CoChat

Environmental conditions shape the nature of a minimal bacterial genome

Open access
Antczak, Magdalena, Michaelis, Martin, Wass, Mark N. · 2019 · Nature Communications

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index; 10 of the 19 sources its claims cite were checked for support.

Abstract

Of the 473 genes in the genome of the bacterium with the smallest genome generated to date, 149 genes have unknown function, emphasising a universal problem; less than 1% of proteins have experimentally determined annotations. Here, we combine the results from state-of-the-art in silico methods for functional annotation and assign functions to 66 of the 149 proteins. Proteins that are still not annotated lack orthologues, lack protein domains, and/ or are membrane proteins. Twenty-four likely transporter proteins are identified indicating the importance of nutrient uptake into and waste disposal out of the minimal bacterial cell in a nutrient-rich environment after removal of metabolic enzymes. Hence, the environment shapes the nature of a minimal genome. Our findings also show that the combination of multiple different state-of-the-art in silico methods for annotating proteins is able to predict functions, even for difficult to characterise proteins and identify crucial gaps for further development.

Overview

CoChat Summary

TL;DR

Combining 22 in silico methods assigned new functional information to 66 of the 149 unknown-function proteins in the minimal bacterial genome, with many newly identified as transporters. The authors argue that the minimal genome is environment-dependent, requiring a core gene set plus additional genes for life in a nutrient-rich medium.

Key findings

  • New functional information was assigned to 66 of 149 previously unknown-function proteins; only 16 remained hypothetical, compared with 78 proteins left unannotated by Danchin and Fang.
  • Benchmarking on 289 known-function proteins showed 92% (266/289) agreement with Hutchison et al. annotations, with an average of 13 supporting methods versus a maximum of three previously.
  • 24 likely transporter proteins were newly identified; transporters now represent 22% of the minimal genome, and 46 of 61 predicted transmembrane unknown-function proteins are likely membrane transport proteins.
  • No single method predicted the most detailed function for more than one-third of unknown-function proteins, and any combination of the top five methods reached the final annotation for at most 25%.

Methods

  • In silico analysis of the 473 proteins encoded by the minimal *Mycoplasma mycoides* genome, focusing on the 149 unknown-function proteins, using 22 computational methods including Pfam, InterPro, eggNOG-Mapper, Phyre2, TMHMM, FFPred3, Argot2.5, CombFunc, LocTree3, and others.
  • Predictions covered orthologues, domains, disorder, structural models, transmembrane helices, ligand binding, GO terms, and operon context; final functions were assigned by manual review, with confidence estimated from the number of supporting methods and normalized average scores.
  • The approach was benchmarked against Hutchison et al.’s Putative, Probable, and Equivalog classes (289 proteins) and compared with predictions by Danchin and Fang.

Limitations

  • Stated: proteins still not annotated lack orthologues, lack protein domains, and/or are membrane proteins; substrate specificity could not be assigned for transporters; Hutchison et al. method outputs were unavailable, so confidence levels could not be compared; Danchin and Fang did not provide methods.
  • Evident from the text: all annotations are computational predictions rather than experimentally confirmed functions, and the benchmark is limited to proteins with existing annotations.
  • Evident from the text: final function assignment relied on manual review and combination of method outputs, which may introduce subjectivity.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41467-019-10837-2

