Designing Genomes Using Design-simulate-test Cycles

Designing minimal genomes using whole-cell models

Joshua Rees-Garbutt  et al. Nat Commun. 2020 .

Free PMC article

Erratum in

• Author Correction: Designing minimal genomes using whole-cell models.

  Rees-Garbutt J, Chalkley O, Landon S, Purcell O, Marucci L, Grierson C. Rees-Garbutt J, et al. Nat Commun. 2020 May 6;11(1):2347. doi: 10.1038/s41467-020-15994-3. Nat Commun. 2020. PMID: 32376830 Free PMC article.

Abstract

In the future, entire genomes tailored to specific functions and environments could be designed using computational tools. However, computational tools for genome design are currently scarce. Here we present algorithms that enable the use of design-simulate-test cycles for genome design, using genome minimisation as a proof-of-concept. Minimal genomes are ideal for this purpose as they have a simple functional assay whether the cell replicates or not. We used the first (and currently only published) whole-cell model for the bacterium Mycoplasma genitalium. Our computational design-simulate-test cycles discovered novel in silico minimal genomes which, if biologically correct, predict in vivo genomes smaller than JCVI-Syn3.0; a bacterium with, currently, the smallest genome that can be grown in pure culture. In the process, we identified 10 low essential genes and produced evidence for at least two Mycoplasma genitalium in silico minimal genomes. This work brings combined computational and laboratory genome engineering a step closer.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1. Minesweeper algorithm for genome design.

a In silico single-gene knockouts are conducted to identify no/low essential genes (whose knockout does not prevent cell division). b 26 deletion segments, ranging in size from 100 to 12.5% of the no/low essential genes, are simulated. Grey indicates a gene deletion, white indicates a remaining gene. Deletion segments that on removal do not prevent division go to the next stage. c The largest deletion segment is matched with all division-producing, non-overlapping segments. A powerset (all possible unique combinations of this set of matched deletion segments) is generated and each combination simulated. Combination segments that do not prevent division go to the next stage. d The largest combination segment determines the remaining no/low essential genes that have not been deleted. These remaining genes are divided into eight groups (see Methods), a powerset generated for these eight groups, and each member of the powerset individually appended to the current largest deletion combination and simulated. If none of these simulations on removal produces a dividing cell, the remaining genes are appended as single knockouts to the current largest deletion combination, removed and simulated. The individual remaining genes that do not produce a dividing cell are temporarily excluded and a reduced remaining gene list produced. Details of simulations settings are available in the Methods section. Simulation data generated by the Minesweeper algorithm is available (see Data availability section). Powerset* = the complete powerset is not displayed here.

Fig. 2. GAMA algorithm for genome design.

a Only non-essential genes whose knockout does not prevent cell division are deletion candidates and are equally divided into Sets A–D. Four hundred random deletion subsets are produced and simulated per set, each containing 50–100% of the genes within the set. Deletion subsets that do not prevent division ("viable") go to the next stage. b 3000 combinations of deletion subsets are generated and simulated. c This is a cyclical step. The mutation pool targets a random number of genes for alteration (both knock-ins and knockouts), including essential genes. Simulation data generated by the GAMA algorithm is available (see Data availability section). Details of simulations settings are available in the Methods section.

Fig. 3. Behavioural comparison of whole-cell model, Minesweeper_256, and GAMA_237.

One hundred in silico replicates, with second-by-second values plotted for six cellular variables over 13.89 h (the default endtime of the simulations). The top row shows the expected cellular behaviour (previously show by Karr et al.) and is used for comparison. Minesweeper_256 and GAMA_237 show deviations in phenotype caused by gene deletions. Non-aggregated data for each in silico simulation is available.

Fig. 4. Genome comparison of whole-cell model, Minesweeper_256, and GAMA_237.

The outer ring displays the M. genitalium genome (525 genes in total), with modelled genes (401) in navy and unmodelled genes (124, with unknown function) in grey. The middle ring displays the reduced Minesweeper_256 (256 genes) genome in light blue, with genes present in Minesweeper_265 but not in GAMA_237 in dark blue. The inner ring displays the reduced GAMA_237 (237 genes) genome in light yellow, with genes present in GAMA_237 but not in Minesweeper_265 in dark yellow. Figure produced from published M. genitalium genetic data^,, with genetic data for Minesweeper_256 and GAMA_237 available in Supplementary Data 5.

Fig. 5. Comparing Minesweeper_256 and 2954 GAMA genomes.

The genomes of Minesweeper_256 and all the genomes found by GAMA that were 270 genes and smaller were collated. Each point represents a single genome and is plotted based on a similarity metric (see Methods), showing the pathway of convergence to GAMA_237. The circled genome in the top left is Minesweeper_256 and the circled genome in the bottom right is GAMA_237. The binary formatted genome data used to produce this figure is available.

Similar articles

• Minimal Genome Design Algorithms Using Whole-Cell Models.

  Rees-Garbutt J, Chalkley O, Grierson C, Marucci L. Rees-Garbutt J, et al. Methods Mol Biol. 2021;2189:183-198. doi: 10.1007/978-1-0716-0822-7_14. Methods Mol Biol. 2021. PMID: 33180302

• Essential genes of a minimal bacterium.

