" while the Reporting Summary says "DNA damage foci and RNA FISH imaging was carried out using an Eclipse Ti-2 spinning disk confocal microscope (Nikon) equipped with Nikon NIS-Elements software (v6.02.03)".
- p. 13 Methods, Immunofluorescence microscopy
Raw images were exported from Nikon NIS-Elements software (v.5.41.02)
- p. 1 Reporting Summary, Software and code, Data collection item 3
DNA damage foci and RNA FISH imaging was carried out using an Eclipse Ti-2 spinning disk confocal microscope (Nikon) equipped with Nikon NIS-Elements software (v6.02.03)
- p. 13 Methods, Immunofluorescence microscopy
Nikon NIS-Elements software (v.5.41.02)
- p. 26 Reporting Summary, Data collection
Nikon NIS-Elements software (v6.02.03)
Only the authors can say what is right here.
Judge's note: Read crop C019-a.png / C019-c.png (page 13 Methods: Nikon Ti2 Eclipse confocal system, NIS-Elements v.5.41.02) and the page image main-p026.png (Reporting Summary item 3, v6.02.03; identical text on supp-01 page 1, viewed as supp-01-p001.png). I checked the RNA-FISH Methods section (page 15 text) for a second version: it names a Nikon Eclipse Ti-2 with a CSU-W1 spinning disk unit but gives no software version, so the two instruments/sessions may differ. Because the Reporting Summary attaches v6.02.03 explicitly to the DNA-damage-foci imaging as well, the two documents disagree for that experiment; only the authors can say which version applies to which acquisition. Minor, affects no result.
Second check: another model agreed
Report a problem with this finding
Wording Fig. 4 legend reads 'b, Targeting within the E6 transcript of HPV18.' although panel b draws both an E7 and an E6 transcript row under 'Integrated HPV18 genome' and the Results describe gRNAs against transcripts 'encoding E6 or E7 oncoproteins (Fig. 4b, Supplementary Tables 6 and 7)'.
- p. 6 Fig. 4 legend, panel b
b, Targeting within the E6 transcript of HPV18.
- p. 6 Fig. 4b, panel labels
Integrated HPV18 genome
- p. 5 Results, Cas12a2 eliminates virally infected cells
we therefore designed gRNAs targeting the moderately expressed HPV transcripts encoding E6 or E7 oncoproteins (Fig. 4b, Supplementary Tables 6 and 7)
- p. 5 Results, HPV
(Fig. 4b, Supplementary Tables 6 and 7)
Only the authors can say what is right here.
Judge's note: Read crop C008-a (Fig. 4 legend, page 6), crop C008-b (Fig. 4 panels a–f) and the full page main-p006.png: panel b shows the integrated HPV18 genome from 1 kb to 8 kb with PE and PL promoters and two labelled transcript rows, E7 and E6; panel c reports NT, GFP, E6 and E7 targets. The legend names only E6, so it under-describes the panel. Rated S3: the panel and data are correct, only the legend's description is incomplete. On the other reviewer notes in this cluster: I read crop C008-c/C008-d (page 5, '(Fig. 4b, Supplementary Tables 6 and 7)'), crop C008-e (page 14 Methods, 'all potential activating RNAs were identified and reported (Supplementary Table 6)') and crop C008-f (page 4, '(Supplementary Tables 4–6)'). These are consistent with Table 6 holding the predicted activating-RNA list, which the HPV sentence cites alongside Table 7 for the guides; the supplementary tables themselves are not among the provided files (supp-01 is only the Reporting Summary), so no pointer defect is demonstrable there.
Second check: another model agreed
Report a problem with this finding
Wording The five protospacer-flanking positions are numbered "+1" to "+5" on the schematic and the x-axis ("PFS Position") of Extended Data Fig. 3a, but "−5" to "−1" on the schematic and x-axis of Extended Data Fig. 2b, on the target rulers of Extended Data Fig. 5a, and in the Methods ("log2 fold-change values for depletion scores were computed for nucleotides at PFS positions (−1 to −5)"), with no statement reconciling the two sign conventions.
