Limits...
Metabolic modeling of endosymbiont genome reduction on a temporal scale.

Yizhak K, Tuller T, Papp B, Ruppin E - Mol. Syst. Biol. (2011)

Bottom Line: A fundamental challenge in Systems Biology is whether a cell-scale metabolic model can predict patterns of genome evolution by realistically accounting for associated biochemical constraints.The model's network-based predictive ability outperforms predictions obtained using genomic features of individual genes, reflecting the effect of selection imposed by metabolic stoichiometric constraints.Thus, while the timing of gene loss might be expected to be a completely stochastic evolutionary process, remarkably, we find that metabolic considerations, on their own, make a marked 40% contribution to determining when such losses occur.

View Article: PubMed Central - PubMed

Affiliation: The Blavatnik School of Computer Science, Tel Aviv University, Tel Aviv, Israel. kerenyiz@post.tau.ac.il

ABSTRACT
A fundamental challenge in Systems Biology is whether a cell-scale metabolic model can predict patterns of genome evolution by realistically accounting for associated biochemical constraints. Here, we study the order in which genes are lost in an in silico evolutionary process, leading from the metabolic network of Escherichia coli to that of the endosymbiont Buchnera aphidicola. We examine how this order correlates with the order by which the genes were actually lost, as estimated from a phylogenetic reconstruction. By optimizing this correlation across the space of potential growth and biomass conditions, we compute an upper bound estimate on the model's prediction accuracy (R=0.54). The model's network-based predictive ability outperforms predictions obtained using genomic features of individual genes, reflecting the effect of selection imposed by metabolic stoichiometric constraints. Thus, while the timing of gene loss might be expected to be a completely stochastic evolutionary process, remarkably, we find that metabolic considerations, on their own, make a marked 40% contribution to determining when such losses occur.

Show MeSH

Related in: MedlinePlus

Correlation results obtained by comparing in silico predicted gene loss times to the times these genes were estimated to be lost during evolution, for four different Buchnera strains. This estimation is based on a phylogenetic reconstruction of the ancestral gene content for the sub-tree, leading from different Buchnera aphidicola strains to their common ancestor with E. coli (Figure 1A). The in silico time estimations were simulated in three different situations: (1) literature-based viable medium and E. coli's biomass function (literature-based viable medium), (2) minimal medium and E. coli's biomass function as used by Pál et al (2006) (minimal conditions), and two control conditions: (3) five random media and the E. coli biomass function, and (4) five random media and random biomass functions (that together yet still form viable growth conditions, Supplementary Material).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3094061&req=5

f2: Correlation results obtained by comparing in silico predicted gene loss times to the times these genes were estimated to be lost during evolution, for four different Buchnera strains. This estimation is based on a phylogenetic reconstruction of the ancestral gene content for the sub-tree, leading from different Buchnera aphidicola strains to their common ancestor with E. coli (Figure 1A). The in silico time estimations were simulated in three different situations: (1) literature-based viable medium and E. coli's biomass function (literature-based viable medium), (2) minimal medium and E. coli's biomass function as used by Pál et al (2006) (minimal conditions), and two control conditions: (3) five random media and the E. coli biomass function, and (4) five random media and random biomass functions (that together yet still form viable growth conditions, Supplementary Material).

Mentions: First, it should be emphasized that the maximal mean Spearman's correlation between in silico and reconstructed gene loss times that one can possibly obtain in our setup is 0.86 and not 1 (due to numerous ties in the vector representing the phylogenetic loss time). Simulating the in silico evolutionary process described above under the literature-based viable medium and computing the correlation between the resulting in silico and reconstructed gene loss times for each of the four B. aphidicola strains, we obtain a mean Spearman's correlation of 0.46 (53% of the maximal correlation, empirical P-value <9.9e−4, see Materials and methods) averaged over these four strains (Figure 2). Notably, when excluding from the analysis those end point genes that are always retained in the pertaining species, we still obtain a significant mean Spearman's correlation of 0.37 (43% of the maximal correlation, empirical P-value <9.9e−4). Interestingly, repeating this analysis under the minimal medium (Supplementary Table 2.a) used in Pál et al (2006), we obtain a similar high mean Spearman's correlation (Supplementary Material). It should be emphasized that, in accordance with an earlier report (Pál et al, 2006), the model accurately predicts that the most preserved pathways are those involved in essential amino-acid metabolism and in central metabolism, including the pentose phosphate pathway, glycolysis and so on, while genes associated with cell envelope synthesis, lipopolysaccharides synthesis and membrane lipid metabolism are not fully retained in the final networks (Supplementary Table 1.f). It is reassuring to see that these predictions match the reports known from the literature, where it is known that B. aphidicola lacks genes for the biosynthesis of cell surface components, including lipopolysaccharides and phospholipids (Shigenobu et al, 2000). Furthermore, the extensive loss of transport capabilities and conservation of essential amino acids biosynthetic pathways are prime characteristics of the aphid symbiont (van Ham et al, 2003).


