Automated Prototyping of Genetic Codes
Nature, 2026 · paper · Biomedicine
Describes AGENTEX, a robotic cell-free platform for prototyping new genetic codes. It finds that most non-CCA-3′ tRNAs can be aminoacylated by natural E. coli enzymes, enabling compressed 34-codon codes and tests of two codes with reassignment of up to three codons.
Checked: PDF from nature.com, fetched 5 Oct 2026. Page numbers are those of that PDF.
10 findings
Substance Abstract states "we developed cell-free translation systems enabling compressed genetic codes of 34 aaRSs for 34 codons", although no set of 34 aminoacyl tRNA synthetases is described anywhere in the article (aaRS is defined in the same abstract as "aminoacyl tRNA synthetases"; 34 elsewhere always counts codons).
- p. 1 Abstract, final third
we developed cell-free translation systems enabling compressed genetic codes of 34 aaRSs for 34 codons
- p. 2 Fig. 1 caption, panel b
Alternatively, all 34 codons could be used for making peptides analogous to non-ribosomal peptide synthase
- p. 10 Discussion, paragraph beginning "On the basis of these discoveries"
specified by 22 tRNAs using AGENTEX (21 tRNAs with sense anticodons and one tRNA with a stop anticodon)
- p. 1 Abstract
compressed genetic codes of 34 aaRSs for 34 codons
Only the authors can say what is right here.
Judge's note: Read the abstract on the crop and on the full page 1 image: the phrase is printed as quoted. Searched the whole text layer for "34": it occurs only in the abstract ("34 aaRSs for 34 codons"), in the Fig. 1b caption ("all 34 codons could be used...", page 2 image), and in the Discussion (page 10 image: "a theoretical genetic code of 20 canonical amino acids and 14 open codons (34-codon genetic code)"), plus unrelated reference/number strings. The compressed codes actually built are specified by 22 tRNAs (Discussion, page 10) and 20 canonical aaRSs; Extended Data Fig. 1 (page 15) shows 20 canonical codons plus 14 open codons = 34 codons. No 34-aaRS quantity exists in the article, so the abstract gives a number without a referent; only the authors can say whether "tRNAs" or another term was intended.
Second check: another model agreed
Report a problem with this finding
Substance The Fig. 5 caption prints the reporter peptides as "MHHHHHHSKVPGAGVPGAGVPGAGKGGG (ELP-AAA) or MHHHHHHAKVPGSG VPGSGVPGSGKGGG (ELP-SSS)" (28 residues), whereas the construct bars above the chromatograms in the very panels described (Fig. 5c,e,g,i) read MHHHHHHSKGPGKVPGAGVPGAGVPGAGKGGG / MHHHHHHAKGPGKVPGSGVPGSGVPGSGKGGG (32 residues) — the caption omits the GPGK segment.
- p. 7 Fig. 5 caption, panels c–j
MHHHHHHSKVPGAGVPGAGVPGAGKGGG (ELP-AAA) or MHHHHHHAKVPGSG VPGSGVPGSGKGGG (ELP-SSS)
- p. 6 Fig. 5c, construct header bar above the EIC
MHHHHHHSKGPGKVPGAGVPGAGVPGAGKGGG
- p. 9 Results, ELP-SAS description
As ELP-SAS is 32 amino acids in length
Judge's note: Read the caption on the page-7 crop (C016-a) and the Fig. 5 panels on extract/png/main-p006.png: the orange/blue header bars over panels c and e read MHHHHHHSKGPGKVPGAGVPGAGVPGAGKGGG and over g and i MHHHHHHAKGPGKVPGSGVPGSGVPGSGKGGG. The same GPGK linker is visible in the Fig. 6d,g headers on extract/png/main-p008.png (MHHHHHHSKGPGKVPGSGVPGAGVPGSGKGGG), and the Results on page 9 (crop C016-c) state "As ELP-SAS is 32 amino acids in length", which is only reached with GPGK (9+4+16+3=32); without it the sequence is 28. The figure text is not in the text layer (grep for GPGK returns nothing), so this rests on the page images. Caption and panels describe the same experiment, so the two sequences are not legitimately different constructs.
Second check: another model agreed
Report a problem with this finding
Substance Methods, 'tRNA in vitro production': "After reverse transcription, polynucleotide kinase treatment was performed to dephosphorylate the tRNA 3′ terminus62" — the step sits in the T7 in vitro transcription/HDV ribozyme workflow, where no reverse transcription has occurred.
- p. 12 Methods, 'tRNA in vitro production'
After reverse transcription, polynucleotide kinase treatment was performed to dephosphorylate the tRNA 3′ terminus
- p. 12 Methods, "tRNA library preparation and sequencing"
Primer-dependent reverse transcription reaction was carried out using Maxima H Minus Reverse Transcriptase
Judge's note: Read crop C024-a and crop C024-b (page 12). The sentence is embedded between "IVT reactions were carried out at 37 °C ... and T7 RNA polymerase" and "tRNA produced by IVT was purified and size-selected using SPRIselect beads", i.e. entirely within the in vitro transcription protocol; the oligonucleotides are described in the same paragraph as carrying an HDV ribozyme. Reverse transcription is introduced only later, in 'tRNA library preparation and sequencing' ("Primer-dependent reverse transcription reaction was carried out using Maxima H Minus Reverse Transcriptase"), after the cell-free incubation, and the Fig. 2b workflow caption (page 3 text: "periodate oxidation, β-elimination (6) and dephosphorylation (7) ... followed by reverse transcription") places dephosphorylation before reverse transcription. As printed the methods state a step order the paper's own workflow contradicts; the evident fix is "After transcription".
