CoChat
Sign Up
Cited by 21

The DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.

Real or fake? Measuring the impact of protein annotation errors on estimates of domain gain and loss events

Kress, Arnaud, Poch, Olivier, Lecompte, Odile, Thompson, Julie D.·2023·Frontiers in Bioinformatics
DOI 10.3389/fbinf.2023.1178926PMC PMC10158824
Open this paper in CoChat — free account

Abstract

Protein annotation errors can have significant consequences in a wide range of fields, ranging from protein structure and function prediction to biomedical research, drug discovery, and biotechnology. By comparing the domains of different proteins, scientists can identify common domains, classify proteins based on their domain architecture, and highlight proteins that have evolved differently in one or more species or clades. However, genome-wide identification of different protein domain architectures involves a complex error-prone pipeline that includes genome sequencing, prediction of gene exon/intron structures, and inference of protein sequences and domain annotations. Here we developed an automated fact-checking approach to distinguish true domain loss/gain events from false events caused by errors that occur during the annotation process. Using genome-wide ortholog sets and taking advantage of the high-quality human and Saccharomyces cerevisiae genome annotations, we analyzed the domain gain and loss events in the predicted proteomes of 9 non-human primates (NHP) and 20 non-S. cerevisiae fungi (NSF) as annotated in the Uniprot and Interpro databases. Our approach allowed us to quantify the impact of errors on estimates of protein domain gains and losses, and we show that domain losses are over-estimated ten-fold and three-fold in the NHP and NSF proteins respectively. This is in line with previous studies of gene-level losses, where issues with genome sequencing or gene annotation led to genes being falsely inferred as absent. In addition, we show that insistent protein domain annotations are a major factor contributing to the false events. For the first time, to our knowledge, we show that domain gains are also over-estimated by three-fold and two-fold respectively in NHP and NSF proteins. Based on our more accurate estimates, we infer that true domain losses and gains in NHP with respect to humans are observed at similar rates, while domain gains in the more divergent NSF are observed twice as frequently as domain losses with respect to S. cerevisiae. This study highlights the need to critically examine the scientific validity of protein annotations, and represents a significant step toward scalable computational fact-checking methods that may 1 day mitigate the propagation of wrong information in protein databases.

Overview

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 · 10.3389/fbinf.2023.1178926

