CoChat
Sign Up
Cited by 57Algorithms and Data CompressionFractal and DNA sequence analysisAdvanced Data Compression Techniques

Efficient DNA sequence compression with neural networks

Open access
Silva, Milton, Pratas, Diogo, Pinho, Armando J·2020·GigaScience
DOI 10.1093/gigascience/giaa119PMC PMC7657843
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: The increasing production of genomic data has led to an intensified need for models that can cope efficiently with the lossless compression of DNA sequences. Important applications include long-term storage and compression-based data analysis. In the literature, only a few recent articles propose the use of neural networks for DNA sequence compression. However, they fall short when compared with specific DNA compression tools, such as GeCo2. This limitation is due to the absence of models specifically designed for DNA sequences. In this work, we combine the power of neural networks with specific DNA models. For this purpose, we created GeCo3, a new genomic sequence compressor that uses neural networks for mixing multiple context and substitution-tolerant context models. Findings: We benchmark GeCo3 as a reference-free DNA compressor in 5 datasets, including a balanced and comprehensive dataset of DNA sequences, the Y-chromosome and human mitogenome, 2 compilations of archaeal and virus genomes, 4 whole genomes, and 2 collections of FASTQ data of a human virome and ancient DNA. GeCo3 achieves a solid improvement in compression over the previous version (GeCo2) of $2.4\%$, $7.1\%$, $6.1\%$, $5.8\%$, and $6.0\%$, respectively. To test its performance as a reference-based DNA compressor, we benchmark GeCo3 in 4 datasets constituted by the pairwise compression of the chromosomes of the genomes of several primates. GeCo3 improves the compression in $12.4\%$, $11.7\%$, $10.8\%$, and $10.1\%$ over the state of the art. The cost of this compression improvement is some additional computational time (1.7–3 times slower than GeCo2). The RAM use is constant, and the tool scales efficiently, independently of the sequence size. Overall, these values outperform the state of the art. Conclusions: GeCo3 is a genomic sequence compressor with a neural network mixing approach that provides additional gains over top specific genomic compressors. The proposed mixing method is portable, requiring only the probabilities of the models as inputs, providing easy adaptation to other data compressors or compression-based data analysis tools. GeCo3 is released under GPLv3 and is available for free download at https://github.com/cobilab/geco3.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/gigascience/giaa119

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Heuristic methods, such as MUMmer and BLASTZ, generate approximate alignments rapidly but may not always produce the optimal alignment.

    “Heuristic methods, such as MUMmer and BLASTZ, generate approximate alignments rapidly but may not always produce the optimal alignment [57,58].”

    From Whole-Genome Alignment: Methods, Challenges, and Future Directions (2024) · cites this paper as [58]

    The Introduction discusses DNA sequence compression, and the Results benchmark compression tools rather than alignment methods. Searches of our copy for MUMmer, BLASTZ, heuristic alignment, and optimal alignment found no passage addressing whether those methods produce rapid but potentially non-optimal alignments. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 95

