Initial Sequencing and Analysis of the Human Genome

• p. 7 Table 1, row 'Total all libraries', BAC-end sequence column

  100,775

• p. 7 Table 1, row 'Total of top eight libraries', BAC-end sequence column

  297,997

Only the authors can say what is right here.

Judge's note: Read the column on the p7 image. The four library rows sum to the printed top-eight subtotal for all three components (805,320; 484,278; 297,997). The all-libraries row includes those eight, so its third component cannot be below 297,997; the first two components do rise (812,594 and 488,017). The footnote defines the three numbers the same way for every row and gives no different basis for the last row. The printed 100,775 is wrong; the correct value cannot be derived from the paper (300,775 is a guess), so the authors must supply it.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 9 Table 3, Total row, total human sequence (kb)

  4,338,224

Only the authors can say what is right here.

Judge's note: Read every row from a 4x render of Table 3 and checked against the layout text; all 21 values match and sum to 4,338,498. The table note explains only why the total differs from Tables 1 and 2 (padding characters, clones not used in assembly), not a row-to-total difference inside Table 3. Values are integers in kb, so rounding cannot account for it. The gap is tiny (0.006%) and it is not evident whether a row or the total is wrong.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 9 Table 3, Total row, finished human sequence (kb)

  842,027

• p. 9 Table 3, finished column rows

  284,353

Only the authors can say what is right here.

Judge's note: Read every row from a 4x render of Table 3 (46,560; 284,353; 175,279; ... 35,911) and checked against the layout text; the sum is 844,527. The table note addresses only differences from Tables 1 and 2. The text figure of 'more than 835 Mb' of finished sequence is a different (nonredundant, 7 October) quantity and does not explain the within-table sum. An exact 2,500 difference suggests a single wrong digit in one row or in the total, but which cell is wrong cannot be told from the PDF.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 Table 7, chromosome 17, Sequence contigs: Number

  4,339

• p. 14 Table 8, chromosome 17, Sequence gaps: Number

  4,261

• p. 14 Table 7, All, Sequence contigs: Number

  149,821

Judge's note: Read all rows of Tables 6, 7 and 8 from 3x renders and the layout text. The sequence-contig column (chr 1-22, X, Y, UL) sums to 149,761, 60 below the All row, whereas the initial-sequence-contig column sums exactly to 396,913. For every chromosome except 6 and 17, Table 8 sequence gaps = Table 7 sequence contigs minus Table 6 sequenced-clone contigs. For 17: 4,339 - 138 = 4,201, but 4,261 is printed; 4,261 + 138 = 4,399, and replacing 4,339 by 4,399 makes the column sum exactly 149,821. Two independent checks point to the same digit transposition (4,399 printed as 4,339).

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 Table 7, All, Sequence-contig scaffolds: Number

  87,757

• p. 14 Table 7, chromosome 10 row, scaffold Number

  6,886

Only the authors can say what is right here.

Judge's note: Read the scaffold Number column from a 3x render of Table 7 and the layout text; the 25 rows sum to 85,747. The same rows sum exactly to the All value in the initial-sequence-contig column (396,913) and, after the chromosome 17 fix, in the sequence-contig column, so an unlisted category is not the explanation and the legend offers none. Chromosome 10 ("6,886" scaffolds for 9,126 contigs) is out of line with the other chromosomes, but changing it to 4,886 still leaves the sum 10 short (87,747), so the wrong cell cannot be identified from the PDF.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 Table 8, chromosome 6, Sequence gaps: Number

  5,225

• p. 14 Table 7, chromosome 6, Sequence contigs: Number

  5,547

Judge's note: Read Tables 6, 7 and 8 from 3x renders. Sequence gaps equal sequence contigs minus sequenced-clone contigs for every chromosome except 6 and 17 (17 is the Table 7 error in C023). For chromosome 6: 5,547 - 292 = 5,255, not 5,225. The printed rows (chr 1-22, X, Y, NA, UL) sum to 145,484; with 5,255 they sum to the printed 145,514, which is also the figure quoted in the text on page 15 ('145,514 gaps'). Evident digit slip: 5,255 printed as 5,225.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 Table 8, All row, Total estimated chromosome size

  3,289

• p. 14 Table 8 footnote (double dagger)

  The sum of the five lengths in the preceding columns.