Cites

6 of 6 references verified0 citation errors · 0 not checkable

Statistics

Figures

FIGURE 1 Definition of different domain content used in this analysis. As an example, a reference protein is shown with a single copy of domain D1 and two copies of domain D2. Different domain content is defined as a change in the list of distinct domains, corresponding to either a missing domain (MD) or an additional domain (AD) in a given ortholog from a non-reference species with respect to the reference protein.
FIGURE 1 Definition of different domain content used in this analysis. As an example, a reference protein is shown with a single copy of domain D1 and two copies of domain D2. Different domain content is defined as a change in the list of distinct domains, corresponding to either a missing domain (MD) or an additional domain (AD) in a given ortholog from a non-reference species with respect to the reference protein.CC BY · PubMed Central
FIGURE 2 Fact checking protocol for potential missing domain (MD) and additional domain (AD) events in a given ortholog from a non-reference species with respect to the protein from the reference species. True domain events are classified as either domain switching (the ortholog sequence does not contain domain D2, but contains a domain D2′ belonging to the same Pfam clan as D2), or a MD/AD event.
FIGURE 2 Fact checking protocol for potential missing domain (MD) and additional domain (AD) events in a given ortholog from a non-reference species with respect to the protein from the reference species. True domain events are classified as either domain switching (the ortholog sequence does not contain domain D2, but contains a domain D2′ belonging to the same Pfam clan as D2), or a MD/AD event.CC BY · PubMed Central
FIGURE 3 Example of a MD event due to a gene prediction error. (A). Domain annotations extracted from the Interpro database, with a potential MD in the bushbaby sequence (B5FW94). (B). Part of the Tblastn alignment identifying an additional exon at the 3′ end of the corresponding genome region (Ensembl: ENSOGAG00000008310). Exons are shown in uppercase and introns in lowercase. The number in brackets indicates the length of the potential intron. (C). Part of the protein sequence alignment of the human sequence (Q15904), the B5FW94 bushbaby sequence, and the corrected bushbaby sequence with the C-terminal extension coded by the additional exon.
FIGURE 3 Example of a MD event due to a gene prediction error. (A). Domain annotations extracted from the Interpro database, with a potential MD in the bushbaby sequence (B5FW94). (B). Part of the Tblastn alignment identifying an additional exon at the 3′ end of the corresponding genome region (Ensembl: ENSOGAG00000008310). Exons are shown in uppercase and introns in lowercase. The number in brackets indicates the length of the potential intron. (C). Part of the protein sequence alignment of the human sequence (Q15904), the B5FW94 bushbaby sequence, and the corrected bushbaby sequence with the C-terminal extension coded by the additional exon.CC BY · PubMed Central
FIGURE 4 Results of error detection protocol for (A) NHP proteins and (B) NSF proteins with potential MD.
FIGURE 4 Results of error detection protocol for (A) NHP proteins and (B) NSF proteins with potential MD.CC BY · PubMed Central
FIGURE 5 Results of error detection protocol for (A) NHP proteins and (B) NSF proteins with potential AD events.
FIGURE 5 Results of error detection protocol for (A) NHP proteins and (B) NSF proteins with potential AD events.CC BY · PubMed Central
FIGURE 6 Rates of MD and AD events before and after filtering of false positives for (A) NHP and (B) NSF proteins.