Second check: another model agreed
Report a problem with this finding
Substance Extended Data Table 1 rows 1-2 print "VPGAGVPGAGVPGAGK 645.3582 2" and "VPGSGVPGSGVPGSGK 669.3506 2", values that equal (M+H)/2 rather than the doubly protonated (M+2H)/2 expected for the stated charge z = 2 (645.8618 and 669.8542), while other rows of the same table use (M+2H)/2.
- p. 24 Extended Data Table 1, row 1
VPGAGVPGAGVPGAGK 645.3582 2
- p. 24 Extended Data Table 1, row 2
VPGSGVPGSGVPGSGK 669.3506 2
Judge's note: Read extract/png/main-p024.png; the two rows and the z column are legible. Recomputed from monoisotopic residue masses (V 99.06841, P 97.05276, G 57.02146, A 71.03711, S 87.03203, K 128.09496) plus water 18.01056: VPGAGVPGAGVPGAGK M = 1289.70912, so (M+2H)/2 = 645.8618 but (M+H)/2 = 645.3582 — exactly the printed value; VPGSGVPGSGVPGSGK M = 1337.69388, (M+2H)/2 = 669.8542 but (M+H)/2 = 669.3506 — again exactly the printed value. Both rows are low by 0.5036 = half a proton. By contrast row 4 of the same table (GVPGAGVPGSGK, M = 981.5243) is printed as 491.7696, which is (M+2H)/2 = 491.7694, so the table is internally inconsistent. Average masses (646.25) are further away, and a 0.5 Da offset is far outside the accuracy of the instruments described in the Methods, so these are not plausibly observed values. I found no footnote or Methods statement defining a different m/z convention for the two asterisked rows (the asterisk only states 'Run on Q Exactive Orbitrap mass spectrometer').
Second check: another model agreed
Report a problem with this finding
Reference or label A PURE-specific comparison is pointed to the lysate panel: "Although CCA-tRNAs were generally more highly aminoacylated than CGA-tRNAs in PURE (Extended Data Fig. 4a)", whereas Extended Data Fig. 4's caption assigns (a) to "E. coli lysate-based cell free translation system" and (b) to "E. coli PURExpress reconstituted cell free translation system".
- p. 5 Results, 'Impact of tRNA modifications on charging'
Although CCA-tRNAs were generally more highly aminoacylated than CGA-tRNAs in PURE (Extended Data Fig. 4a)
- p. 18 Extended Data Fig. 4 caption
either incubated in (a) E. coli lysate-based cell free translation system or (b) E. coli PURExpress reconstituted cell free translation system
- p. 5 Impact of tRNA modifications, para 1
more highly aminoacylated than CGA-tRNAs in PURE (Extended Data Fig. 4a)
- p. 18 Extended Data Fig. 4 legend
(b) E. coli PURExpress reconstituted cell free translation system
Judge's note: Read the sentence on the page 5 image and the Extended Data Fig. 4 caption and panels on the page 18 image: panel a is the lysate CCA dataset (bars generally higher, consistent with the stated 67% lysate average) and panel b the PURE CCA dataset (consistent with 56%). The assignment is independently corroborated by the Extended Data Fig. 2 caption (lysate: "Note controls in Extended Data Fig. 4a", page 16 image) and the Extended Data Fig. 3 caption (PURExpress: "Note controls in Extended Data Fig. 4b", page 17 image). Panel 4a contains neither PURE data nor CGA data, so the pointer for this clause should be to Extended Data Fig. 4b (with Extended Data Fig. 3a); the following clause already cites 4b for PURE.
Second check: another model agreed
Report a problem with this finding
Reference or label The Results cite panels of a figure that has none: "the two terminal bases, 74 and 75, before the final adenosine was degenerate (Fig. 4a)" and "Many positions were found to be permissive to substitutions (Fig. 4b)", while Fig. 4 is a single unlettered heatmap array whose caption contains no panel letters.
- p. 7 Results, 'Landscape of acceptor stem charging'
the two terminal bases, 74 and 75, before the final adenosine was degenerate (Fig. 4a)
- p. 7 Results, 'Landscape of acceptor stem charging'
Many positions were found to be permissive to substitutions (Fig. 4b)
- p. 5 Fig. 4 (whole figure and caption)
Fig. 4 | E. coli tRNAs with full variation of the CCA 3′ end are aminoacylated by natural E. coli aaRSs.
- p. 7 Results, "Landscape of acceptor stem charging"
with the two terminal bases, 74 and 75, before the final adenosine was degenerate (Fig. 4a)
Judge's note: Looked at the whole of page 5 (extract/png/main-p005.png) and the caption crops C017-c/g/i: Fig. 4 consists of five unlettered columns of heatmaps plus a colour bar; no a/b (or any) panel letters appear anywhere in the figure, and the caption runs continuously ("A library of all E. coli isoacceptor tRNAs was synthesized ... All data represent n = 2 independent replicates") with no panel descriptions. The two in-text pointers are visible on the page-7 crops (C017-a/b/d/e/f/h and C018-a). Pointer error only; the underlying heatmap data are unaffected.
Second check: another model agreed
Report a problem with this finding
Reference or label Fig. 5 caption pairs each reporter with the other reporter's tryptic product: "we used dsDNA templates for peptide MHHHHHHSKVPGAGVPGAGVPGAGKGGG (ELP-AAA) or MHHHHHHAKVPGSGVPGSGVPGSGKGGG (ELP-SSS), in which the trypsin-digested product, VPGSGVPGSGVPGSGK or VPGAGVPGAGVPGAGK, respectively, can be robustly detected by LC–MS/MS".
