Back

Kinbiont: From time series to ecological and evolutionary responses in microbial systems

Angaroni, F.; Peruzzi, A.; Alvarenga, E. Z.; Pinheiro, F.

2024-09-09 microbiology
10.1101/2024.09.09.611847 bioRxiv
Show abstract

Microbial behavior is quantitatively characterized by observables inferred from kinetics experiments. Growth rate and biomass yield, for example, are used to map response patterns across different conditions including antibiotic growth inhibition and yield dependence on substrate. As microbial kinetics datasets grow, there is immense potential to advance our understanding of ecological and evolutionary processes. But how can we turn these data into actionable insights about microbial responses? Here we introduce Kinbiont - an ecosystem of numerical methods integrating advanced ordinary differential equation solvers, non-linear optimization, signal processing, and interpretable machine learning algorithms. Kinbiont offers a model-based data analysis pipeline covering all aspects of microbial kinetics, from pre-processing to result interpretation. We demonstrate Kinbionts performance using synthetic and real datasets, including bacterial growth, diauxic curves, phage-bacteria co-cultures, and ecotoxicological responses. Kinbiont can aid biological discovery through data-driven generation of hypotheses that can be tested in targeted experiments.

Published in Nature Communications (predicted rank #3) · training set

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.