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Basic characterisation of proteins encoded by the minimal bacterial genome. a Orthologues identified in bacteria. Results for each functional class are represented by a different colour: gold for the Unknown class, yellow–Generic, light turquoise–Putative, turquoise–Probable and dark turquoise–Equivalog. b The domain architecture for proteins in each of the five functional confidence classes is plotted (Unknown [Un], Generic [Gn], Putative [Pt], Probable [Pr] and Equivalog [Eq]). Proteins with no domains are displayed in yellow, grey represents single domain proteins and dark blue multi-domain proteins. c Predicted protein disorder in the minimal genome proteins. The results are shown for the five confidence classes from b and coloured according to the percentage of disorder present. Proteins with a percentage disorder >30% are represented by yellow, 20–30% disorder by green, 10–20%-turquoise and 0–10%-blue. Purple indicates proteins without disordered regions. d The percentage of protein structure that can be confidently modified by Phyre2. Functional class colouring as for a
Fig. 1 Basic characterisation of proteins encoded by the minimal bacterial genome. a Orthologues identified in bacteria. Results for each functional class are represented by a different colour: gold for the Unknown class, yellow–Generic, light turquoise–Putative, turquoise–Probable and dark turquoise–Equivalog. b The domain architecture for proteins in each of the five functional confidence classes is plotted (Unknown [Un], Generic [Gn], Putative [Pt], Probable [Pr] and Equivalog [Eq]). Proteins with no domains are displayed in yellow, grey represents single domain proteins and dark blue multi-domain proteins. c Predicted protein disorder in the minimal genome proteins. The results are shown for the five confidence classes from b and coloured according to the percentage of disorder present. Proteins with a percentage disorder >30% are represented by yellow, 20–30% disorder by green, 10–20%-turquoise and 0–10%-blue. Purple indicates proteins without disordered regions. d The percentage of protein structure that can be confidently modified by Phyre2. Functional class colouring as for aCC BY · PubMed Central
Fig. 2 Transmembrane proteins encoded by the minimal bacterial genome. a The number of proteins predicted by TMHMM to have transmembrane helices. Brown indicates proteins with one or more transmembrane helix. Yellow for those without transmembrane helices. b The number of transmembrane helices present in each of the proteins in the minimal genome that is predicted to have one or more transmembrane helix. Results for each functional class are represented by a different colour: gold for the Unknown class, yellow–Generic, light turquoise–Putative, turquoise–Probable and dark turquoise–Equivalog class
Fig. 2 Transmembrane proteins encoded by the minimal bacterial genome. a The number of proteins predicted by TMHMM to have transmembrane helices. Brown indicates proteins with one or more transmembrane helix. Yellow for those without transmembrane helices. b The number of transmembrane helices present in each of the proteins in the minimal genome that is predicted to have one or more transmembrane helix. Results for each functional class are represented by a different colour: gold for the Unknown class, yellow–Generic, light turquoise–Putative, turquoise–Probable and dark turquoise–Equivalog classCC BY · PubMed Central
Fig. 3 Predictions for proteins of known function encoded by the minimal genome. a Assessment of the specificity of functions predicted by Hutchison et al. across all five initial functional classes (Unknown to Equivalog). Functions from different initial specificity classes are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specific. b Comparison of our predictions with the functions predicted by Hutchison et al. for proteins of known function, i.e., from the Putative, Probable and Equivalog functional classes. Colouring indicates the level of agreement between the initial functions and the predictions made here. Dark blue where the functions exactly match, medium blue where the predictions made here were less specific than the initial ones, light blue where our predictions were more specific, dark purple where there were minor differences between the functions and light purple where the function did not agree. c Number of methods supporting the function and the average score of those methods. Each point represents a protein. Methods include those used in the first step of function prediction. Specificity class colouring as for a
Fig. 3 Predictions for proteins of known function encoded by the minimal genome. a Assessment of the specificity of functions predicted by Hutchison et al. across all five initial functional classes (Unknown to Equivalog). Functions from different initial specificity classes are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specific. b Comparison of our predictions with the functions predicted by Hutchison et al. for proteins of known function, i.e., from the Putative, Probable and Equivalog functional classes. Colouring indicates the level of agreement between the initial functions and the predictions made here. Dark blue where the functions exactly match, medium blue where the predictions made here were less specific than the initial ones, light blue where our predictions were more specific, dark purple where there were minor differences between the functions and light purple where the function did not agree. c Number of methods supporting the function and the average score of those methods. Each point represents a protein. Methods include those used in the first step of function prediction. Specificity class colouring as for aCC BY · PubMed Central
Fig. 4 Assigning function to proteins in the minimal genome. The flowchart outlines how functions were assigned to the proteins using MMSYN1_0879 as an example. The top row of methods are used to identify a likely function. The methods in the three groups of boxes (predicted GO terms, ligand binding predictions and membrane protein predictions) are then used to see if they support the function identified by the first group. Where the first group does not predict a function then this second group was used. The figure shows the results obtained for MMSYN1_0879, which was annotated as the gene mgtA, a magnesium importing P-type ATPase
Fig. 4 Assigning function to proteins in the minimal genome. The flowchart outlines how functions were assigned to the proteins using MMSYN1_0879 as an example. The top row of methods are used to identify a likely function. The methods in the three groups of boxes (predicted GO terms, ligand binding predictions and membrane protein predictions) are then used to see if they support the function identified by the first group. Where the first group does not predict a function then this second group was used. The figure shows the results obtained for MMSYN1_0879, which was annotated as the gene mgtA, a magnesium importing P-type ATPaseCC BY · PubMed Central