- p. 7 Fig. 5 caption, panels c–j
in which the trypsin-digested product, VPGSG VPGSGVPGSGK or VPGAGVPGAGVPGAGK, respectively, can be robustly detected by LC–MS/MS
- p. 6 Fig. 5d, spectrum header and panel label
Reporter + CGA pool
- p. 6 Fig. 5c,d panel labels
V P G A G V P G A G V P G A G K
- p. 7 Fig. 5 legend
the trypsin-digested product, VPGSG VPGSGVPGSGK or VPGAGVPGAGVPGAGK, respectively
Judge's note: Read the Fig. 5 caption on the page 7 image (crops C015-a/d) and the Fig. 5 panels on the page 6 image. ELP-AAA is listed first and contains VPGAGVPGAGVPGAGK, ELP-SSS second containing VPGSGVPGSGVPGSGK, but the products are listed in the reverse order under "respectively". The panels confirm the true pairing: panels c,d (reporter MHHHHHHSKGPGKVPGAGVPGAGVPGAGKGGG, "Reporter + CGA pool") show and annotate VPGAGVPGAGVPGAGK, while panels g–j (swapped reporter MHHHHHHAKGPGKVPGSGVPGSGVPGSGKGGG) show VPGSGVPGSGVPGSGK. Trypsin cleaves C-terminal to K, so ELP-AAA yields VPGAGVPGAGVPGAGK. The data are correct; the caption's pairing is reversed.
Second check: another model agreed
Report a problem with this finding
Reference or label Extended Data Table 1 rows 4-7 name the crosstalk peptides with 11 residues ("GVPGAGVPGSK", "GVPGSGVPGAK", "GVPGSGVPGSK", "GVPGAGVPGAK"), while the text, Fig. 6 and the table's own m/z correspond to the 12-residue peptides (GVPGAGVPGSGK etc.); a glycine is missing from each name.
- p. 24 Extended Data Table 1, rows 4–7
GVPGAGVPGSK
- p. 9 Results, AGENTEX crosstalk paragraph
We monitored GVPGJGVPGJGK across all samples, where J = A or S.
- p. 8 Fig. 6e, annotated MS/MS sequence
GVPGAGVPGSGK
- p. 9 Results, crosstalk analysis
both GVPGAGVPGSGK (expanded genetic code) and GVPGSGVPGAGK (standard genetic code) were detected
Judge's note: Read extract/png/main-p024.png: the four rows print GVPGAGVPGSK 491.7696 z=2, GVPGSGVPGAK 491.7696 z=2, GVPGSGVPGSK 499.7610 z=2, GVPGAGVPGAK 483.7665 z=2. Read adjudicate/crops/C039-b.png, C039-d.png and C039-e.png (p. 9): 'We monitored GVPGJGVPGJGK across all samples, where J = A or S. The extracted ion chromatograms (EICs) for the expected masses of the peptide (Extended Data Table 1) were analysed', and 'both GVPGAGVPGSGK ... and GVPGSGVPGAGK ... were detected'; Fig. 6e on p. 8 annotates GVPGAGVPGSGK. I recomputed the 12-residue peptide from monoisotopic residue masses (G 57.02146, V 99.06841, P 97.05276, A 71.03711, S 87.03203, K 128.09496) plus water: M = 981.5243, [M+2H]2+ = 491.7694, matching the tabulated 491.7696; the printed 11-residue name (one G less) gives 463.2587, which appears nowhere in the table. So the masses are right and the sequence strings in the table are one residue short. I searched the Methods and the Extended Data Table footnotes ('*Run on Q Exactive Orbitrap mass spectrometer'; 'ND = Not detected') for any statement defining a shorter digestion product; there is none.
Second check: another model agreed
Report a problem with this finding
Wording Methods, 'aaRS protein production and purification': "The frozen cell pellet was thawed in lysis buffer (100 mM HEPES pH 7.2, 500 mM NaCl, 5 mM BME). The cell paste was suspended in 15 ml of lysis buffer (50 mM Tris (pH 7.5), 300 mM NaCl, 20 mM imidazole)" — two different compositions carry the same name, and the later steps "equilibrated with 20 ml lysis buffer" and "washed with 20 ml lysis buffer" do not say which is meant.
- p. 13 Methods, 'aaRS protein production and purification'
thawed in lysis buffer (100 mM HEPES pH 7.2, 500 mM NaCl, 5 mM BME). The cell paste was suspended in 15 ml of lysis buffer (50 mM Tris (pH 7.5), 300 mM NaCl, 20 mM imidazole)
- p. 13 Methods, "aaRS protein production and purification"
The frozen cell pellet was thawed in lysis buffer (100 mM HEPES pH 7.2, 500 mM NaCl, 5 mM BME). The cell paste was suspended in 15 ml of lysis buffer (50 mM Tris (pH 7.5), 300 mM NaCl, 20 mM imidazole)
- p. 13 aaRS protein production and purification
lysis buffer (100 mM HEPES pH 7.2, 500 mM NaCl, 5 mM BME)
- p. 13 same paragraph
15 ml of lysis buffer (50 mM Tris (pH 7.5), 300 mM NaCl, 20 mM imidazole)
Only the authors can say what is right here.
Judge's note: Read crops C028-a, C028-b, C028-c and the full page extract/png/main-p013.png (left column). Both definitions appear in consecutive sentences exactly as quoted, and the same paragraph then refers twice more to "lysis buffer" without qualification ("previously equilibrated with 20 ml lysis buffer", "The column was washed with 20 ml lysis buffer"). Since only the imidazole-containing buffer is compatible with Ni-NTA equilibration/wash, the ambiguity is real but affects reproducibility of the protocol description, not a reported result, hence S3; only the authors can say which buffer was used at each step.
