English

Unsupervised Discovery of Toxoplasma gondii Motility Phenotypes

Computer Vision and Pattern Recognition 2018-01-15 v2 Quantitative Methods

Abstract

Toxoplasma gondii is a parasitic protozoan that causes dis- seminated toxoplasmosis, a disease that afflicts roughly a third of the worlds population. Its virulence is predicated on its motility and ability to enter and exit nucleated cells; therefore, studies elucidating its mechanism of motility and in particular, its motility patterns in the context of its lytic cycle, are critical to the eventual development of therapeutic strate- gies. Here, we present an end-to-end computational pipeline for identifying T. gondii motility phenotypes in a completely unsupervised, data-driven way. We track the parasites before and after addition of extracellular Ca2+ to study its effects on the parasite motility patterns and use this information to parameterize the motion and group it according to similarity of spatiotemporal dynamics.

Keywords

Cite

@article{arxiv.1801.02591,
  title  = {Unsupervised Discovery of Toxoplasma gondii Motility Phenotypes},
  author = {Mojtaba S. Fazli and Stephen A. Vella and Silvia N. J. Moreno and Shannon Quinn},
  journal= {arXiv preprint arXiv:1801.02591},
  year   = {2018}
}

Comments

4 pages, Accepted to 2018 IEEE International Symposium on Biomedical Imaging