Fig. 5 Proteins assigned new functions. This figure shows the 51 proteins where the specificity class was increased. Results for each final specificity class are represented by a different colour: orange for the General specificity class, light brown–Specific and dark brown–Highly specific. a Each column represents a protein in the minimal genome and the squares show the methods that made predictions (darker colours indicate support of the final prediction), grey squares indicate predictions that did not support the function, light squares indicate that a method did not make a prediction. Proteins are grouped by their initial specificity class (Hypothetical, General, Specific and Highly specific) and then by their final specificity class. b Boxplot demonstrating the distribution of the scores across proteins. Proteins grouped by their initial specificity class and then by their final specificity class. Horizontal lines represent the median, the lower and upper hinge show respectively first quartile and third quartile, and lower and upper whisker include scores from first quartile to (distance between the first and third quartile) × 1.5 (for lower whisker) and from third quartile to (distance between the first and third quartile) × 1.5 (for upper whisker). Any scores outside of these intervals are shown as points. c The number of methods supporting the function and the average score. Each point represents a protein
Fig. 5 Proteins assigned new functions. This figure shows the 51 proteins where the specificity class was increased. Results for each final specificity class are represented by a different colour: orange for the General specificity class, light brown–Specific and dark brown–Highly specific. a Each column represents a protein in the minimal genome and the squares show the methods that made predictions (darker colours indicate support of the final prediction), grey squares indicate predictions that did not support the function, light squares indicate that a method did not make a prediction. Proteins are grouped by their initial specificity class (Hypothetical, General, Specific and Highly specific) and then by their final specificity class. b Boxplot demonstrating the distribution of the scores across proteins. Proteins grouped by their initial specificity class and then by their final specificity class. Horizontal lines represent the median, the lower and upper hinge show respectively first quartile and third quartile, and lower and upper whisker include scores from first quartile to (distance between the first and third quartile) × 1.5 (for lower whisker) and from third quartile to (distance between the first and third quartile) × 1.5 (for upper whisker). Any scores outside of these intervals are shown as points. c The number of methods supporting the function and the average score. Each point represents a proteinCC BY · PubMed Central
Fig. 6 Confident predictions of protein function in the minimal genome. Both a MMSYN1_0298 and b MMSYN1_0302 were previously classified as hypothetical proteins. The results from prediction methods and the function assigned are shown
Fig. 6 Confident predictions of protein function in the minimal genome. Both a MMSYN1_0298 and b MMSYN1_0302 were previously classified as hypothetical proteins. The results from prediction methods and the function assigned are shownCC BY · PubMed Central
Fig. 7 Multiple methods supporting existing annotations. For all proteins where the predicted function agreed with the existing annotation (i.e., the specificity class was not changed), the number of methods that predicted the function is plotted against the average score from these methods. Points for each of the final specificity classes are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specific
Fig. 7 Multiple methods supporting existing annotations. For all proteins where the predicted function agreed with the existing annotation (i.e., the specificity class was not changed), the number of methods that predicted the function is plotted against the average score from these methods. Points for each of the final specificity classes are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specificCC BY · PubMed Central
Fig. 8 Functional annotations of the minimal bacterial genome. The number of proteins in each of the a protein biological process categories (light and dark purple indicate initial and final categories, respectively). b Specificity classes is shown with the original minimal genome annotation and the annotations identified here. c Shows the change in specificity classes, coloured based on the original specificity class. Results for each initial specificity class are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specific
Fig. 8 Functional annotations of the minimal bacterial genome. The number of proteins in each of the a protein biological process categories (light and dark purple indicate initial and final categories, respectively). b Specificity classes is shown with the original minimal genome annotation and the annotations identified here. c Shows the change in specificity classes, coloured based on the original specificity class. Results for each initial specificity class are represented by a different colour: beige for the Hypothetical specificity class, orange–General, light brown–Specific and dark brown–Highly specificCC BY · PubMed Central
Fig. 9 Prediction of membrane related functions. Each point represents a protein with initially unknown function for which we assigned cell membrane related functions (e.g., transmembrane, transporter). The number of methods that supported the prediction is plotted against the average score from these methods. Points for each of the final specificity classes are represented by a different colour: orange for the General specificity class, light brown–Specific and dark brown–Highly specific
Fig. 9 Prediction of membrane related functions. Each point represents a protein with initially unknown function for which we assigned cell membrane related functions (e.g., transmembrane, transporter). The number of methods that supported the prediction is plotted against the average score from these methods. Points for each of the final specificity classes are represented by a different colour: orange for the General specificity class, light brown–Specific and dark brown–Highly specificCC BY · PubMed Central
Fig. 10 MMSYN1_0325 is predicted to be a transporter and member of the Major facilitator Superfamily. The results from Phyre2, TMHMM, the combination of GO term prediction methods (numbers shown are probability associated with each function) and InterPro are shown. All of these methods supported a transporter function with Phyre2 and InterPro confidently identifying association with the Major facilitator superfamily
Fig. 10 MMSYN1_0325 is predicted to be a transporter and member of the Major facilitator Superfamily. The results from Phyre2, TMHMM, the combination of GO term prediction methods (numbers shown are probability associated with each function) and InterPro are shown. All of these methods supported a transporter function with Phyre2 and InterPro confidently identifying association with the Major facilitator superfamilyCC BY · PubMed Central