Metabolic modeling of endosymbiont genome reduction on a temporal scale.

Yizhak K, Tuller T, Papp B, Ruppin E - Mol. Syst. Biol. (2011)

Correlation results obtained by comparing in silico predicted gene loss times to the times these genes were estimated to be lost during evolution, for four different Buchnera strains. This estimation is based on a phylogenetic reconstruction of the ancestral gene content for the sub-tree, leading from different Buchnera aphidicola strains to their common ancestor with E. coli (Figure 1A). The in silico time estimations were simulated in three different situations: (1) literature-based viable medium and E. coli's biomass function (literature-based viable medium), (2) minimal medium and E. coli's biomass function as used by Pál et al (2006) (minimal conditions), and two control conditions: (3) five random media and the E. coli biomass function, and (4) five random media and random biomass functions (that together yet still form viable growth conditions, Supplementary Material).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3094061&req=5

f2: Correlation results obtained by comparing in silico predicted gene loss times to the times these genes were estimated to be lost during evolution, for four different Buchnera strains. This estimation is based on a phylogenetic reconstruction of the ancestral gene content for the sub-tree, leading from different Buchnera aphidicola strains to their common ancestor with E. coli (Figure 1A). The in silico time estimations were simulated in three different situations: (1) literature-based viable medium and E. coli's biomass function (literature-based viable medium), (2) minimal medium and E. coli's biomass function as used by Pál et al (2006) (minimal conditions), and two control conditions: (3) five random media and the E. coli biomass function, and (4) five random media and random biomass functions (that together yet still form viable growth conditions, Supplementary Material).
Mentions: First, it should be emphasized that the maximal mean Spearman's correlation between in silico and reconstructed gene loss times that one can possibly obtain in our setup is 0.86 and not 1 (due to numerous ties in the vector representing the phylogenetic loss time). Simulating the in silico evolutionary process described above under the literature-based viable medium and computing the correlation between the resulting in silico and reconstructed gene loss times for each of the four B. aphidicola strains, we obtain a mean Spearman's correlation of 0.46 (53% of the maximal correlation, empirical P-value <9.9e−4, see Materials and methods) averaged over these four strains (Figure 2). Notably, when excluding from the analysis those end point genes that are always retained in the pertaining species, we still obtain a significant mean Spearman's correlation of 0.37 (43% of the maximal correlation, empirical P-value <9.9e−4). Interestingly, repeating this analysis under the minimal medium (Supplementary Table 2.a) used in Pál et al (2006), we obtain a similar high mean Spearman's correlation (Supplementary Material). It should be emphasized that, in accordance with an earlier report (Pál et al, 2006), the model accurately predicts that the most preserved pathways are those involved in essential amino-acid metabolism and in central metabolism, including the pentose phosphate pathway, glycolysis and so on, while genes associated with cell envelope synthesis, lipopolysaccharides synthesis and membrane lipid metabolism are not fully retained in the final networks (Supplementary Table 1.f). It is reassuring to see that these predictions match the reports known from the literature, where it is known that B. aphidicola lacks genes for the biosynthesis of cell surface components, including lipopolysaccharides and phospholipids (Shigenobu et al, 2000). Furthermore, the extensive loss of transport capabilities and conservation of essential amino acids biosynthetic pathways are prime characteristics of the aphid symbiont (van Ham et al, 2003).

Bottom Line: A fundamental challenge in Systems Biology is whether a cell-scale metabolic model can predict patterns of genome evolution by realistically accounting for associated biochemical constraints.The model's network-based predictive ability outperforms predictions obtained using genomic features of individual genes, reflecting the effect of selection imposed by metabolic stoichiometric constraints.Thus, while the timing of gene loss might be expected to be a completely stochastic evolutionary process, remarkably, we find that metabolic considerations, on their own, make a marked 40% contribution to determining when such losses occur.

View Article: PubMed Central - PubMed

Affiliation: The Blavatnik School of Computer Science, Tel Aviv University, Tel Aviv, Israel. kerenyiz@post.tau.ac.il

ABSTRACT
A fundamental challenge in Systems Biology is whether a cell-scale metabolic model can predict patterns of genome evolution by realistically accounting for associated biochemical constraints. Here, we study the order in which genes are lost in an in silico evolutionary process, leading from the metabolic network of Escherichia coli to that of the endosymbiont Buchnera aphidicola. We examine how this order correlates with the order by which the genes were actually lost, as estimated from a phylogenetic reconstruction. By optimizing this correlation across the space of potential growth and biomass conditions, we compute an upper bound estimate on the model's prediction accuracy (R=0.54). The model's network-based predictive ability outperforms predictions obtained using genomic features of individual genes, reflecting the effect of selection imposed by metabolic stoichiometric constraints. Thus, while the timing of gene loss might be expected to be a completely stochastic evolutionary process, remarkably, we find that metabolic considerations, on their own, make a marked 40% contribution to determining when such losses occur.

Show MeSH
Related in: MedlinePlus