  1. [1]Stephens ZD, Lee SY, Faghri F, et al. Big data: astronomical or genomical?. PLoS Biol. 2015; 13(7): e1002195. 26151137 PMC4494865
  2. [2]Pratas D, Pinho AJ, Ferreira PJ. Efficient compression of genomic sequences. In: 2016 Data Compression Conference (DCC). IEEE; 2016: 231– 240.
  3. [3]Pratas D, Hosseini M, Pinho AJ. GeCo2: An optimized tool for lossless compression and analysis of DNA sequences. In: International Conference on Practical Applications of Computational Biology and Bioinformatics. Springer; 2019: 137– 145.
  4. [5]Pratas D, Hosseini M, Pinho AJ. Substitutional tolerant Markov models for relative compression of DNA sequences. In: International Conference on Practical Applications of Computational Biology and Bioinformatics. Springer; 2017: 265– 272.
  5. [6]Polikar R. Ensemble based systems in decision making. IEEE Circuits Syst Mag. 2006; 6(3): 21– 45.
  6. [7]Wolpert DH. Stacked generalization. Neural Netw. 1992; 5(2): 241– 59.
  7. [8]Khalid S, Khalil T, Nasreen S. A survey of feature selection and feature extraction techniques in machine learning. In: 2014 Science and Information Conference. IEEE; 2014: 372– 378.
  8. [9]Fritz MHY, Leinonen R, Cochrane G, et al. Efficient storage of high throughput DNA sequencing data using reference-based compression. Genome Res. 2011; 21(5): 734– 40. 21245279 PMC3083090
  9. [10]Giancarlo R, Scaturro D, Utro F. Textual data compression in computational biology: A synopsis. Bioinformatics. 2009; 25(13): 1575– 86. 19251772
  10. [11]Pratas D, Silva RM, Pinho AJ, et al. An alignment-free method to find and visualise rearrangements between pairs of DNA sequences. Sci Rep. 2015; 5(1): 10203. 25984837 PMC4434998
  11. [12]Pratas D, Pinho AJ. Metagenomic composition analysis of sedimentary ancient DNA from the Isle of Wight. In: 2018 26th European Signal Processing Conference (EUSIPCO). IEEE; 2018: 1177– 81.
  12. [13]Covas C, Caetano T, Cruz A, et al. Pedobacter lusitanus sp. nov., isolated from sludge of a deactivated uranium mine. Int J Syst Evol Microbiol. 2017; 67(5): 1339– 48. 28109203
  13. [14]Pyöriä L, Jokinen M, Toppinen M, et al. HERQ-9 is a new multiplex PCR for differentiation and quantification of all nine human herpesviruses. Msphere. 2020; 5(3): e00265– 20. 32581076 PMC7316487
  14. [15]Toppinen M, Pratas D, Väisänen E, et al. The landscape of persistent human DNA viruses in femoral bone. Forensic Sci Int Genet. 2020; 48: 102353. 32668397
  15. [16]Duggan AT, Perdomo MF, Piombino-Mascali D, et al. 17th century variola virus reveals the recent history of smallpox. Curr Biol. 2016; 26(24): 3407– 12. 27939314 PMC5196022
  16. [17]Teixeira H, Berg T, Uusitalo L, et al. A catalogue of marine biodiversity indicators. Front Mar Sci. 2016; 3: 207.
  17. [18]Cowan DA, Ramond JB, Makhalanyane TP, et al. Metagenomics of extreme environments. Curr Opin Microbiol. 2015; 25: 97– 102. 26048196
  18. [20]Roeder GS, Fink GR. DNA rearrangements associated with a transposable element in yeast. Cell. 1980; 21(1): 239– 49. 6250713
  19. [21]Sajantila A. Editors’ Pick: Contamination has always been the issue!. Investig Genet. 2014; 5: 106. PMC4279886 25551017
  20. [23]Jeong C, Di Rienzo A. Adaptations to local environments in modern human populations. Curr Opin Genet Dev. 2014; 29: 1– 8. 25129844 PMC4258478
  21. [24]Beres S, Kachroo P, Nasser W, et al. Transcriptome remodeling contributes to epidemic disease caused by the human pathogen Streptococcus pyogenes. mBio. 2016: 00403– 16. PMC4895104 27247229
  22. [25]Fumagalli M, Sironi M. Human genome variability, natural selection and infectious diseases. Curr Opin Immunol. 2014; 30: 9– 16. 24880709
  23. [26]Long H, Sung W, Kucukyildirim S, et al. Evolutionary determinants of genome-wide nucleotide composition. Nat Ecol Evol. 2018; 2(2): 237– 40. 29292397 PMC6855595
  24. [27]Golan A. Foundations of Info-Metrics: Modeling and Inference with Imperfect Information. Oxford University Press; 2017.
  25. [28]Hernaez M, Pavlichin D, Weissman T, et al. Genomic data compression. Annu Rev Biomed Data Sci. 2019; 2: 19– 37.