Second check: another model agreed
Report a problem with this finding
Wording Extended Data Fig. 6 caption and Methods call the detected species "aminoacyl adenylates" / "amino acid adenylate", but panel a of the same figure labels it "AA-adenosine" and the masses plotted (431.1674, 454.1833) are those of the aminoacyl-adenosine, not of an aminoacyl adenylate.
- p. 20 Extended Data Fig. 6 caption and panel a label
The aminoacyl adenylates from exogenous tRNAs can be distinguished from native tRNAs by LC-MS by a 5 Da mass difference
- p. 20 Extended Data Fig. 6b, EIC labels
431.1674
- p. 13 Methods, 'Mass spectrometry analysis of tRNA aminoacylation'
All amino acid adenylate masses were monitored
Judge's note: Read extract/png/main-p020.png and adjudicate/crops/C036-a.png: panel a labels the RNase A product 'AA-adenosine' and 'AA-adenosine (heavy)', while the caption says 'The aminoacyl adenylates from exogenous tRNAs can be distinguished from native tRNAs by LC-MS by a 5 Da mass difference' and 'the aminoacyl adenylates of the exogenous tRNAs are expected to have a 5 Da greater mass'. Panel b prints native Tyr 431.1674 and native Trp 454.1833. I recomputed: Tyr esterified to adenosine, [M+H]+ = 267.09675 (adenosine) + 181.07389 (Tyr) - 18.01056 + 1.00728 = 431.16736, and Trp-adenosine = 267.09675 + 204.08988 - 18.01056 + 1.00728 = 454.18335 — both match the printed values to 0.0001. The corresponding aminoacyl adenylate (aa-AMP, adenosine 5'-monophosphate 347.06308) gives Tyr-AMP [M+H]+ = 511.1337, which appears nowhere in the figure. The printed exogenous/native gap 436.1525 - 431.1674 = 4.9851 is also consistent with five 15N atoms of the adenosine ring, as drawn in panel a. Methods p. 13 ('All amino acid adenylate masses were monitored', confirmed in adjudicate/crops/C036-c.png) and the Extended Data Fig. 5 caption use the same term. The term misdescribes the measured species and contradicts the figure's own label; the data themselves are unaffected.
Second check: another model agreed
Report a problem with this finding
24 minor (typos and the like; most checked by the judge only)
Reference or label The opening sentence "Synthetic genetic codes1,2,4,5 have enabled construction of organisms with new properties" includes reference 1, whose own title in the reference list is "Complete chemical synthesis, assembly, and cloning of a Mycoplasma genitalium genome" (Science 319, 1215–1220 (2008)), a genome-synthesis report rather than a synthetic genetic code.
- p. 1 Introduction, first sentence
Synthetic genetic codes1,2,4,5 have enabled construction of organisms with new properties
- p. 10 Reference list, entry 1
Gibson, D. G. et al. Complete chemical synthesis, assembly, and cloning of a Mycoplasma genitalium genome. Science 319, 1215–1220 (2008).
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Gibson et al. 2008 assembled a genome carrying the standard genetic code and did not produce a recoded organism. From the files alone I verified the citation group on the page 1 image crop and the reference entries on the page 10 image/text: refs 2 (Lajoie), 4 (Fredens) and 5 (Grome) are recoded-genome papers, while refs 3 (Ostrov, 57-codon genome) and 6 (Robertson, 57-codon code) are cited separately later for sense-codon reassignment. The title of ref 1 mentions no genetic-code change, so its placement in this group is at least mismatched; the authors may have intended it as general genome-synthesis context.
Second check: another model agreed
Report a problem with this finding
Reference or label Methods, 'Production of custom E. coli cell-free translation system': "DTT (3 ml) was added per ml of the sample immediately after sonication" — a volume three times that of the lysate, with no concentration given.
- p. 13 Methods, 'Production of custom E. coli cell-free translation system'
DTT (3 ml) was added per ml of the sample immediately after sonication.
- p. 13 Production of custom E. coli cell-free translation system
DTT (3 ml) was added per ml of the sample immediately after sonication
Only the authors can say what is right here.
Judge's note: Read the sentence on crop C028-a and on the full page extract/png/main-p013.png (right column): it follows "resuspended in 0.8 ml g−1 of pellet mass and sonicated ... about 600 J delivered per ml of sample" and precedes "The lysate was centrifuged at 18,000 × g ... the supernatant underwent a run-off reaction". As printed, 3 ml added per 1 ml of sample is a four-fold dilution of an extract that is then used directly as a translation system, and no DTT stock concentration is stated anywhere in the paragraph (the S30 wash buffer separately specifies 2 mM dithiothreitol). The volume/unit as printed is not usable as written; only the authors can say whether µl or a concentration was intended, so this is an author query rather than an evident fix.
Second check: another model agreed
Report a problem with this finding
Wording The same 14 open codons carry different numbers 1–14 in Fig. 1b (1 = UUG, 2 = CUC, 3 = AUA, 4 = GUC, 5 = UCG, 6 = CCC, 7 = ACG, 8 = GCC, 9 = UAG, 10 = UGA, 11 = CGG, 12 = AGC, 13 = AGG, 14 = GGG) and in Extended Data Fig. 1 (1 = GCC, 2 = CGG, 3 = AGG, 4 = GGG, 5 = ATA, 6 = TTG, 7 = CTC, 8 = CCC, 9 = TCG, 10 = AGT, 11 = ACG, 12 = GTC, 13 = TGA, 14 = TAG), although Fig. 1b states "Further details on codon choice and design of open codons are provided in Extended Data Fig. 1."