• p. 14 Table 8 footnote ††

  The sum of the five lengths in the preceding columns

Only the authors can say what is right here.

Judge's note: Read the All row and all chromosome rows from a 3x render of Table 8. Every chromosome row reconciles with the five-column sum to within rounding (e.g. chr 1: 278.5 vs 279), but the All row gives 3,280.2 against 3,289. The chromosome totals themselves sum to 3,254; adding the NA and UL sequenced bases (5.1 + 9.3) and their gap bases brings this to about 3,280, agreeing with the five-column sum and not with 3,289. No footnote explains an extra 9 Mb. The text conclusion ('about 3,200 Mb' after removing about 100 Mb) is not materially affected. Which figure the authors intended cannot be settled from the PDF.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 24 Comparison with other organisms, point (3)

  (25%, 49% and 87%, respectively)

• p. 23 Table 12, Worm, DNA and Total

  5.30%

• p. 23 Table 12, Worm column

  6.50%

Only the authors can say what is right here.

Judge's note: Read the sentence on PDF page 24 and Table 12 on a zoomed crop of PDF page 23. The same sentence's human 6% (2.8/44.4) and mustard weed 49% (5.1/10.5) reproduce from the Total row, and the neighbouring 75% and 5–25% LINE/SINE figures do too, so the Total row is the denominator. Worm gives 81.5%; rounding of the table entries allows at most 5.35/6.45 = 83%, and the sum of the three listed classes as denominator gives 93%. No denominator in the table gives 87%. Fly is 22.6% against 25%, which is just reachable within rounding (0.75/3.05 = 24.6%). Either the 87% or the worm column is wrong; only the authors can say which.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 28 Fast living on chromosome Y

  1.57 corresponding to

• p. 28 Left column, paragraph 3

  = 1.57 corresponding to

• p. 28 Left column, paragraph 2 (formula)

  + 2

• p. 28 Fast living on chromosome Y

  mY:mX = 1.57 corresponding to am:af = 2.1

Only the authors can say what is right here.

Judge's note: Read the formula, the Page et al. conversion and the sentence on PDF page 28 (the image confirms the '+ 2αf' and '/3'). Solving for the ratio gives α = 2r/(3 − r). For r = 1.57 this is 2.196; any slope that rounds to 1.57 (1.565 to 1.575) gives 2.18 to 2.21, so rounding cannot produce 2.1. A ratio of 2.1 corresponds to a slope of 1.537. The same formula reproduces the quoted Page et al. pair (1.36 gives 1.66, printed 1.7), so the formula is being applied as written. Either the slope or the ratio is slightly off; the conclusion (about twofold, in reasonable agreement with Page et al.) is not affected.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 30 Table 14, repeat unit length 2, average number of SSR elements per Mb

  43.1

• p. 30 Table 15, AC

  27.7

Only the authors can say what is right here.

Judge's note: Read both tables on the page-30 image. Dinucleotide rows of Table 15 sum to 55.4, not 43.1, whereas the ten trinucleotide rows sum to 11.8, exactly the Table 14 value for length 3. The text shares (AC 50%, AT 35%, AG 15%) fit the 55.4 total. Table 15's footnote says the same identification method was used, and I found no statement of a different threshold or counting rule for dinucleotides. Which table is wrong cannot be told from the paper.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 39 Alternative splicing paragraph

  transcripts, corresponding to 544 genes (average 3.2 distinct tran-

• p. 39 Alternative splicing paragraph

  1,859 transcripts, corresponding to 544 genes (average 3.2 distinct transcripts per gene)

• p. 39 Alternative splicing paragraph

  1,859 transcripts, corresponding to 544 genes (average 3.2 distinct tran

• p. 39

  corresponding to 544 genes (average 3.2 distinct tran-

Only the authors can say what is right here.