  26. [29]Hosseini M, Pratas D, Pinho AJ. A survey on data compression methods for biological sequences. Information. 2016; 7(4): 56.
  27. [30]Grumbach S, Tahi F. Compression of DNA sequences. In: DCC '93: Data Compression Conference, Snowbird, UT. 1993: 340– 50.
  28. [31]Grumbach S, Tahi F. A new challenge for compression algorithms: genetic sequences. Inf Process Manag. 1994; 30(6): 875– 86.
  29. [32]Rivals E, Delahaye JP, Dauchet M, et al. A guaranteed compression scheme for repetitive DNA sequences. In: DCC '96: Data Compression Conference, Snowbird, UT. 1996: 453.
  30. [33]Loewenstern D, Yianilos PN. Significantly lower entropy estimates for natural DNA sequences. J Comput Biol. 1999; 6(1): 125– 42. 10223669
  31. [34]Allison L, Edgoose T, Dix TI. Compression of strings with approximate repeats. Proc Int Conf Intell Syst Mol Biol. 1998; 6: 8– 16. 9783204
  32. [35]Apostolico A, Lonardi S. Compression of biological sequences by greedy off-line textual substitution. In: DCC '00: Proceedings of the Conference on Data Compression. Washington, DC: IEEE; 2000: 143– 52.
  33. [36]Chen X, Li M, Ma B, Tromp J. DNACompress: Fast and effective DNA sequence compression. 2002; 18(12): 1696– 8. 12490460
  34. [37]Matsumoto T, Sadakane K, Imai H. Biological sequence compression algorithms. In: Dunker AK, Konagaya A, Miyano S et al., et al., eds. Genome Informatics 2000: Proc. of the 11th Workshop, Tokyo. 2000: 43– 52. 11700586
  35. [38]Tabus I, Korodi G, Rissanen J. DNA sequence compression using the normalized maximum likelihood model for discrete regression. In: DCC '03: Proceedings of the Conference on Data Compression. Washington, DC: IEEE; 2003: 253– 62.
  36. [39]Korodi G, Tabus I. An efficient normalized maximum likelihood algorithm for DNA sequence compression. ACM Trans Inf Syst. 2005; 23(1): 3– 34.
  37. [41]Manzini G, Rastero M. A simple and fast DNA compressor. J Softw Pract Exp. 2004; 34: 1397– 411.
  38. [42]Behzadi B, Le Fessant F. DNA compression challenge revisited: A dynamic programming approach. In: Apostolico A, Crochemore M, Park K, eds. Combinatorial Pattern Matching: Proc. of CPM-2005. Berlin, Heidelberg: Springer; 2005, doi.org/10.1007/11496656_17. · DOI
  39. [43]Cao MD, Dix TI, Allison L, et al. A simple statistical algorithm for biological sequence compression. In: 2007 Data Compression Conference (DCC'07), Snowbird, UT. IEEE; 2007: 43– 52.
  40. [44]Vey G. Differential direct coding: a compression algorithm for nucleotide sequence data. Database (Oxford). 2009; 2009, doi: 10.1093/database/bap013. PMC2797453 20157486 · DOI
  41. [45]Mishra KN, Aaggarwal A, Abdelhadi E, et al. An efficient horizontal and vertical method for online dna sequence compression. Int J Comput Appl. 2010; 3(1): 39– 46.
  42. [46]Rajeswari PR, Apparao A. GENBIT Compress-Algorithm for repetitive and non repetitive DNA sequences. Int J Comput Sci Inf Technol. 2010; 2: 25– 9.
  43. [47]Gupta A, Agarwal S. A novel approach for compressing DNA sequences using semi-statistical compressor. Int J Comput Appl. 2011; 33(3): 245– 51.
  44. [48]Zhu Z, Zhou J, Ji Z, et al. DNA sequence compression using adaptive particle swarm optimization-based memetic algorithm. IEEE Trans Evol Comput. 2011; 15(5): 643– 58.
  45. [49]Pinho AJ, Pratas D, Ferreira PJSG. Bacteria DNA sequence compression using a mixture of finite-context models. In: Proc. of the IEEE Workshop on Statistical Signal Processing, Nice, France. 2011: 125– 28.
  46. [50]Pinho AJ, Ferreira PJSG, Neves AJR, et al. On the representability of complete genomes by multiple competing finite-context (Markov) models. PLoS One. 2011; 6(6): e21588. 21738720 PMC3128062
  47. [51]Roy S, Khatua S, Roy S et al. An efficient biological sequence compression technique using lut and repeat in the sequence. arXiv. 2012: 1209.5905.
  48. [52]Satyanvesh D, Balleda K, Padyana A, et al. GenCodex - A novel algorithm for compressing DNA sequences on multi-cores and GPUs. In: Proc. IEEE, 19th International Conf. on High Performance Computing (HiPC), Pune, India. IEEE; 2012.