- p. 2 Fig. 1b, codon table and caption
14 codons available for reassignment, indicated in red
- p. 15 Extended Data Fig. 1, codon/anticodon boxes and caption
14 codons available for reassignment, indicated in red and numbered
- p. 2 Fig. 1b grid (red numbers) and caption
20 canonical amino acids can each be decoded by a single tRNA and 14 codons available for reassignment, indicated in red
- p. 1 Introduction, paragraph 2
ates 14 codons for non-standard building blocks (Fig. 1b and Extended Data Fig. 1)
Judge's note: Read the Fig. 1b codon table on the page 2 image (red numbers beside the red codons) and all 14 red numbers in Extended Data Fig. 1 on the page 15 image (ordered alphabetically by amino acid: Ala GCC = 1, Arg CGG = 2, Arg AGG = 3, Gly GGG = 4, Ile ATA = 5, Leu TTG = 6, Leu CTC = 7, Pro CCC = 8, Ser TCG = 9, Ser AGT = 10, Thr ACG = 11, Val GTC = 12, TGA = 13, TAG = 14). The two sets of codons are identical; only the indices differ, and Fig. 1b's caption cross-references Extended Data Fig. 1. I searched the text for references to a numbered open codon (grep "open codon"): the article never cites an open codon by number, so no statement in the text is affected; the numbering is left ambiguous rather than wrong.
Second check: another model agreed
Report a problem with this finding
Wording Discussion: "In contrast to previous work, with tSCAN we can rapidly test multiple exogenous tRNA designs using laboratory automation, with control over translation system composition, and determine the identity of amino acids charged to tRNA" attributes identification of the charged amino acid to tSCAN, which the paper elsewhere describes as a sequencing readout of charged versus uncharged tRNA, the amino-acid identification being the stated purpose of tSCAN-M.
- p. 9 Discussion, paragraph 1
In contrast to previous work, with tSCAN we can rapidly test multiple exogenous tRNA designs using laboratory automation, with control over translation system composition, and determine the identity of amino acids charged to tRNA.
- p. 3 Results, development of tSCAN
By quantifying ratios of intact to truncated reads for each species, we quantify aminoacylation.
- p. 6 Results, introduction of tSCAN-M
charging analysis by mass (tSCAN-M), which uses isotopically labelled tRNAs incubated in lysate and subsequent digestion, isolation and LC–MS detection of the charged amino acid species
Judge's note: Read crop C021-a (page 9 Discussion), C021-b (page 3: "By quantifying ratios of intact to truncated reads for each species, we quantify aminoacylation") and C021-c (page 6: tSCAN-M "uses isotopically labelled tRNAs ... and LC–MS detection of the charged amino acid species"). The sentence immediately preceding the flagged one names both methods together ("we developed new methods, tSCAN and tSCAN-M, to quantify aminoacylation ... and identify their esterified amino acids"), so "with tSCAN" can be read as shorthand for the platform; the following sentence then says "Using this technology". Either reading is defensible, so this is a wording ambiguity only, not a misattributed result.
Second check: another model agreed
Report a problem with this finding
Wording Position numbering of the tRNA 3′ end is inconsistent: the main text and Extended Data Fig. 7c,d call the terminal adenosine A75 ("π–π stacking between A75 of tRNA glyT and Y80 of GlyRS"; panel labels C73, C74, A75), while Fig. 4 numbers the two cytidines of the CCA end as bases 74 and 75 ("Bases 74 and 75 comprising the 3′ CCA sequence were degenerate in all tRNA isoacceptors"; rows = third-to-last base 74, columns = penultimate base 75).
- p. 5 Results, structural modelling sentence
including π–π stacking between A75 of tRNA glyT and Y80 of GlyRS
- p. 21 Extended Data Fig. 7c,d labels and caption
π-π stacking is present between A75 of WT tRNA glyT and Y80 of glyRS
- p. 5 Fig. 4 caption
Bases 74 and 75 comprising the 3′ CCA sequence were degenerate in all tRNA isoacceptors
Only the authors can say what is right here.
Judge's note: Read the Fig. 4 axes and caption on the page 5 image: rows are labelled "Base 74" (third-to-last position) and columns "Base 75" (penultimate position), so the terminal adenosine is position 76 in that scheme. On the page 21 image, Extended Data Fig. 7c,d label the glyT 3′ end C73, C74, A75, and panels a,b label the leuW end C84/G84, A85 — i.e. residue numbering taken straight from the structural models rather than the standard tRNA numbering. The main-text sentence adopts the model numbering ("A75") without saying so, so within the article position 75 denotes a cytidine in Fig. 4 and the terminal adenosine in Extended Data Fig. 7. The underlying structures and conclusions are unaffected; only the authors can say which numbering they intend in the text.
Second check: another model agreed
Report a problem with this finding
Wording The text lists "some tRNAs exhibited a generally larger flexibility to substitution, such as valT, valV, valU and pheV", but Fig. 4 contains a single combined row "valTUXYZ (UAC)" (so valT and valU are not separable rows) and no pheV row, only "pheUV (GAA)".
- p. 7 Results, 'Landscape of acceptor stem charging'
some tRNAs exhibited a generally larger flexibility to substitution, such as valT, valV, valU and pheV
- p. 5 Fig. 4, heatmap row labels
valTUXYZ (UAC)
Only the authors can say what is right here.
Judge's note: Read the sentence on crop C018-a and the Fig. 4 row labels on crop C018-b and the full page extract/png/main-p005.png: the valine rows are valTUXYZ (UAC), valV (GAC) and valW (GAC), and the phenylalanine rows are pheUV (GAA) and pheV is not present separately. The paper does elsewhere use single-gene shorthand for grouped rows (e.g. "only valT was aminoacylated" on page 5), so naming valT is defensible; naming valT and valU as two separate examples points at one and the same row, and pheV cannot be distinguished from pheU in the figure. An ambiguity of wording rather than a wrong datum, hence S3.