Verification

Claims the paper makes

7 Valid · 1 Not valid · 9 Not checked · 107 Own results and interpretations

57 more claims

Create a free account to see all 124 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 28
  • It may state two separate claims as one. 39

Create a free account to see these claims.

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?

2 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

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

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 44 references

Cited by

See all 55

Metadata

Authors
Antczak, M., Michaelis, M., & Wass, M. N.
Affiliations
University of Kent (GB)
Corresponding author
  • Michaelis, Martin
  • Wass, Mark N.
Venue
Nature Communications
Published
2019-07-15
Volume
10
Issue
1
Pages
3100-3100
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2957309664
Cited by
55
References
44
Citations by year
2026: 3 · 2024: 8 · 2023: 6 · 2022: 11 · 2021: 21 · 2020: 6
FWCI
3.14
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Kent Academic Repository (University of Kent) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Bacterial Genetics and Biotechnology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Microbial Metabolic Engineering and Bioproduction
Keywords
protein annotation, gene function prediction, unknown function genes, orthologs, protein domains, membrane proteins, membrane transporters, nutrient uptake, waste disposal, environmental adaptation
MeSH terms
Adaptation, Biological, Bacteria, Software, Genome, Bacterial, Computational Biology, Genes, Essential, Molecular Sequence Annotation

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Environmental conditions shape the nature of a minimal bacterial genome. https://app.cochat.ai/papers/10.1038/s41467-019-10837-2
The paper
Antczak, M., Michaelis, M., & Wass, M. N. (2019). Environmental conditions shape the nature of a minimal bacterial genome. Nature Communications. https://doi.org/10.1038/s41467-019-10837-2

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.