FIGURE 6 Rates of MD and AD events before and after filtering of false positives for (A) NHP and (B) NSF proteins.CC BY · PubMed Central
FIGURE 7 Location of MD and AD events at the N/C-termini or in the middle of the protein for (A) NHP and (B) NSF proteins.
FIGURE 7 Location of MD and AD events at the N/C-termini or in the middle of the protein for (A) NHP and (B) NSF proteins.CC BY · PubMed Central
FIGURE 8 Schematic views of the example domain events observed in NHP proteomes. (A). Domain organizations for human protein NRP1_HUMAN and baboon protein A0A096P459, showing the shared domains CUB (PF00431) and FA58C (PF00754) and the missing MAM (PF00629) domain in baboon. (B). Domain organizations for human protein SP100_HUMAN and marmoset protein A0A2R8MEX0, showing the shared domains HSR (PF03172) and SAND (PF01342) and the missing HMG (PF00505) domain in marmoset. The marmoset protein contains 2 additional domains: PHD (PF00628) and bromodomain (PF00439). (C). Domain organizations for human protein CMAH_HUMAN and all NHP orthologs, showing the additional Rieske-type domain (PF00355) in NHP orthologs.
FIGURE 8 Schematic views of the example domain events observed in NHP proteomes. (A). Domain organizations for human protein NRP1_HUMAN and baboon protein A0A096P459, showing the shared domains CUB (PF00431) and FA58C (PF00754) and the missing MAM (PF00629) domain in baboon. (B). Domain organizations for human protein SP100_HUMAN and marmoset protein A0A2R8MEX0, showing the shared domains HSR (PF03172) and SAND (PF01342) and the missing HMG (PF00505) domain in marmoset. The marmoset protein contains 2 additional domains: PHD (PF00628) and bromodomain (PF00439). (C). Domain organizations for human protein CMAH_HUMAN and all NHP orthologs, showing the additional Rieske-type domain (PF00355) in NHP orthologs.CC BY · PubMed Central
FIGURE 9 Schematic views of the example domain events observed in NSF proteomes. (A). Domain organizations for Saccharomyces cerevisiae protein GAL10_YEAST, Schizosaccharomyces pombe protein GAL10_SCHPO and other NSF orthologs, showing the shared domain GDP-mannose 4,6 dehydratase (PF16363) and the missing Aldose 1-epimerase (PF01263) domain in NSF orthologs. (B). Domain organizations for Saccharomyces cerevisiae protein GAL4_YEAST, A. fumigatus protein Q4WWQ9 and Neurospora crassa protein Q7SAJ9, showing the shared Zn2C6 (PF00172) and Fungal_trans domains (PF04082) and the missing dimerization domain (Dimer_dom) (PF03902) in the Neurospora crassa protein. (C). (B). Domain organizations for Saccharomyces cerevisiae protein LDH1_YEAST, Y. lipolytica protein Q6CEG0 and A. oligospora protein G1XEC8, showing the shared AB hydrolase (PF12697) domain and the additional DSPc (PF00782) domain in Y. lipolytica and A. oligospora proteins.
FIGURE 9 Schematic views of the example domain events observed in NSF proteomes. (A). Domain organizations for Saccharomyces cerevisiae protein GAL10_YEAST, Schizosaccharomyces pombe protein GAL10_SCHPO and other NSF orthologs, showing the shared domain GDP-mannose 4,6 dehydratase (PF16363) and the missing Aldose 1-epimerase (PF01263) domain in NSF orthologs. (B). Domain organizations for Saccharomyces cerevisiae protein GAL4_YEAST, A. fumigatus protein Q4WWQ9 and Neurospora crassa protein Q7SAJ9, showing the shared Zn2C6 (PF00172) and Fungal_trans domains (PF04082) and the missing dimerization domain (Dimer_dom) (PF03902) in the Neurospora crassa protein. (C). (B). Domain organizations for Saccharomyces cerevisiae protein LDH1_YEAST, Y. lipolytica protein Q6CEG0 and A. oligospora protein G1XEC8, showing the shared AB hydrolase (PF12697) domain and the additional DSPc (PF00782) domain in Y. lipolytica and A. oligospora proteins.CC BY · PubMed Central

