Designing minimal genomes using whole-cell models

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.

Understanding metabolic behaviour in whole-cell model output

“…If the way that two processes are coupled together in silico in the whole-cell model yields output that matches experimental data, this can help to develop insight into how these processes are linked in a real cell. This could aid genome design, where insights from modelling can rationally guide in vitro experiments and gene editing (Landon et al 2019, Rees-Garbutt et al 2020.…”

Testing theoretical minimal genomes using whole-cell models

“…To prevent confusion, we named the sets after the main location where the set was constructed ( Table 1) Prior to simulations, we analysed the comparative genomics and single gene essentiality designed minimal gene sets for design commonalities (the protocell designed minimal gene sets were excluded due to their much reduced size) and identified 96 genes in common (Supplementary Data 1 (col L) and 3). 87 of these were classified as essential (genes that when removed stop the cell successfully dividing, classified in-silico previously 6,15 ), eight were non-essential (removal did not prevent successful cell division), and one gene was unmodelled. 6 and in-vivo 4 , they all play a part in essential functions, hence their inclusion.…”

“…Analysis found that every one of the sets deleted essential genes (classified in-silico previously Additionally, Rockville grew cells in mixed pools with DNA isolated from these mixtures rather than from isolated pure colonies of cells 22 . For the Stanford set, the removal of MG_203, MG_250, and MG_470 is likely due to averaging multiple simulation's data together before computational assessment, genes instead found to be essential (Supplementary Table 3 15 ).…”

“…In an attempt to restore in-silico division (Figure 1), we reintroduced essential genes to the minimal gene sets. Based on previous research 15 , we also reintroduced low essential genes (i.e. genes dispensable in some contexts, such as redundant essential genes and gene complexes 19 ).…”

“…genes dispensable in some contexts, such as redundant essential genes and gene complexes 19 ). We did this by comparing the gene content of the individual minimal gene sets with a complete list of the M.genitalium in-silico genes and their essentiality classifications 15 ( Supplementary Data 6-14). For example, the original Agreed set removed low essential genes MG_291 (phosphonate transport) and MG_412 (phosphate transport); by disrupting both these processes, the in-silico cell has no functioning source of phosphate, which has been established previously 15 .…”