Second check: another model agreed
Report a problem with this finding
Wording The fold-changes quoted for the HiBiT signal — "approximately six-fold higher HiBiT production with MP2 and pazF (+pazF, +pazFRS) than WT lysate, and about 10-fold higher than the no-template control (Fig. 6c)" and, later, "about six-fold over background, P < 0.001" — are not readable from the bar heights plotted in Fig. 6c, where the CGA +pazF +pazFRS bar is about 95,000 a.u. against about 20,000 a.u. for the corresponding WT bar (about 4.7-fold) and about 29,000 a.u. for the CGA −pazFRS bar.
- p. 7 Compressed genetic codes with AGENTEX
approximately six-fold higher HiBiT production with MP2 and pazF (+pazF, +pazFRS) than WT lysate, and about 10-fold
- p. 9 p9 first paragraph
about six-fold over background, P < 0.001
- p. 7 Results, HiBiT assay
approximately six-fold higher HiBiT production
- p. 7 Compressed genetic codes with AGENTEX
approximately six-fold higher HiBiT production with MP2 and pazF (+pazF, +pazFRS) than WT lysate, and about 10-fold higher than the no-template control
Only the authors can say what is right here.
Judge's note: Read the two text passages on crops C019-a/c/d (page 7) and C019-b/f (page 9) and the Fig. 6c bar chart on crop C019-e and the full page extract/png/main-p008.png. Axis ticks are 0 / 50,000 / 100,000 / 150,000; by eye the CGA +pazF +pazFRS bar tops out just below 100,000 (points ~88,000–105,000), the WT +pazF +pazFRS bar is ~20,000, the CGA +pazF −pazFRS bar ~29,000 and the two no-template bars (h-TAG-1 "−") ~8,000–9,000. The ~10-fold statement against no template is consistent (~11). The six-fold statement is not reachable from the plotted means for either comparator unless the extreme ends of the ranges are used (100,000/17,000 ≈ 5.9), and the same "six-fold" is applied once to WT lysate and once to "background" (the −pazFRS condition), which are different bars; the page-9 sentence does state that the WT/−pazFRS signals are "at background levels", which partly reconciles the two baselines. Bar heights read by eye cannot settle the exact ratio, so flagged as an author query rather than a wrong number. The *** on the bar is consistent with the stated P < 0.001.
Second check: another model agreed
Report a problem with this finding
Typo The abbreviation otRNA is expanded two different ways: abstract "the extent to which non-CCA-3′ tRNAs (otRNAs)" (and Introduction "tRNAs carrying non-CCA 3′ ends (otRNAs)") versus Fig. 6 caption "an MP2 orthogonal tRNA (otRNA) pool".
- p. 1 Abstract
the extent to which non-CCA-3′ tRNAs (otRNAs) would be aminoacylated
- p. 2 Introduction, column 1
tRNAs carrying non-CCA 3′ ends (otRNAs)
- p. 9 Fig. 6 caption, panel a
ribosomes (G2251C/G2553C in 23S rRNA) and an MP2 orthogonal tRNA (otRNA) pool
Judge's note: Checked the abstract crop/page 1 image (definition as non-CCA-3′ tRNAs), page 2 image (Introduction, same expansion) and the Fig. 6 caption crop (page 9) where the abbreviation is re-introduced as "orthogonal tRNA (otRNA)". Both wordings refer to the same molecules, so nothing substantive is affected; it is a cosmetic inconsistency in how a defined abbreviation is expanded.
Second check: none (judge only)
Report a problem with this finding
Typo Author contributions (PDF p. 14) use the initials "H.-M.B." twice, while the byline (PDF p. 1) gives the author as "Hana M. Burgess" (i.e. H.M.B.).
- p. 14 Author contributions
F.R., B.B., N.S., H.-M.B., L.O. and G.M.C. devised the methodology.
- p. 1 Author list
Hana M. Burgess1,3
- p. 14 Author contributions
H.-M.B.
- p. 1 Author list
Hana M. Burgess
Judge's note: Read adjudicate/crops/C033-a.png and C033-c.png (Author contributions): "F.R., B.B., N.S., H.-M.B., L.O. and G.M.C. devised the methodology" and "F.R., B.B., N.S. and H.-M.B. performed the visualization". Read adjudicate/crops/C033-b.png and C033-d.png (byline): "Felix Radford, Nayan Sapers, Hana M. Burgess, Lucy Ort, Bogdan Budnik & George M. Church" — no hyphenated given name. The hyphen is not used for any other author's initials (F.R., B.B., N.S., L.O., G.M.C.). A typographic slip in initials; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Introduction prints "universality in tRNAs charged by both class I and Class II aaRSs", capitalising "Class II" inconsistently with "class I" and with the lower-case usage elsewhere.
- p. 2 Introduction, column 2, paragraph 2
universality in tRNAs charged by both class I and Class II aaRSs
- p. 5 Results, "Impact of tRNA modifications on charging"
tRNAs charged by class II aaRSs generally showed lower aminoacyla-
- p. 2 Introduction
both class I and Class II aaRSs
Judge's note: Verified on the page 2 image (crop C008-a at high magnification): the capital C in "Class II" is clearly set in the published text, so it is not an extraction artefact. Elsewhere the article uses lower case, e.g. page 5 "tRNAs charged by class II aaRSs generally showed lower aminoacylation than class I" and the Fig. 3 caption "class I (a,c) or class II (b,d) aaRSs". Cosmetic only.