Judge's note: Numbers read on the page-39 image. 1,859/544 = 3.42. The chromosome 22 figures in the same paragraph divide correctly (642/245 = 2.6; 145/245 = 59%). No exclusion or different denominator is described for chromosome 19, so one of the three numbers is wrong; which one only the authors can say.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 41 Table 22, Ensembl row, Matches to nonhuman proteins

  81,126 (63%)

• p. 41 Table 22, Total row, 'Matches to RIKEN mouse cDNA set'

  219,873 (69%)

Judge's note: Checked on the page-41 image and a 4x render of the table: the cell is printed "81,126 (63%)" with no footnote marker. The column total minus the other two rows gives 8,116, and 8,116/12,839 = 63.2%, matching the printed percentage. The similar fault in the Total row of the RIKEN column is a separate cell, handled in C058.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 41 Table 22, Total row, Matches to RIKEN mouse cDNA set

  219,873 (69%)

Judge's note: Checked on the page-41 image and a 4x render: the cell is printed "219,873 (69%)". The rows above (11,599, 3,016, 7,372) sum to 21,987, and 21,987/31,778 = 69.2%, so the printed figure is a garbled 21,987. A different cell from C057, so not a duplicate.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 41 Comparison with RIKEN mouse cDNAs

  69% (22,013/31,898) of the IGI matches the RIKEN

• p. 41 Table 22, Total row, Number

  31,778

• p. 41 Comparison with RIKEN mouse cDNAs, left column

  Conversely, 69% (22,013/31,898) of the IGI matches the RIKEN

• p. 41 Comparison with RIKEN mouse cDNAs

  69% (22,013/31,898) of the IGI matches the RIKEN cDNA set

Only the authors can say what is right here.

Judge's note: Read the sentence and Table 22 on the page-41 image. The denominator 31,898 differs from the 31,778 in Table 22 (rows sum to 31,778) and in the Table 23 footnote; the numerator 22,013 differs from the sum of the table's RIKEN column, 21,987. Both pairs give 69%. The text presents Table 22 as the breakdown of these same matches and mentions no other version of the set, so the mismatch is unexplained.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 43 Table 23, Number of distinct domain architectures, Human

  1,695

• p. 45 New architectures from old domains

  1.8 times

• p. 46 Figure 40, Human bar

  1,800

• p. 45 New architectures from old domains

  1.8 times as many protein architectures as worm or ¯y and 5.8 times as many as yeast

Only the authors can say what is right here.

Judge's note: Checked Table 23 on the p43 image (1,695 / 1,036 / 1,018 / 310), the text on p45 and the Fig. 40 bars on p46. The fly (about 1,040), worm (about 1,020) and yeast (about 310) bars agree with the table; only the human bar (about 1,810, read from the axis) does not. The text ratios match the figure, not the table. Both the Table 23 footnote and the Fig. 40 legend say SMART was used; the legend's note about collapsing tandem repeats is not said to differ from the table, and no other explanation was found. Which human value is right only the authors can say.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 43 Table 23, Percentage of 1-1-1-1, Human

  1.40

• p. 44 Genes shared with fly, worm and yeast

  In 564 of the 1,308 groups

• p. 44 Genes shared with fly, worm and yeast

  In 564 of the 1,308 groups, one orthologue

• p. 43 Table 23 footnote

  Table is based on 31,778 known genes and gene predictions.

Only the authors can say what is right here.

Judge's note: Read Table 23 and its footnote on the p43 image and the 564 figure on p44. The footnote defines the fraction as the percentage of the genome in groups with one member in each of the four species, so each species contributes 564 genes. 564 over the gene count reproduces the three other columns to the printed precision but gives 1.77% for human (1.76% with 32,000). 1.40% would need about 40,300 genes; no such human denominator is stated for this table. Either the percentage or the denominator is wrong; only the authors can say which.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 62 Present addresses

  63039 Clermont-Ferrand Cedex 2, France (L.C.)

• p. 2 Genome Analysis Group author list

  Lorenzo Cerutti28

• p. 2 Genome Analysis Group list

  Lorenzo Cerutti

• p. 62 Present addresses footnote

  (L.C.)