- p. 11 Extended Data Fig. 3a, schematic and plot axis
PFS Position
- p. 11 Methods, PFS library sequencing and analysis
log2 fold-change values for depletion scores were computed for nucleotides at PFS positions (−1 to −5)
- p. 18 Extended Data Fig. 2b, PFS depletion plot
PFS region
Judge's note: Opened main-p019.png (Extended Data Fig. 3a): the BARAN schematic carries tick labels '+1' and '+5' and the SuCas12a2 depletion plot x-axis is '+1 +2 +3 +4 +5' under the boxed label 'PFS Position'. Opened main-p018.png and crop C023-c.png (Extended Data Fig. 2b): the NNNNN schematic is labelled '-5' and '-1' and the plot x-axis is '-5 -4 -3 -2 -1' under 'PFS Position'. Opened main-p021.png (Extended Data Fig. 5a): target rulers read '-5 -1 1 5 10 ...' with the PFS highlighted at -5 to -1. Crop C016-b.png confirms the Methods wording '(−1 to −5)'. I corrected the reviewer's description: the two schemes do not run in opposite directions — in both, position 1 is the base adjacent to the target and position 5 the most distal — so only the sign differs. Extended Data Fig. 3a's legend does state that the right-hand plot reproduces previously published SuCas12a2 depletion data (ref. 15), which may explain the retained convention, but the change of sign is nowhere noted, so the same coordinate label appears with two signs in one paper. Labelling point only; no data are affected.
Second check: another model agreed
Report a problem with this finding
Wording The intratumoural dose, "25 µl of the LNP formulation (containing 10 µg total RNA)", implies 0.4 mg ml−1, whereas the only LNP concentration given in the Methods is "concentrations were adjusted to 0.25 mg ml–1" (which would give 6.25 µg in 25 µl), and nothing states that the in vivo batch was formulated differently.
- p. 14 Methods, Animal experiments
25 µl of the LNP formulation (containing 10 µg total RNA)
- p. 13 Methods, LNP production
concentrations were adjusted to 0.25 mg ml–1
Only the authors can say what is right here.
Judge's note: Read crop C021-b.png (page 13, LNP production and RNA delivery: 0.2 mg of formulated payload, concentrations adjusted to 0.25 mg ml−1) and the full page image main-p014.png (Animal experiments: '25 µl of the LNP formulation (containing 10 µg total RNA) or 25 µl of saline solution'). I recomputed: 0.25 mg ml−1 × 25 µl = 6.25 µg, not 10 µg. I searched the LNP, animal and in vitro delivery passages for a second concentration or a statement of concentrating the in vivo material and found none; the LNP section describes only the in vitro application (40 ng per well). A separately formulated in vivo batch remains possible, so this is an unresolved ambiguity for the authors rather than a demonstrable wrong number.
Second check: another model agreed
Report a problem with this finding
Typo Misspelling in the Reporting Summary, Software and code, Data collection item 5: "RNA-seq was perfomed using Hiseq 4000 (Illumina) or NovaSeq X series (Illumina)."
- p. 1 Reporting Summary, Software and code, Data collection item 5
RNA-seq was perfomed using Hiseq 4000 (Illumina) or NovaSeq X series (Illumina).
- p. 26 Reporting Summary bound into the article, Data collection item 5
RNA-seq was perfomed using Hiseq 4000
- p. 1
RNA-seq was perfomed using Hiseq 4000 (Illumina)
Judge's note: These pages carry no text layer, so I read the rendered page images rather than the OCR: supp-01-p001.png and main-p026.png both show item 5 with 'perfomed' (missing r), which I transcribed from the images before consulting the OCR text. Items 1, 2 and 6 in the same box print 'performed' correctly, confirming it is a one-off typographical slip and not a rendering artefact. Cosmetic only.
Second check: none (judge only)
Report a problem with this finding
Typo Fig. 4h legend reads 'The red dashes represent the potential indels produced from FnCas12a2 editing.' one sentence after 'Enriching indels in the heterologous GFP gene created by FnCas12a.'; the Extended Data Fig. 8 legend likewise reads 'non-target GeCas12a RNPs'.
- p. 6 Fig. 4 legend, panel h
The red dashes represent the potential indels produced from FnCas12a2 editing.
- p. 6 Fig. 4 legend, panel h (preceding sentence)
Enriching indels in the heterologous GFP gene created by FnCas12a.
- p. 6 Fig. 4 legend, h
potential indels produced from FnCas12a2 editing
- p. 24 Extended Data Fig. 8 legend
non-target GeCas12a RNPs
Judge's note: Read crops C007-a, C007-b, C007-c (Fig. 4 legend, page 6) and C007-d plus the full page main-p024.png (Extended Data Fig. 8 legend). The Fig. 4 legend text 'potential indels produced from FnCas12a2 editing' is on the page as printed, in the same sentence pair as 'created by FnCas12a' and 'The same gRNA was used for both FnCas12a and GeCas12a2'. In the Extended Data Fig. 8 legend, (b) reads 'on-target FnCas12a and non-target GeCas12a RNPs' while (c) reads 'on-target FnCas12a and on-target GeCas12a2 RNPs', so the '2' is dropped in (b). No FnCas12a2 orthologue appears anywhere in the article; the enzymes used are FnCas12a, GeCas12a2 and SuCas12a2. Cosmetic naming slips with an evident fix.