  Glass JI, Assad-Garcia N, Alperovich N, Yooseph S, Lewis MR, Maruf M, Hutchison CA 3rd, Smith HO, Venter JC. Glass JI, et al. Proc Natl Acad Sci U S A. 2006 Jan 10;103(2):425-30. doi: 10.1073/pnas.0510013103. Epub 2006 Jan 3. Proc Natl Acad Sci U S A. 2006. PMID: 16407165 Free PMC article.

• Synthetic genomes brought closer to life.

  Holt RA. Holt RA. Nat Biotechnol. 2008 Mar;26(3):296-7. doi: 10.1038/nbt0308-296. Nat Biotechnol. 2008. PMID: 18327239 No abstract available.

• Building the blueprint of life.

  Henry C, Overbeek R, Stevens RL. Henry C, et al. Biotechnol J. 2010 Jul;5(7):695-704. doi: 10.1002/biot.201000076. Biotechnol J. 2010. PMID: 20665643 Review.

• [Whole genome transplantation: bringing natural or synthetic bacterial genomes back to life].

  Labroussaa F, Baby V, Rodrigue S, Lartigue C. Labroussaa F, et al. Med Sci (Paris). 2019 Oct;35(10):761-770. doi: 10.1051/medsci/2019154. Epub 2019 Oct 18. Med Sci (Paris). 2019. PMID: 31625898 Review. French.

Cited by 5 articles

• Genome-scale metabolic modeling reveals key features of a minimal gene set.

  Lachance JC, Matteau D, Brodeur J, Lloyd CJ, Mih N, King ZA, Knight TF, Feist AM, Monk JM, Palsson BO, Jacques PÉ, Rodrigue S. Lachance JC, et al. Mol Syst Biol. 2021 Jul;17(7):e10099. doi: 10.15252/msb.202010099. Mol Syst Biol. 2021. PMID: 34288418 Free PMC article.

• A forecast for large-scale, predictive biology: Lessons from meteorology.

  Covert MW, Gillies TE, Kudo T, Agmon E. Covert MW, et al. Cell Syst. 2021 Jun 16;12(6):488-496. doi: 10.1016/j.cels.2021.05.014. Cell Syst. 2021. PMID: 34139161 Review.

• Atlas: Automatic modeling of regulation of bacterial gene expression and metabolism using rule-based languages.

  Santibáñez R, Garrido D, Martin AJM. Santibáñez R, et al. Bioinformatics. 2020 Dec 26;36(22-23):5473-80. doi: 10.1093/bioinformatics/btaa1040. Online ahead of print. Bioinformatics. 2020. PMID: 33367504 Free PMC article.

• Computer-Aided Whole-Cell Design: Taking a Holistic Approach by Integrating Synthetic With Systems Biology.

  Marucci L, Barberis M, Karr J, Ray O, Race PR, de Souza Andrade M, Grierson C, Hoffmann SA, Landon S, Rech E, Rees-Garbutt J, Seabrook R, Shaw W, Woods C. Marucci L, et al. Front Bioeng Biotechnol. 2020 Aug 7;8:942. doi: 10.3389/fbioe.2020.00942. eCollection 2020. Front Bioeng Biotechnol. 2020. PMID: 32850764 Free PMC article.

• A biochemical network modeling of a whole-cell.

  Burke PEP, Campos CBL, Costa LDF, Quiles MG. Burke PEP, et al. Sci Rep. 2020 Aug 6;10(1):13303. doi: 10.1038/s41598-020-70145-4. Sci Rep. 2020. PMID: 32764598 Free PMC article.

References

1. 1. Rancati G, Moffat J, Typas A, Pavelka N. Emerging and evolving concepts in gene essentiality. Nat. Rev. Genet. 2018;19:34–49. doi: 10.1038/nrg.2017.74. - DOI - PubMed
2. 1. Hutchison CA, et al. Design and synthesis of a minimal bacterial genome. Science. 2016;351:1414–U73. doi: 10.1126/science.aad6253. - DOI - PubMed
3. 1. Iwadate Y, Honda H, Sato H, Hashimoto M, Kato J-I. Oxidative stress sensitivity of engineered Escherichia coli cells with a reduced genome. FEMS Microbiol. Lett. 2011;322:25–33. doi: 10.1111/j.1574-6968.2011.02331.x. - DOI - PubMed
4. 1. Hirokawa Y, et al. Genetic manipulations restored the growth fitness of reduced-genome Escherichia coli. J. Biosci. Bioeng. 2013;116:52–58. doi: 10.1016/j.jbiosc.2013.01.010. - DOI - PubMed
5. 1. Zhou J, Wu R, Xue X, Qin Z. CasHRA (Cas9-facilitated Homologous Recombination Assembly) method of constructing megabase-sized DNA. Nucleic Acids Res. 2016;44:e124. doi: 10.1093/nar/gkw475. - DOI - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources

• Full Text Sources

  • Europe PubMed Central
  • Nature Publishing Group
  • PubMed Central

Designing Genomes Using Design-simulate-test Cycles

Source: https://pubmed.ncbi.nlm.nih.gov/32047145/

Share this post