Judge's note: Viewed the page 2 image: "Glen A. Evans21†" has the dagger, and "Lorenzo Cerutti28" in the Genome Analysis Group has only the affiliation number. The page 62 image shows the dagger footnote giving addresses for (G.A.E.) and (L.C.). Cerutti is the only listed author with the initials L.C. (Lisa L. Cook has a middle initial), and page 61 separately credits fellowships to L.C. in the analysis group. The footnote mark is missing from his name.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 3 Background, item (4)

  Schimmel12 and Schimmel and Sutcliffe13

• p. 56 Reference 13

  Milner, R. J. & Sutcliffe, J. G. Gene expression in rat brain.

• p. 3 Background, item (4)

  Schimmel

Judge's note: Read the sentence on the p3 image and references 12 and 13 on the p56 image. Ref 12 is Putney, Herlihy & Schimmel (Nature 302, 1983), which fits "Schimmel". Ref 13 is Milner & Sutcliffe (Nucleic Acids Res. 11, 1983), with no Schimmel. The in-text name and the reference list disagree; "Milner and Sutcliffe" appears to be meant.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 7 Table 1 column header

  (ends/clones/

• p. 7 Table 1 footnote (double dagger)

  (clones/ends/clones with both ends sequenced)

• p. 7 Table 1 column header

  (ends/clones/clones with both ends sequenced)

Judge's note: Both strings read on the p7 image. The data fit the header order: the first number is always the largest (for example 21,956/14,445/7,255 and 403,589/226,068/156,631), and ends must outnumber clones. The footnote has the first two items transposed.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 20 Comparison of genetic and physical distance, first paragraph on page

  effects have been observed for chromosome 21 (refs 110, 134)

• p. 57 Reference 110

  RefSeq and LocusLink: NCBI gene-centered resources

• p. 20 Left column, first paragraph

  effects have been observed for chromosome 21 (refs 110, 134).

• p. 57 Reference 110

  RefSeq and LocusLink: NCBI gene-centered resources.

Only the authors can say what is right here.

Judge's note: Page 20 image: the sentence cites '(refs 110, 134)'. Page 57 image, reference list: 110 is the RefSeq/LocusLink paper; 134 is 'Lynn, A. et al. Patterns of meiotic recombination on the long arm of human chromosome 21', which fits. Reference 110 is cited appropriately on page 16 for 'the RefSeq dataset', so its use here is a wrong number. The intended reference is not determinable from the PDF (candidates in the list include 102, Broman et al., sex-specific variation in recombination, or 93, the chromosome 21 sequence paper), so the authors must supply it.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 23 Figure 19 legend

  Median ages and per cent of the genome covered by subfamilies of DNA

• p. 23 Figure 19 axis labels

  Average nucleotide substitution level for family

• p. 23 Age distribution discussion

  Figure 19 shows the mean ages of various

• p. 23 Figure 19 legend

  Median ages and per cent of the genome covered by subfamilies

Only the authors can say what is right here.

Judge's note: Read the legend, both axis labels and the body text on PDF page 23 and in the crop. The y axis is in kb (0–8,000), not per cent of the genome. The statistic is called median (legend, and the Fig. 24 legend 'median substitution level (see Fig. 19)'), average (x axis) and mean (text). The legend's later 'the average or median divergence' does not say which was plotted. Only the authors can say which statistic was used; the kb vs per cent label mismatch is plain.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 29 Table 13, row 8 (Tc1/Pogo/Pogo, no gene name)

  PID 10047247

• p. 29 Table 13, GenBank ID column

  pID 4263748

• p. 29 Table 13, first row

  nID 3150436

• p. 29 Table 13, Buster2 row and a later hAT/Charlie row

  PID 7243087

Only the authors can say what is right here.