  49. [53]Bose T, Mohammed MH, Dutta A, et al. BIND–An algorithm for loss-less compression of nucleotide sequence data. J Biosci. 2012; 37(4): 785– 9. 22922203
  50. [54]Li P, Wang S, Kim J, et al. DNA-COMPACT: DNA compression based on a pattern-aware contextual modeling technique. 2013; 8(11): e80377. PMC3840021 24282536
  51. [55]Pratas D, Pinho AJ. Exploring deep Markov models in genomic data compression using sequence pre-analysis. In: 22nd European Signal Processing Conference (EUSIPCO), Lisbon. IEEE; 2014: 2395– 9.
  52. [56]Sardaraz M, Tahir M, Ikram AA, et al. SeqCompress: An algorithm for biological sequence compression. Genomics. 2014; 104(4): 225– 8. 25173568
  53. [57]Guo H, Chen M, Liu X, et al. Genome compression based on Hilbert space filling curve. In: Proceedings of the 3rd International Conference on Management, Education, Information and Control (MEICI 2015), Shenyang, China. 2015: 29– 31.
  54. [58]Xie X, Zhou S, Guan J. CoGI: Towards compressing genomes as an imag. IEEE/ACM Trans Comput Biol Bioinform. 2015; 12(6): 1275– 85. 26671800
  55. [59]Chen M, Shao J, Jia X. Genome sequence compression based on optimized context weighting. Genet Mol Res. 2017; 16(2), doi: 10.4238/gmr16026784. 28525649 · DOI
  56. [60]Bakr NS, Sharawi AA. Improve the compression of bacterial DNA sequence. In: 2017 13th International Computer Engineering Conference (ICENCO). IEEE; 2017: 286– 90.
  57. [61]Mansouri D, Yuan X. One-Bit DNA Compression Algorithm. In: Cheng L, Leung A, Ozawa S, eds. International Conference on Neural Information Processing. Cham: Springer; 2018: 378– 86.
  58. [62]Wang R, Bai Y, Chu YS, et al. DeepDNA: A hybrid convolutional and recurrent neural network for compressing human mitochondrial genomes. In: 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE; 2018: 270– 4.
  59. [63]Wang R, Zang T, Wang Y. Human mitochondrial genome compression using machine learning techniques. Hum Genomics. 2019; 13(1): 49. 31639043 PMC6805717
  60. [64]Pratas D, Hosseini M, Silva JM, et al. A reference-free lossless compression algorithm for DNA sequences using a competitive prediction of two classes of weighted models. Entropy. 2019; 21(11): 1074.
  61. [65]Mohammed MH, Dutta A, Bose T, et al. DELIMINATE—A fast and efficient method for loss-less compression of genomic sequences: sequence analysis. Bioinformatics. 2012; 28(19): 2527– 9. 22833526
  62. [66]Pinho AJ, Pratas D. MFCompress: A compression tool for FASTA and multi-FASTA data. Bioinformatics. 2014; 30(1): 117– 8. 24132931 PMC3866555
  63. [67]Kryukov K, Ueda MT, Nakagawa S, et al. Nucleotide Archival Format (NAF) enables efficient lossless reference-free compression of DNA sequences. Bioinformatics. 2019; 35(19): 3826– 8. 30799504 PMC6761962
  64. [68]Christley S, Lu Y, Li C, et al. Human genomes as email attachments. Bioinformatics. 2009; 25(2): 274– 5. 18996942
  65. [69]Brandon MC, Wallace DC, Baldi P. Data structures and compression algorithms for genomic sequence data. Bioinformatics. 2009; 25(14): 1731– 8. 19447783 PMC2705231
  66. [70]Ochoa I, Hernaez M, Weissman T. iDoComp: A compression scheme for assembled genomes. Bioinformatics. 2015; 31(5): 626– 33. 25344501 PMC5855886
  67. [71]Deorowicz S, Danek A, Niemiec M. GDC 2: Compression of large collections of genomes. Sci Rep. 2015; 5: 11565. 26108279 PMC4479802
  68. [72]Kuruppu S, Puglisi SJ, Zobel J. Relative Lempel-Ziv compression of genomes for large-scale storage and retrieval. In: International Symposium on String Processing and Information Retrieval. Springer; 2010: 201– 206.
  69. [73]Wang C, Zhang D. A novel compression tool for efficient storage of genome resequencing data. Nucleic Acids Res. 2011; 39(7): e45. 21266471 PMC3074166
  70. [74]Kuruppu S, Puglisi SJ, Zobel J. Optimized relative Lempel-Ziv compression of genomes. In: Proceedings of the Thirty-Fourth Australasian Computer Science Conference-Volume 113. 2011: 91– 8.
  71. [75]Deorowicz S, Grabowski S. Robust relative compression of genomes with random access. Bioinformatics. 2011; 27(21): 2979– 86. 21896510