Verification

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?

Self-check

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

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

Citations

Cites

Show all 6 references

Research works printed but not linked: 48

  1. Altenhoff, 2020, The Quest for Orthologs benchmark service and consensus calls in 2020, Nucleic Acids Res., 48, W538, 10.1093/nar/gkaa308 · DOI
  2. Bányai, 2016, Putative extremely high rate of proteome innovation in lancelets might be explained by high rate of gene prediction errors, Sci. Rep., 6, 30700, 10.1038/srep30700 · DOI
  3. Barrera, 2014, Analysis of the protein domain and domain architecture content in fungi and its application in the search of new antifungal targets, PLoS Comput. Biol., 10, e1003733, 10.1371/journal.pcbi.1003733 · DOI
  4. Berbee, 2017, Early diverging fungi: Diversity and impact at the dawn of terrestrial life, Annu. Rev. Microbiol., 71, 41, 10.1146/annurev-micro-030117-020324 · DOI
  5. Blum, 2021, The InterPro protein families and domains database: 20 years on, Nucleic Acids Res., 49, D344, 10.1093/nar/gkaa977 · DOI · Paper page
  6. Buljan, 2009, The evolution of protein domain families, Biochem. Soc. Trans., 37, 751, 10.1042/bst0370751 · DOI
  7. Cheng, 2018, 10KP: A phylodiverse genome sequencing plan, Gigascience, 7, 1, 10.1093/gigascience/giy013 · DOI · Paper page
  8. Deutekom, 2019, Measuring the impact of gene prediction on gene loss estimates in Eukaryotes by quantifying falsely inferred absences, PLoS Comput. Biol., 15, e1007301, 10.1371/journal.pcbi.1007301 · DOI
  9. Dohmen, 2020, The modular nature of protein evolution: Domain rearrangement rates across eukaryotic life, BMC Evol. Biol., 20, 30, 10.1186/s12862-020-1591-0 · DOI
  10. Dornburg, 2022, Placing human gene families into their evolutionary context, Hum. Genomics, 16, 56, 10.1186/s40246-022-00429-5 · DOI
  11. Forslund, 2011, Domain architecture conservation in orthologs, BMC Bioinforma., 12, 326, 10.1186/1471-2105-12-326 · DOI
  12. Forslund, 2018, Gearing up to handle the mosaic nature of life in the quest for orthologs, Bioinformatics, 34, 323, 10.1093/bioinformatics/btx542 · DOI
  13. Forslund, 2012, Evolution of protein domain architectures, Methods Mol. Biol., 856, 187, 10.1007/978-1-61779-585-5_8 · DOI
  14. Goudey, 2022, Propagation, detection and correction of errors using the sequence database network, Brief. Bioinform, 23, bbac416, 10.1093/bib/bbac416 · DOI
  15. Hittinger, 2004, Parallel inactivation of multiple GAL pathway genes and ecological diversification in yeasts, Proc. Natl. Acad. Sci. U. S. A., 101, 14144, 10.1073/pnas.0404319101 · DOI
  16. Jain, 2019, The evolutionary traceability of a protein, Genome Biol. Evol., 11, 531, 10.1093/gbe/evz008 · DOI
  17. Kuznetsov, 2021, NCBI genome Workbench: Desktop software for comparative genomics, visualization, and GenBank data submission, Methods Mol. Biol., 2231, 261, 10.1007/978-1-0716-1036-7_16 · DOI · Paper page
  18. Lewin, 2018, Earth BioGenome project: Sequencing life for the future of life, Proc. Natl. Acad. Sci. U. S. A., 115, 4325, 10.1073/pnas.1720115115 · DOI · Paper page
  19. Li, 2021, A genome-scale phylogeny of the kingdom Fungi, Curr. Biol., 31, 1653, 10.1016/j.cub.2021.01.074 · DOI
  20. Lin, 2006, An initial strategy for comparing proteins at the domain architecture level, Bioinformatics, 22, 2081, 10.1093/bioinformatics/btl366 · DOI
  21. Martin, 2022, Ensembl 2023, Nucleic Acids Res., 1, D933, 10.1093/nar/gkac958 · DOI
  22. Martin, 2012, Primates, Curr. Biol., 22, R785, 10.1016/j.cub.2012.07.015 · DOI
  23. Meyer, 2020, Understanding the causes of errors in eukaryotic protein-coding gene prediction: A case study of primate proteomes, BMC Bioinforma., 21, 513, 10.1186/s12859-020-03855-1 · DOI
  24. Mistry, 2021, Pfam: The protein families database in 2021, Nucleic Acids Res., 49, D412, 10.1093/nar/gkaa913 · DOI · Paper page
  25. Moore, 2012, The dynamics and evolutionary potential of domain loss and emergence, Mol. Biol. Evol., 29, 787, 10.1093/molbev/msr250 · DOI