Second check: none (judge only)
Report a problem with this finding
Typo Page 2 prints "the Opentrons OT2 (Fig. 2b, ...)" whereas the instrument is written "OT-2" everywhere else in the article.
- p. 2 tSCAN section
the Opentrons OT2
Judge's note: Read the page 2 image and crops C006-a/C006-c, which show "the Opentrons OT2". The Fig. 2 caption on page 3 (crop C011-a) prints "the Opentrons OT-2 robotic system", and the text layer shows "OT-2" at all other occurrences (page 7 "OT-2 robot", Methods "Opentrons OT-2 equipped with two Magnetic Modules", "Jupyter notebook of OT-2"). A single inconsistent spelling of the instrument name.
Second check: none (judge only)
Report a problem with this finding
Typo Extended Data Fig. 7 caption (PDF p. 21): "there is a steric hinderance between A85 and G84 with P105. This hinderance is not present in the WT leuW tRNA" — 'hinderance' twice, where the main text (p. 5) writes 'steric hindrance'.
- p. 21 Extended Data Fig. 7 caption
there is a steric hinderance between A85 and G84 with P105. This hinderance is not present
- p. 21 Extended Data Fig. 7 caption
there is a steric hinderance between A85 and G84 with P105. This hinderance is not present in the WT leuW tRNA
- p. 5 Results, "Impact of tRNA modifications on charging"
Structural modelling (AlphaFold3) suggested steric hindrance
- p. 21 Extended Data Fig. 7 legend
there is a steric hinderance between A85 and G84 with P105. This hinderance
Judge's note: Read adjudicate/crops/C037-a.png and C037-e.png together with the page-21 text layer (which is a real text layer, not OCR): lines 2698-2699 read 'there is a steric hinderance between A85 and G84 with P105. This hinderance is not present in the WT leuW tRNA where cytosine is at position 84.' The main text on p. 5 (adjudicate/crops/C037-c.png) reads 'Structural modelling (AlphaFold3) suggested steric hindrance differences between CCA and CGA ends'. Misspelling, twice; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Fig. 5 caption: "The elution times of the respective peptides is indicated with shading." — plural subject with singular verb.
- p. 7 Fig. 5 legend
The elution times of the respective peptides is
Judge's note: Read on crop C016-a (the Fig. 5 caption on page 7), right-hand column: the sentence reads exactly "The elution times of the respective peptides is indicated with shading." Cosmetic subject–verb disagreement; fix is "are indicated".
Second check: none (judge only)
Report a problem with this finding
Typo Missing spaces in two places: reference 54 prints "Front. Genet.5, 158 (2014)" (no space before the volume number) and the Extended Data Fig. 8 legend prints "stop codon(TAG) reassigned for nsAA incorporation".
- p. 11 Ref. 54
Front. Genet.5, 158
- p. 22 Extended Data Fig. 8 legend
stop codon(TAG)
Judge's note: Read crop C023-a (page 11 reference list) — ref. 54 reads "Front. Genet.5, 158 (2014)." while the neighbouring entries are correctly spaced ("Biomolecules 7, 35 (2017)", "Nat. Biotechnol. 19, 751–755 (2001)"), confirming the missing space is in the typesetting and not an extraction artefact. Read crop C023-b (page 22) — the Extended Data Fig. 8 legend reads "or stop codon(TAG) reassigned for nsAA incorporation". Both are cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, 'tRNA library preparation and sequencing': "Libraries were purified using SPRIselect and quantified using QuBit."
- p. 12 tRNA library preparation
quantified using QuBit
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the Thermo Fisher fluorometer is branded "Qubit", not "QuBit". I confirmed on crop C025-a (page 12, right column) that the page prints "quantified using QuBit"; the capitalisation of the brand itself is not verifiable from the files. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, 'Robotic procedure for tSCAN': "Three.xlsx spreadsheets were provided containing layouts of two source plates as well as reactions" — missing space before the file extension.
- p. 12 Robotic procedure for tSCAN
Three.xlsx spreadsheets
Judge's note: Read crop C026-a (page 12, right column): the line reads "Three.xlsx spreadsheets were provided". The parallel sentence in the automation section on page 14 (extract/main.layout.txt line 1282, same wording) is set correctly as "Three .xlsx spreadsheets ... A .csv file", whereas page 12 also prints "A CSV file"; so the missing space on page 12 is a typesetting slip, not an extraction artefact. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods: "1.5 U μl−1 RNAse A" (PDF p. 13) against "1.5 U μl−1 RNase A" in the next Methods section, and "reads were aligned with MMseq" (PDF p. 12) against "MMSeqs2" in the Reporting Summary.
- p. 13 tSCAN-M methods
1.5 U μl−1 RNAse A
- p. 12 Analysis of NGS data
aligned with MMseq
Judge's note: Read extract/png/main-p013.png: the tSCAN-M paragraph prints 'RNAse A' while the 'In vitro aminoacylation activity assay' paragraph on the same page prints 'RNase A' (text layer lines 2373 vs 2408; 'RNase A' also at 2371 and 2466, and 'RNAase A' in the Extended Data Fig. 6 caption). Read adjudicate/crops/C032-b.png: 'reads were aligned with MMseq to a library of all E. coli tRNAs' is legible, and extract/png/main-p025.png (Reporting Summary, Data analysis) lists 'GraphPad Prism 10, Python version 3.10, MMSeqs2'. Both are naming inconsistencies within the article; no numbers or claims are affected, so cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, 'In vitro aminoacylation activity assay': "Aminoacylation reactions were incubated at 1 h at 37 °C. The reactions were stopped by the adding 20 µl digestion solution".
