Back

An evolutionary timescale for Bacteria calibrated using the Great Oxidation Event

Davin, A. A.; Woodcroft, B. J.; Soo, R. M.; Murali, R.; Schrempf, D.; Clark, J.; Boussau, B.; Moody, E. R. R.; Szantho, L. L.; Richy, E.; Pisani, D.; Hemp, J.; Fischer, W. W.; Donoghue, P.; Spang, A.; Hugenholtz, P.; Williams, T. A.; Szöllosi, G. J.

2023-08-11 evolutionary biology
10.1101/2023.08.08.552427 bioRxiv
Show abstract

Most of lifes diversity and history is microbial but it has left a meagre fossil record, greatly hindering understanding of evolution in deep time. However, the co-evolution of life and the Earth system has left signatures of bacterial metabolism in the geochemical record, most conspicuously the Great Oxidation Event (GOE) [~]2.33 billion years ago (Ga, (Poulton et al. 2021)), in which oxygenic photosynthesis and tectonism (Eguchi, Seales, and Dasgupta 2019) transformed Earths biosphere from dominantly anaerobic to aerobic. Here, we combine machine learning and phylogenetic reconciliation to infer ancestral transitions to aerobic lifestyles during bacterial evolution. Linking these transitions to the GOE provides new constraints to infer the timetree of Bacteria. We find that extant bacterial phyla are truly ancient, having radiated in the Archaean and the Proterozoic: the oldest include Bacillota (Firmicutes), which radiated 3.1-3.7 Ga, Cyanobacteria (3.3-3.5 Ga) and Patescibacteria (3-3.5 Ga). We show that most bacterial phyla were ancestrally anaerobic and that most transitions to an aerobic lifestyle post-dated the GOE. Our analyses trace oxygen production and consumption back to Cyanobacteria. From that starting point, horizontal transfer seeded aerobic lifestyles across bacterial diversity over hundreds of millions of years. Our analyses demonstrate that the diversification of aerobes proceeded in two waves corresponding to the GOE and to a second sustained rise in atmospheric O2 at the dawn of the Palezoic (Krause et al. 2022).

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.