  26. Nagy, 2011, Reassessing domain architecture evolution of metazoan proteins: Major impact of gene prediction errors, Genes. (Basel)., 2, 449, 10.3390/genes2030449 · DOI · Paper page
  27. Naranjo-Ortiz, 2019, Fungal evolution: Major ecological adaptations and evolutionary transitions, Biol. Rev. Camb Philos. Soc., 94, 1443, 10.1111/brv.12510 · DOI
  28. Nevers, 2020, Orthology: Promises and challenges, 10.1007/978-3-030-57246-4_9 · DOI
  29. Nevers, 2022, The Quest for Orthologs orthology benchmark service in 2022, Nucleic Acids Res., 50, W623, 10.1093/nar/gkac330 · DOI · Paper page
  30. Nevers, 2019, OrthoInspector 3.0: Open portal for comparative genomics, Nucleic Acids Res., 47, D411, 10.1093/nar/gky1068 · DOI
  31. Peisajovich, 2010, Rapid diversification of cell signaling phenotypes by modular domain recombination, Science, 328, 368, 10.1126/science.1182376 · DOI
  32. Persson, 2019, Domainoid: Domain-oriented orthology inference, BMC Bioinforma., 20, 523, 10.1186/s12859-019-3137-2 · DOI
  33. Przytycka, 2006, Graph theoretical insights into evolution of multidomain proteins, J. Comput. Biol., 13, 351, 10.1089/cmb.2006.13.351 · DOI
  34. Rembeza, 2021, Experimental and computational investigation of enzyme functional annotations uncovers misannotation in the EC 1.1.3.15 enzyme class, PLoS Comput. Biol., 17, e1009522, 10.1371/journal.pcbi.1009446 · DOI
  35. Rogers, 2014, Comparative primate genomics: Emerging patterns of genome content and dynamics, Nat. Rev. Genet., 15, 347, 10.1038/nrg3707 · DOI
  36. Schikora-Tamarit, 2022, Using genomics to understand the mechanisms of virulence and drug resistance in fungal pathogens, Biochem. Soc. Trans., 50, 1259, 10.1042/bst20211123 · DOI
  37. Schmidt, 2022, Modeling genetic diseases in nonhuman primates through embryonic and germline modification: Considerations and challenges, Sci. Transl. Med., 14, eabf4879, 10.1126/scitranslmed.abf4879 · DOI
  38. Schnoes, 2009, Annotation error in public databases: Misannotation of molecular function in enzyme superfamilies, PLoS Comput. Biol., 5, e1000605, 10.1371/journal.pcbi.1000605 · DOI
  39. Sjolander, 2011, Ortholog identification in the presence of domain architecture rearrangement, Brief. Bioinform, 12, 413, 10.1093/bib/bbr036 · DOI
  40. Stamboulian, 2020, The ortholog conjecture revisited: The value of orthologs and paralogs in function prediction, Bioinformatics, 36, i219, 10.1093/bioinformatics/btaa468 · DOI
  41. Syamaladevi, 2013, An alignment-free domain architecture similarity search (ADASS) algorithm for inferring homology between multi-domain proteins, Bioinformation, 9, 491, 10.6026/97320630009491 · DOI
  42. Wang, 2020, Structures of a complete human V-ATPase reveal mechanisms of its assembly, Mol. Cell., 80, 501, 10.1016/j.molcel.2020.09.029 · DOI
  43. Wang, 2021, Protein domain identification methods and online resources, Comput. Struct. Biotechnol. J., 19, 1145, 10.1016/j.csbj.2021.01.041 · DOI
  44. Weirather, 2017, Comprehensive comparison of pacific biosciences and oxford nanopore technologies and their applications to transcriptome analysis, F1000Res., 6, 100, 10.12688/f1000research.10571.1 · DOI
  45. Weisman, 2022, Mixing genome annotation methods in a comparative analysis inflates the apparent number of lineage-specific genes, Curr. Biol., 32, 2632, 10.1016/j.cub.2022.04.085 · DOI · Paper page
  46. Wu, 2022, Reconstruction of gene innovation associated with major evolutionary transitions in the kingdom Fungi, BMC Biol., 20, 144, 10.1186/s12915-022-01346-8 · DOI
  47. Wu, 2022, Initiation of the primate genome project, Zool. Res., 43, 147, 10.24272/j.issn.2095-8137.2022.001 · DOI
  48. Yu, 2019, Grammar of protein domain architectures, Proc. Natl. Acad. Sci. U. S. A., 116, 3636, 10.1073/pnas.1814684116 · DOI

Cited by

Sorted by citations

Metadata

Authors
Kress, A., Poch, O., Lecompte, O., & Thompson, J. D.
Venue
Frontiers in Bioinformatics
Publisher
Frontiers Media SA
Published
2023-04-20
Volume
3
ISSN
2673-7647
DOI
10.3389/fbinf.2023.1178926
PMC
PMC10158824
Cited by
21
References
6
Licence
creativecommons.org/licenses/by/4.0/ (vor)

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