Second check: none (judge only)
Report a problem with this finding
Typo Methods state 'Branch confidence was reported as ultrafast bootstrap values (ranging from 0 to 100).' while the Extended Data Fig. 2 legend states 'Branch confidence is shown on a 1-100 scale as ultrafast bootstrap values (UFBoot)76.'
- p. 11 Methods, Phylogenetic analysis
Branch confidence was reported as ultrafast bootstrap values (ranging from 0 to 100).
- p. 18 Extended Data Fig. 2 legend, panel a
Branch confidence is shown on a 1-100 scale as ultrafast bootstrap values (UFBoot)
Judge's note: Read crop C015-a (page 11, Methods 'Phylogenetic analysis') and crop C015-b (page 18, Extended Data Fig. 2 legend, panel a). Both are printed as quoted, so the same UFBoot support measure for the same dendrogram is given two different ranges, 0–100 and 1–100. This is a trivial inconsistency in describing the scale of a support statistic, with no effect on any reported value; recorded as a cosmetic consistency point.
Second check: none (judge only)
Report a problem with this finding
Typo Redundant repeated sentence in the Animal experiments Methods: "On the day of inoculation, tissue of passage 7 was collected. For transplantation of HN11303 fragments, tissue of passage 7 was collected and sliced into 3 × 3 to 4 × 4 mm3 pieces."
- p. 14 Methods, Animal experiments
On the day of inoculation, tissue of passage 7 was collected. For transplantation of HN11303 fragments, tissue of passage 7 was collected and sliced into 3 × 3 to 4 × 4 mm3 pieces.
Judge's note: Read crop C020-a.png and the full page image main-p014.png; the two consecutive sentences appear exactly as quoted, both stating that passage-7 HN11303 tissue was collected, the second adding the slicing step. The first sentence carries no information not in the second. Purely cosmetic duplication.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 76 spells the first inventor's surname "Scholtz, P., Zurek, C. & Krohn, M.", whereas the same person is the article's corresponding author "Paul Scholz" and is identified as an inventor of the same patent in Competing interests.
- p. 15 Reference 76
Scholtz, P., Zurek, C. & Krohn, M.
Judge's note: Read crops C024-a.png and C023-a.png (page 15, reference 76 reads 'Scholtz, P.'), C025-b.png (page 1 byline, 'Paul Scholz1,10'), and C025-a.png (page 16 Competing interests: 'BRAIN Biotech AG has filed patent applications that cover the GeCas12a2 nuclease sequence (WO 2022/017633) ... with P.S., C.Z. and M.K. as inventors'; the same page gives 'Correspondence ... should be addressed to Paul Scholz'). The patent number, the initials and the co-inventors Zurek and Krohn tie the reference entry to the article's own authors, so the 't' in 'Scholtz' is an internal inconsistency with an evident fix. Judged as a distinct defect from the UFBoot citation handled under C023.
Second check: none (judge only)
Report a problem with this finding
Typo The Extended Data Fig. 8 legend drops the '2' from the nuclease name in panel b: "following electroporation with (b) on-target FnCas12a and non-target GeCas12a RNPs or (c) on-target FnCas12a and on-target GeCas12a2 RNPs".
- p. 24 Extended Data Fig. 8 legend, panels b,c
following electroporation with (b) on-target FnCas12a and non-target GeCas12a RNPs
- p. 24 Extended Data Fig. 8 legend, panel c
(c) on-target FnCas12a and on-target GeCas12a2 RNPs
- p. 24 Extended Data Fig. 8 legend, b,c
on-target FnCas12a and non-target GeCas12a RNPs
Judge's note: Read crops C027-a.png, C027-b.png and C027-c.png and the full page image main-p024.png. The legend reads 'non-target GeCas12a RNPs' for panel b and 'on-target GeCas12a2 RNPs' for panel c in the same sentence; the panel c plot title in the figure itself reads 'FnCas12a g-GFP + GeCas12a2 g-GFP (enriched)'. No enzyme named GeCas12a (without the 2) appears anywhere else in the article, and the only Cas12a orthologue used, FnCas12a, is named separately in the same clause. Evident typographical fix.
Second check: none (judge only)