Computational palaeobiology 

Machine learning for evolving data in life sciences


Computational palaeobiology is about extracting and interpreting patterns from rich and multifaceted fossil data to understand biological, ecological and environmental change processes and interrelations among them over time and being able to extrapolate them into the future. 

Fossil record is “big data” in a sense that it comes from many different sources, it is biased due to how fossils preserve and how they are collected, identification is interpretation based and uncertain in many ways, records are of varying quality over time.

Fossils are preserved remains or traces of animals, plants, and other organisms from the remote past. The fossil record gives clues about the history of life and environmental changes over millions of years, it helps to understand environmental change, biological and ecological processes and can be used to reconstruct past climate and human impact on the environment. 

Selected publications

  • A survey of computational methods for fossil data analysis by Žliobaitė et al. 2017 in Evolutionary Ecology ResearchPDF
  • Ecosystem evolution and hominin paleobiology at East Turkana, northern Kenya between 2.0 and 1.4 Ma by Patterson et al. 2017 in Palaeogeography, Palaeoclimatology, PalaeoecologyDOI
  • Herbivore teeth predict climatic limits in Kenyan ecosystems by Žliobaitė et al. 2016 in PNASDOI (open access)
  • An ecometric analysis of the fossil mammal record of the Turkana Basin by Fortelius et al. 2016 in Philosophical Transactions B. DOI (open access)
  • Improving Adaptation through Evolution and Learning: A Response to Watson and Szathmáry by Žliobaitė and Stenseth 2016 in Trends in Ecology and Evolution. DOI  PDF

Tutorial

Media coverage