  72. [76]Pinho AJ, Pratas D, Garcia SP. GReEn: A tool for efficient compression of genome resequencing data. Nucleic Acids Res. 2012; 40(4): e27. 22139935 PMC3287168
  73. [77]Wandelt S, Leser U. FRESCO: Referential compression of highly similar sequences. IEEE/ACM Trans Comput Biol Bioinform. 2013; 10(5): 1275– 88. 24524158
  74. [78]Liu Y, Peng H, Wong L, et al. High-speed and high-ratio referential genome compression. Bioinformatics. 2017; 33(21): 3364– 72. 28651329
  75. [79]Fan W, Dai W, Li Y, et al. Complementary contextual models with FM-Index for DNA compression. In: 2017 Data Compression Conference (DCC). IEEE; 2017: 82– 91.
  76. [80]Yao H, Ji Y, Li K, et al. HRCM: An efficient hybrid referential compression method for genomic big data. BioMed Res Int. 2019; 2019, doi: 10.1155/2019/3108950. PMC6930768 31915686 · DOI
  77. [82]Goyal M, Tatwawadi K, Chandak S, et al. DeepZip: Lossless data compression using recurrent neural networks. arXiv. 2018: 1811.08162.
  78. [83]Absardi ZN, Javidan R. A fast reference-free genome compression using deep neural networks. In: 2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE), Sofia, Bulgaria. IEEE; 2019: 1– 7.
  79. [84]Robbins H, Monro S. A stochastic approximation method. Ann Math Stat. 1951; 22(3): 400– 7.
  80. [85]Hiransha M, Gopalakrishnan EA, Menon VK, et al. NSE stock market prediction using deep-learning models. Procedia Comput Sci. 2018; 132: 1351– 62.
  81. [86]Glorot X, Bengio Y. Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. 2010: 249– 56.
  82. [87]LeCun YA, Bottou L, Orr GB, et al. Efficient backprop. In: Neural Networks: Tricks of the trade. Springer; 2012: 9– 48.
  83. [88]Pratas D, Toppinen M, Pyöriä L, et al. A hybrid pipeline for reconstruction and analysis of viral genomes at multi-organ level. Gigascience. 2020; 9: giaa086. 32815536 PMC7439602
  84. [90]Kryukov K, Ueda MT, Nakagawa S, et al. Sequence Compression Benchmark (SCB) database—A comprehensive evaluation of reference-free compressors for FASTA-formatted sequences. Gigascience. 2020; 9(7): giaa072. 32627830 PMC7336184
  85. [91]Pratas D, Pinho AJ. A DNA sequence corpus for compression benchmark. In: International Conference on Practical Applications of Computational Biology and Bioinformatics. Springer; 2018: 208– 15.
  86. [92]Ijdo J, Baldini A, Ward D, et al. Origin of human chromosome 2: An ancestral telomere-telomere fusion. Proc Natl Acad Sci U S A. 1991; 88(20): 9051– 5. 1924367 PMC52649
  87. [93]Hagedoorn H. AMD Ryzen 5 3600 review - Power Consumption and temperatures. https://www.guru3d.com/articles-pages/amd-ryzen-5-3600-review,7.html. Accessed in 2020-07-30.
  88. [97]Lewin HA, Robinson GE, Kress WJ, et al. Earth BioGenome Project: Sequencing life for the future of life. Proc Natl Acad Sci U S A. 2018; 115(17): 4325– 33. 29686065 PMC5924910
  89. [98]Pratas D, Pinho AJ. On the approximation of the Kolmogorov complexity for DNA sequences. In: Iberian Conference on Pattern Recognition and Image Analysis. Springer; 2017: 259– 66.
  90. [99]Hosseini M, Pratas D, Morgenstern B, et al. Smash++: An alignment-free and memory-efficient tool to find genomic rearrangements. Gigascience. 2020; 9(5), doi: 10.1093/gigascience/giaa048. PMC7238676 32432328 · DOI
  91. [100]Cilibrasi R, Vitányi PM. Clustering by compression. IEEE Trans Inf Theor. 2005; 51(4): 1523– 45.
  92. [101]Li M, Li X, Ma B, et al. The similarity metric. arXiv. 2008: cs/0111054. · arXiv
  93. [104]Kim TY, Oh KJ, Kim C, et al. Artificial neural networks for non-stationary time series. Neurocomputing. 2004; 61: 439– 47.
  94. [105]Lau AK, Dörrer S, Leimeister CA, et al. Read-SpaM: Assembly-free and alignment-free comparison of bacterial genomes with low sequencing coverage. BMC Bioinformatics. 2019; 20(20): 638. 31842735 PMC6916211
  95. [106]Silva M, Pratas D, Pinho AJ. Supporting data for “Efficient DNA sequence compression with neural networks.”. GigaScience Database; 2020. 10.5524/100808 PMC7657843 33179040 · DOI

