Extended Data Fig. 8: Correlation of microbial and geochemical data.
From: Genome-centric view of carbon processing in thawing permafrost
a, CO2 and CH4 concentrations in porewater derived from the bog and fen. The blue line shown is a line of best fit, forced through the origin. Dots indicate the samples, with colours indicating the sample depth. The concentrations are correlated, and the CH4 concentrations are much lower than the CO2 concentrations in both sites. Sample numbers: n = 51 (bog) and 61 (fen) biologically independent samples. b, Methanogenesis versus methanotrophy rates. Each point represents the average relative abundance of methanotrophs and methanogens across all samples in a single core, multiplied by the rate of methane generation or consumption inferred from previous culture-based measurements (2.345 and 20.1 fmol CH4 h–1 per cell of methanogenesis and methanotrophy, respectively, see Methods). The line represents the 1:1 ratio. Inferred fluxes were calculated using relative abundance of methanogenic or methanotrophic lineages so rates are only intended for comparison between the x and y axes, rather than as an absolute measure of CH4 flux. Methanotrophy appears to mitigate a significant proportion of the CH4 generated in the bog sites. c, Correlation of the relative abundance of Ca. ‘Methanoflorens stordalenmirensis’ with the isotopic fractionation of methane (αC) dissolved in paired porewater samples taken from the bog. Previously observed in 2011 using 16S rRNA gene amplicon sequencing33, the correlation is confirmed here using genome-centric metagenomic techniques on the 2011 samples, as well as in a new year of sampling in 2012. Sample numbers: n = 23 (2011) and 24 (2012) biologically independent samples. d, e, Expression of methanogenesis marker gene mcrA across the thaw gradient. Samples analysed with metatranscriptomics are described by the date of sampling, core number and depth. d, Relative contribution of each methanogenic order to the total TPM. e, Relative contribution of all mcrA genes in the metatranscriptome. Metaproteomes revealed the expression of 289 hydrogenotrophic McrA proteins across 13 samples, as well as 78 acetoclastic McrA proteins across eight samples (Supplementary Data 2). f, Linear regression analysis for predicting effective fractionation (αc) of CH4 from environmental variables and Ca. ‘Methanoflorens stordalenmirensis’ abundances in the bog. Ca. ‘Methanoflorens stordalenmirensis’ abundance exceeds bulk geochemical parameters in predicting the effective fractionation of CH4. Each line is the result of a linear regression of the specified measurement against the αc of CH4 in bog porewater samples taken in 2011 and 2012 (n = 47 biologically independent samples).