- p. 13 Methods, 'In vitro aminoacylation activity assay'
Aminoacylation reactions were incubated at 1 h at 37 °C.
- p. 13 In vitro aminoacylation activity assay
were incubated at 1 h at 37 °C. The reactions were stopped by the adding
Judge's note: Read crop C029-a and the full page extract/png/main-p013.png (left column, 'In vitro aminoacylation activity assay'). The page prints "incubated at 1 h at 37 °C" (should be "for 1 h") and "stopped by the adding 20 µl digestion solution" (should be "by adding"). Both are visible in the image, so not extraction artefacts. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, 'Production of custom E. coli cell-free translation system': "At an OD600 of 0.6, cultures were inoculated with 1 mM IPTG and grown to an OD of 3.0" — IPTG induces, it is not an inoculum.
- p. 13 Methods, 'Production of custom E. coli cell-free translation system'
At an OD600 of 0.6, cultures were inoculated with 1 mM IPTG and grown to an OD of 3.0.
Judge's note: Read crop C030-a and the full page extract/png/main-p013.png: the right-hand column prints "cultures were inoculated with 1 mM IPTG", while the left-hand column of the same page uses the correct verb for the same operation ("Protein production was induced by adding 0.5 mM isopropyl-b-D-1-thiogalactopyranoside"). Wrong word, no effect on the reported quantities; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, 'Peptide production in cell-free translation systems' (PDF p. 13): "25 µl of the NEBxpress synthesis buffer" — the product name is spelled "NEBExpress" in the same paragraph and twice elsewhere in the Methods.
- p. 13 Peptide production
25 µl of the NEBxpress synthesis
Judge's note: Read extract/png/main-p013.png. The same paragraph begins "In vitro translation reactions were set up with NEBExpress Cell-free E. coli Protein Synthesis System (New England Biolabs)" and then writes "25 µl of the NEBxpress synthesis buffer"; the text layer shows the same two spellings (lines 2462 and 2465), and 'NEBExpress' also appears on p. 12 and p. 13. An internal capitalisation inconsistency in a product name; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Methods, proteomic analysis (PDF p. 14): "The Q Exactive instrument was run in WWA mode (wide window isolation)" — the parenthetical expansion does not spell out the abbreviation given.
- p. 14 Methods, 'Proteomic analysis of peptides produced in cell-free translation systems'
The Q Exactive instrument was run in WWA mode (wide window isolation)
Only the authors can say what is right here.
Judge's note: Read adjudicate/crops/C034-a.png and extract/png/main-p014.png; the sentence reads exactly as quoted, followed by 'with MS data-dependent acquisition'. The internal mismatch is verifiable on the page: 'wide window isolation' would give WWI, not WWA. (Identifying WWA as the vendor term 'wide window acquisition' rests on outside knowledge, so which half of the phrase is wrong is for the authors to say; the paper separately writes 'automatic window isolation' for the DIA method in the next paragraph.) Abbreviation/definition mismatch only; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Extended Data Fig. 5c panel headings read "CCA tRNA-SCAN-M" and "CGA tRNA-SCAN-M", whereas the method is called "tSCAN-M" in the same figure's caption and throughout the article.
- p. 19 Extended Data Fig. 5c, panel titles
CCA tRNA-SCAN-M
- p. 19 Extended Data Fig. 5 caption
(c) tSCAN-M was performed on tRNALeu, tRNAPro, and tRNAVal
Judge's note: Read extract/png/main-p019.png: the two panel titles above the panel-c chromatograms are legibly 'CCA tRNA-SCAN-M' and 'CGA tRNA-SCAN-M' (panel b titles are 'CCA in vitro' / 'CGA in vitro'), while the caption below states '(c) tSCAN-M was performed on tRNA-Leu, tRNA-Pro, and tRNA-Val...' and 'determination of the charging ... using tSCAN-M'. No definition of 'tRNA-SCAN-M' appears anywhere; the inconsistency is in the artwork only and the meaning is unambiguous, so cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Extended Data Fig. 9 caption (PDF p. 23): "if the ribosome has a complimentary mutation, it can accept that tRNA for translation (Fig. 6)" — 'complimentary' for 'complementary'.
- p. 23 Extended Data Fig. 9b caption
if the ribosome has a complimentary mutation, it can accept that tRNA for translation
- p. 23 Extended Data Fig. 9 caption, panel b
if the ribosome has a complimentary mutation, it can accept that tRNA for translation (Fig. 6)
- p. 23 Extended Data Fig. 9 legend
the ribosome has a complimentary mutation
- p. 23 Extended Data Fig. 9 legend, (b)
if the ribosome has a complimentary mutation
Judge's note: Read adjudicate/crops/C038-a.png (and extract/png/main-p023.png context); the sentence is legible as quoted and describes base-pairing between an altered tRNA 3' end and a mutated large ribosomal subunit, so the intended word is 'complementary'. The page-23 text layer agrees (line 2736). Single-word misspelling; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Reporting Summary, Replication (PDF p. 26): "The principle next generation sequencing experiments were repeated in duplicate" — 'principle' for 'principal'.
- p. 26 Replication
The principle next generation sequencing experiments
Judge's note: Read extract/png/main-p026.png (the 'Replication' box of the Life sciences study design section); the sentence is legible on the page image as 'Experiments were repeated in triplicate for all assays. The principle next generation sequencing experiments were repeated in duplicate, with > 1000000 measurements per experiment.' Although p. 26 is an OCR page, the word is readable in the image, so this is not an extraction artefact. Adjectival use requires 'principal'; cosmetic.
Second check: none (judge only)