Cited by

Sorted by citations

Authors
Silva, M., Pratas, D., & Pinho, A. J.
Affiliations
  • University of Aveiro (PT)
  • University of Helsinki (FI)
Corresponding author
  • Silva, Milton
  • Pratas, Diogo
ORCID
Venue
GigaScience
Publisher
Oxford University Press (OUP)
Published
2020-11
Volume
9
Issue
11
ISSN
2047-217X (ISSN-L)
DOI
10.1093/gigascience/giaa119
PMC
PMC7657843
OpenAlex
W3098741556
Cited by
57
References
3
Citations by year
2026: 8 · 2025: 12 · 2024: 13 · 2023: 8 · 2022: 11 · 2021: 4 · 2020: 1
FWCI
2.32
Type
journal article
Language
en
Open access
Licence
Locations
  • GigaScience (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Työväentutkimus Vuosikirja (submittedVersion)
  • OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
Funders
Fuel Cell Technologies Program (UIDB/00127/2020, CI-CTTI-94-ARH/2019)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Algorithms and Data Compression
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Fractal and DNA sequence analysis
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Data Compression Techniques
Keywords
DNA sequence compression, lossless compression, artificial neural networks, context model, reference-based compression, genomic data storage, Y chromosome, archaeal genomes, viral genome, ancient DNA, human virome, primates
MeSH terms
Algorithms, Base Sequence, Software, Neural Networks, Computer, Sequence Analysis, DNA, High-Throughput Nucleotide Sequencing

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review