Using thermodynamics to obtain geochemical information from genomes (2023)

Chemical analysis of reference proteomes of methanogens reveals adaptation to redox conditions (Figure 1)

utogig1()

Data sources: Hierarchical clustering tree of methanogenesis genes (Lyu and Lu, 2018); Z_C calculated for reference proteomes (Dick and Tan, 2023) based on RefSeq protein sequences (O’Leary et al., 2016). See Table S1 in the paper’s Supplementary Information for references for optimal growth temperature (T_opt).

Relative stabilities of organic compounds depend on redox conditions (Figure 2)

utogig2()

## [1] "p-values for (a):"
## [1] 1.1e-08 1.0e+00 1.2e-07

Data sources: List of organic compounds from Shock and Canovas (2010) and TCA cycle metabolites from Canovas and Shock (2016). See Table 1 in the paper for references for thermodynamic data.

Thermodynamic model for methanogen niche structure (Figure 3)

utogig3()

Data sources: Lost City T and H₂ concentration (Kelley et al., 2005), Rainbow vent fluid mixing model (Shock and Canovas, 2010), methanogenic and non-methanogenic sediments (Lovley and Goodwin, 1988).

Chemical and thermodynamic analysis of evolutionary divergence along redox gradients (Figure 4)

Table_S6 <- utogig4()

## [1] "logaH2 at dashed lines in (b):"
## [1] -7.4 -7.2 -8.8

Data sources: Methanogens (Lyu and Lu, 2018), Nif-bearing organisms (Poudel et al., 2018), Thaumarchaeota (Ren et al., 2019). Amino acid compositions are from RefSeq proteomes (methanogens and Nif-bearing organisms) or NCBI and IMG genomes (Thaumarchaeota).

Parametric and non-parametric tests of comparisons in Figure 4 (Table S6)

• p_Tukey: One-way ANOVA followed by Tukey’s Honest Significant Difference test.
• p_Dunn: Kruskal-Wallis test followed by Dunn’s test with Bonferroni correction for multiple comparisons.

kable(Table_S6, row.names = FALSE, digits = 99, format.args = list(scientific = TRUE))

Comparison of Z_C of proteomes predicted by Glimmer and downloaded from NCBI (Figure S1)

utogigS1()

Association between redox gradients and Z_C of proteins and lipids in alkaline Yellowstone hot springs (Figure S2)

utogigS2()

Data sources: Proteins (Dick and Shock, 2013) and lipids (Boyer et al., 2020).

Relative stabilities of organic compounds at lower activity (Figure S3)

utogig2(logact = -6)

## [1] "p-values:"
## [1] 6.0e-09 1.0e+00 1.5e-07

loga_H2-T plots for different organic compound classes (Figure S4)

utogigS4()

Convert loga_H2 to Eh

logaH2_15 <- -15
pH <- 7
(logK1 <- subcrt(c("H2", "H+", "e-"), c(-1, 2, 2), T = 25)$out$logK)

## [1] 3.105009

pe <- (- logK1 - 2 * pH - logaH2_15) / 2
Eh <- convert(pe, "Eh")
print(paste0("Given logaH2 = ", logaH2_15, " and pH = ", pH, ", Eh = ", round(1000 * Eh), " mV at equilibrium"))

## [1] "Given logaH2 = -15 and pH = 7, Eh = -62 mV at equilibrium"

Convert p_H2 to loga_H2

(logK2 <- subcrt(c("H2", "hydrogen"), c(-1, 1), T = 25)$out$logK)

## [1] 3.105009

# pH2 = 10^-2
logpH2 <- -2
print(paste0("Given logpH2 = ", logpH2, ", logaH2 = ", round(logpH2 - logK2, 1), " at equilibrium"))

## [1] "Given logpH2 = -2, logaH2 = -5.1 at equilibrium"

# 0.55 ppmv H2
logpH2 <- log10(0.55e-6)
print(paste0("Given logpH2 = ", round(logpH2, 2), ", logaH2 = ", round(logpH2 - logK2, 1), " at equilibrium"))

## [1] "Given logpH2 = -6.26, logaH2 = -9.4 at equilibrium"

Convert loga_H2 to logf_H2 and logf_O2

(logK3 <- subcrt(c("H2O", "oxygen", "hydrogen"), c(-1, 0.5, 1), T = 25)$out$logK)

## [1] -41.55238

logfH2 <- logaH2_15 + logK2
logfO2 <- 2 * (logK3 - logfH2)
print(paste0("Given logaH2 = ", logaH2_15, ", logfH2 = ", round(logfH2, 1), " at equilibrium"))

## [1] "Given logaH2 = -15, logfH2 = -11.9 at equilibrium"

print(paste0("Given logaH2 = ", logaH2_15, ", logfO2 = ", round(logfO2, 1), " at equilibrium"))

## [1] "Given logaH2 = -15, logfO2 = -59.3 at equilibrium"