Judge's note: Read Table 13 on a zoomed crop of PDF page 29 and the layout text. The two rows carry the same number but different families, and neither is in an italic related group, so one ID is presumably wrong; which one only the authors can say. The prefix is 'pID' in the first rows and from hAT/Tip100 onward, and 'PID' in between ('nID' in the first row is a nucleotide ID and is not an inconsistency). The second duplicated ID raised in this cluster, 7243087, is handled under C044; note it is on the Buster2 and KIAA1353 rows, not on an unnamed row as one reviewer note said.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 30 Segmental duplications, intrachromosomal duplications

  Marie–Tooth syndrome 1A, respectively34,234

• p. 56 Reference 34

  The human growth hormone locus: nucleotide sequence, biology, and evolution.

• p. 30 Intrachromosomal duplications, chromosome 17 examples

  Tooth syndrome 1A, respectively

Only the authors can say what is right here.

Judge's note: Read the sentence on PDF page 30 and reference 34 on PDF page 56 and reference 234 on PDF page 58. Reference 234 (Reiter et al., COX10 and the 24 kb CMT1A-REPs) fits the second example. Reference 34 is the growth hormone locus sequence and is used earlier in the paper for early targeted sequencing; its title has nothing to do with the roughly 200-kb chromosome 17 repeats or Smith–Magenis syndrome. The reference list contains no entry mentioning Smith–Magenis (searched the text), so the intended reference is not identifiable from the paper; a different paper by an author named Chen is a likely guess but rests on outside knowledge.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 33 Figure 33, y-axis label of all four panels

  Sum of aligned bases (kb)

• p. 33 Figure 33 legend

  number of aligned bp, for the subset of finished genome sequence

Judge's note: Viewed the crop and the full page. The axis labels say (kb) in all four panels and the legend says bp. In kb, single bins would hold 1 to 3.7 Gb of aligned sequence, far more than the roughly 835 Mb of finished sequence analysed (of which 3.3% is duplicated, Table 16), so the values can only be bp. The axis unit is wrong; the data are unaffected.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 38 Figure 36 legend, panel c

  Dependence of mean exon and intron lengths on GC content

• p. 38 Figure 36c axis labels

  Median intron length

• p. 38 Figure 36 legend (c)

  Dependence of mean exon and intron lengths on GC content.

Only the authors can say what is right here.

Judge's note: Viewed the crop and the full page 38. The legend and the axis labels name different statistics. The plotted exon values (about 119 to 134 bp) sit near the median internal exon of Table 21 (122 bp) rather than the mean (145 bp), which suggests the legend is the wrong one, but the intron curve (about 200 to 2,270 bp) does not settle it, so the authors should confirm which statistic was plotted.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 44 Table 24, epoxide hydrolase row

  BAA1143.21

• p. 44 Table 24, 'Uncharacterized protein' and 'β-Lactamase superfamily hydrolase' rows

  BAA34458.1

Only the authors can say what is right here.

Judge's note: The p44 image shows "BAA1143.21" exactly as quoted, so it is not an extraction artefact. The malformed pattern is evident from the table itself. The likely intended form (BAA11432.1, point misplaced) is a guess that cannot be checked from the files, so this goes to the authors. The second point in one reviewer note (BAA34458.1 listed twice) is handled in C064.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 44 Table 24, 'Uncharacterized protein' row (Gamma-proteobacteria, best hit Escherichia coli)

  BAA34458.1

Only the authors can say what is right here.

Judge's note: Both rows are visible on the p44 image, two rows apart, with different predicted function, bacterial range and best hit. One accession cannot be two different proteins with different best hits, so one of the two is miscopied. The table and text give nothing to say which one, or what the right accession is.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 47 Figure 42, labels above peregrin, Mi-2, Hrx and ALR

  F, H

• p. 47 Figure 42 legend

  (Y, yeast; W, worm; F, fly; V, vertebrate)

Judge's note: Viewed Fig. 42 and its legend on the p47 image. Panel a and the E(Pc)-like diagram use V; peregrin, Mi-2, Hrx and ALR use H. The legend has no entry for H. H presumably means human, but the figure mixes the two codes and leaves one undefined. The data are unaffected; the fix is to define H or use one code.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 48 Table 25, row ranked (25)

  IPR001664

• p. 49 Figure 44 legend

  (25) intermediate filament proteins [IPR001664]

• p. 48 Table 25 row IPR001664

  SH2 domain

• p. 49 Figure 44 legend

  (25) intermediate ®lament proteins [IPR001664]

Only the authors can say what is right here.

Judge's note: Read the Table 25 row on the p48 image and the Fig. 44 legend on p49. The same InterPro ID carries two different names in the paper, and Table 25 has no row for IPR000980 or for intermediate filaments. The text on p47 says the intermediate filament family has "127 family members", which matches neither name's count in the table (119). Either the ID or the name in the Table 25 row is wrong, and the table may be missing a row (see C069); the files do not settle which.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 57 Reference 109 (Riethman et al.)

  Nature 409, 953–958 (2001).

• p. 57 Reference 103 (The BAC Resource Consortium)

  human genome. Nature 409, 953–958 (2001).

• p. 57 Reference 109

  Riethman, H. C. et al. Integration of telomere sequences with the draft human genome sequence.

• p. 57 Ref. 109

  Integration of telomere sequences with the draft human genome sequence. Nature 409, 953–958 (2001).

Only the authors can say what is right here.

Judge's note: Read both entries on the page image of PDF p.57: authors and titles differ, volume and page range are identical. The companion papers from the same issue are listed on PDF p.56 and p.57 as ref. 90: 942–943, ref. 89: 943–945, ref. 91: 945–946, ref. 92: 947–948 and ref. 131: 951–953, which leaves 948–951 unassigned and suggests that ref. 109 is the wrong one. The internal contradiction is certain; which entry is wrong and the correct pages cannot be established from the files, so this goes to the authors.

Second check: another model agreed · checked by hand

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 60 Reference 380 (Kidd et al.)

  Am. J. Hum. Genet. 63, 1882–1899 (2000).

• p. 57 Reference 102 (Broman et al.)

  Am. J. Hum. Genet. 63, 861– 869 (1998).

Only the authors can say what is right here.

Judge's note: Read ref. 380 on the page image of PDF p.60 and ref. 102 on PDF p.57. Within the list, Am. J. Hum. Genet. volume 63 is 1998 (ref. 102), 64 is 1999 (ref. 237), 66 and 67 are 2000 (refs 240, 384, 386, 405, 406, 415) and 68 is 2001 (refs 381, 382). Volume 63 with year 2000 is therefore inconsistent; either the volume (probably 66) or the year is wrong. Which one cannot be settled from the files, so it is a query for the authors.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 21 Figure 17, Fraction of genome, LINEs

  21%

• p. 21 Table 11, LINEs, fraction of the draft genome sequence

  20.42

• p. 22 Census of human repeats

  13%, 20%, 8% and 3%

• p. 22 Census of human repeats

  13%, 20%, 8% and 3% of the sequence

Judge's note: Read Figure 17 in the crop and Table 11 on PDF page 21, and the census text on PDF page 22. The other three classes in Figure 17 (13%, 8%, 3%) are the Table 11 values rounded (13.14, 8.29, 2.84); LINEs at 20.42% round to 20%, not 21%. The copy numbers in Figure 17 are all loosely rounded (1,500,000 vs 1,558k; 450,000 vs 443k; 300,000 vs 294k), so the 850,000 vs 868k difference is in keeping with a schematic figure; only the 21% stands out. No explanation in the legend.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 33 Table 17 footnote

  the total amount of finished sequence then

• p. 33 Table 17 title and footnote

  Fraction of the draft genome sequence in inter- and intrachromo

• p. 33 Table 17 footnote

  total amount of ®nished sequence then

Judge's note: Read the Table 17 footnote on page 33 (zoomed render) and the Table 16 footnote on page 32. Table 16's footnote ends "reflects the duplications found within the total amount of finished sequence then", which is correct there. Table 17's footnote repeats the phrase although its title, the first half of the same sentence and the body text ("entire human draft genome sequence (finished and unfinished)") all say draft sequence. The phrase appears to have been carried over from Table 16.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send